• view interview

• view interview

• view interview

• view interview

• view interview

• view interview

• view interview

• view interview

Started in 2021, by two friends who have seen the journey before, Adam and his co-founder launched Workbounce after being early employees on the revenue team at Peakon through the Workday acquisition.

Adam: We both carried responsibility for ramping and enabling revenue teams globally across five markets and we realized what looked like a documentation issue wasn’t really what needed to be solved for. The problem we were looking at was really ask-specific and answers can live in all different areas like Google Docs, Slack messages, etc.— and salespeople search all those things at once. So that’s why we built Workbounce; to try and unify all those sources. And now, as AI is getting more advanced, we’re layering it in so you can have this coach that you can ask any question to and it knows everything that’s available in your company. So product questions, RFP questions, pricing, whatever it is, you can find.Neil: Walk me through how you guys went about bringing that initial idea to life. How did you find your first customers?Adam: Luckily, I’m quite technical. Even as a kid, I’ve been coding and stuff so I just built the product. And for the first year we bootstrapped—didn’t raise any money. We built a proof of concept, we sold a few deals, and in all honesty we might’ve outsold what we built. The product was nowhere near as good as the deals we were selling but we got some initial traction. Then we had some friends like the founders of Peakon who we had worked with come in for a small friends and family round and used that to further product development for a little while longer.My co-founder and I didn’t take a salary for the first 18 months of the company so we were really trying to be as scrappy as we could be. We’ve been continually iterating from that point until now. Product development is always just a race to product market fit so we’ve been working towards that and getting our positioning right. We’ve had a product change as we’ve led in AI and have become more focused on that.You try and be customer-led with product development. You want customers to tell you what their problems are, but you don’t want them to tell you what the solution is because the customer's job isn’t to come up with solutions. Your job is to come up with a solution but also understand the problem in a really detailed way. We were blessed and cursed that we were solving our own problem so we had a lot of domain knowledge but we also had a lot of bias because of the way we experienced the problem. We’ve definitely learned a lot from seeing how other people experienced the problem of how to share knowledge that may not even be documented. Maybe the answer to your question is just an answer someone gave in Slack two months ago, or maybe it has never been asked before.There are lots of ways people tackle that and they’re all interesting, but you try to balance what you know versus what you don’t—and there's always so much you don’t know.

You don’t learn anything when you build features that work, you learn when things don’t work.

Neil: How have the product iterations you’ve made changed the way you guys go about acquiring customers?Adam: The whole first year is about finding out what doesn’t work. You don’t learn anything when something works. If you build a feature and it sounds like that is exactly what the customer needs you didn’t learn anything from that. Nine out of ten things that you build or that you think should work, don’t work. That’s where you learn where your dead ends are and then you shift. Why we do what we do, and what we actually do has massively evolved since we started. Unless you’re super lucky, you don’t come up with a product and nail it the first time. Almost every product started out as some other weird thing; Instagram was a Foursquare clone, and Uber was limousines. So many things start out with a problem but the solution isn’t super obvious. We’ve definitely tried to embrace that and not fight against it and it will just carry on like that forever.Neil: What were some of the challenges you guys had at launch and what did you learn from them?Adam: I think the first thing is that no one cares what you’re building. No one cares that you raised your seed round, and no one cares that you think you’re very smart and have a great idea. Even if you’re very compelling and good at pitching it, they probably still don't care. What people really care about is their own problems. We even knew it going in, I've worked in sales for a long time. My co-founder has a sales background as well. I know deep down no one cares what we're pitching. They just care about their own world. And still, I think when we launch we still have to be so careful to police that. Like hey, look, we're trying to we're putting out a blog post or some content or a feature. And we're talking about it through the lens of, ‘Aren't we so great?’ And it's like, no, no one cares about that. And that's a lesson that I feel, we just have to keep reminding ourselves of. Early on, we definitely were naive about it. Not even naive because we knew it, but we just fell into this trap of being like, we've raised a seed round, we have these ambitious plans and they're all about us. It's like, everyone's just like ‘We don't care.’ You cannot build it and expect people to care. You’ve just got to tell a really compelling story. And even then, most people won't care.There are some companies that do this extremely well, Gong for example. Gong’s marketing strategy was all about salespeople and the things they were annoyed by. If you didn’t know what Gong did, you probably wouldn’t be able to tell from their marketing. They care about their audience and have built a really good marketing machine because of that.

Not all problems are created equal

Neil: So what were some of the things that you guys thought to be true when you started that you found to not be the case?Adam: That's a good question. I'll give you a generalization and I'll give you a specific example of it. So what I thought was true, very specifically, is that everyone was equally motivated to solve their problems. When you talk to someone and they say, “Yeah, this is a real problem” for me, as a founder, you want to hear that because you're like, ok, I'm doing the right thing, and you think that's great. Then we build the solution, come back to you, and ask “Do you want it?” and the response is something like “I'm not that motivated, solve that problem”. And so, not all problems are created equal.You've really got to understand that. How motivated one is to solve that problem. Are they just complaining or do they really want to solve this problem? And the other thing that I realized that ties in with that is that not everyone wants an ideal solution because what's ideal in theory has to meet the real world, right? So in an early version of Workbounce, you would search for content and it would pull in like Google Drive and PDFs and stuff. All the sales reps could vote on content. upvote downvote, like on Reddit, and we had a karma system. So a certain Google Doc or a pitch that could be upvoted and downvoted by reps to decide which content was most useful in the sales process. What we realized as we were pitching is that the people responsible for content, sales enablement, and sales leadership, don't want sales reps to decide what's important. They want to control what's important. They don't want old content, even if it's still useful surfacing. So we thought we created a perfect solution, and in my mind is still the best possible solution, but it turns out that that upvote downvote system was not a big hit. We learned that you have to really test assumptions about stuff—even if you're like, ‘this is a great idea’, and even if people say this is a great idea, you gotta be like, okay, now get your checkbook out. And if they don't, then it's not a great idea. If you build a boat and it doesn't float, it's not a boat. It sinks and that’s all that matters.Neil: There was something you said there about being able to test assumptions; do you guys have a system for that?Adam: Not so much a system but I think the one thing that I try to get better at as an individual, a salesperson, a product leader, is just being more direct than I used to be. I ask ‘why’ a lot more and don’t take things at surface value. If someone tells me they really like a particular thing I ask why they like that, and what’s missing. Even when you’re talking to an engineer and they say ‘Oh we can’t build it this way, we must build this that way.’ Why do we want to do it that way? That’s been the biggest personal development for me. If you went back two years and told me ‘I really like this idea’ I’d say “Cool" and not really dig any deeper. You save yourself a lot of pain if you can face a little bit of discomfort. Being able to sit with someone you don’t really know that well and ask ‘Why do you think that is?’ is a hard thing to do for a lot of people. You’re going to face discomfort somewhere in your life. It’s like working out—you’re going to be uncomfortable at some point. There's no avoiding it. Whether you do it now or when you’re older and immobile. It’s the same for company or product building. At some point you’re gonna feel uncomfortable—you’re better off doing so as soon as possible.Being as direct as you can, asking why to the point of annoyance like a child is a secret hack. There actually is a framework for it used in product development and engineering called the Five Whys. Say the website or the app goes down, and there's an outage for your customers— you say why did it go down? The surface-level answer is because the servers are overloaded. If you stop there, you say, Okay, we just need to upgrade server capacity and they'll be fine. But you say why were the servers overloaded—the servers are overloaded because the maintenance developers had a full backlog. You start there and say, okay, well, we just need to sort out the backlog or whatever. But when you get down to it you find out that actually the reason the web app went down is because we're behind the hiring plan. If you stop at the servers are overloaded, you don't get to the root cause of the thing. So there is a framework for it.

Learn to fight

Neil: So both you and your co-founder have been on the revenue side previously. What does that split of responsibility look like between you two now and how does that affect your personal development for filling in the gaps?Adam: So my co-founder and I happen to be best friends. We've worked together for a long time at this point and we've worked so closely together. So we spent three years taking Peakon from startup, sub 20 employees, to hundreds of employees and towards the exit with Workday. We spent those three years together at that company, arguing with each other because I was responsible for go-to-market strategy and marketing and he was responsible for operations—and I don't know if you know, revenue operations, people that are quite argumentative and they should be not some of my favorite people but we so we, we have very different approaches to things.He's very detail-focused. He's really in the weeds. He immediately wants to know about practical things. I start at like 20,000 feet and he starts in the weeds, and then we meet in the middle. So the thing that we learned to do at Peakon really well is we learn to fight with each other and have it not be a big deal. You know, after three years, we could have really heated arguments in a conference room about something and then go for a beer and just like, we wouldn't even think about it, you know? And I'm married—I joke he is more entwined with me than my wife is in a lot of ways, right? I talk to him just as much, we’re at least as legally entwined as me and my wife, so you've got to be really careful about who you pick as a co-founder, and I don't know if I got lucky or if I'm smart, but I think I just got lucky. It wasn't analytical, I just trusted my gut.I left Peakon a few months before he did and I took a few months off. I went to Portugal for a few months and I knew I was going to start a company and I knew there was one person I wanted to start the company with and it was him. I don't know if that's repeatable for anyone. I think it may be a one and done kind of deal. I don't know if I could even do it myself again. Why can't I just pick him again? But you have to learn to be able to fight and you have to find someone who naturally complements your gaps. Actually, the hardest thing we've learned is that we don't fully cover all the gaps between each other. So we have overlaps in our biases and things that we're both not good at. That can create problems if you're not aware of what they are. And that's been something we've been working on for like, well, forever. But we both know we have these areas where neither of us are good at it so now we have a deficit. That's something that’s really hard to figure out. It's the bits that aren’t so obvious that trip you up. It’s an ongoing process covering that stuff.Neil: What are some of those gaps that have come up recently and what did you guys do about it?Adam: I personally try to curate processes for things that I'm not naturally good at. If I'm not good at that thing. I have to make a process. I'm so bad at sorting out receipts and doing any sort of admin tasks. I don't know why I just cannot sit down and do it. I will find any excuse not to do it. I'm terrible at that so I create a process for it. One of the ways we've overcome most of the challenges is we create processes because if neither of us is going to fill it naturally, we have to have some system in place.
Hiring the right people is also important. But actually one of the things that you can trip up on is that both of you end up asking questions in the same way. If the interview is entirely set up by us, you end up with the same type of person. A big example is that we’re both very analytical people and we can be quite dry. The best response I ever give is ‘that’s cool’. I’m never like ‘That’s awesome! Let’s go’ and I’m aware of that—and he’s the same. So getting that high momentum is really important for the team. We are very intrinsically motivated and other people aren’t. Other people need different reward mechanisms, and different types of recognition. I think we realized that we weren't giving people that all the time as leaders and founders. So that's something that we've had to work on quite a lot. How do we be what other people need us to be contextual to them and not our natural selves all the time.

Avoid Brickwalls

Neil: Given both you and your co-founder have seen a company through scale before, how do you guys think about getting Workbounce to scale now?Adam: That's a good question. There are so many factors. So there’s the things in our control and there are things outside of our control. Things we can control, well, the biggest one is how much money you spend. That's the main thing you can control. Then everything else you can kind of control, like how full your pipeline is, how well your product develops, and hiring the right people and all that stuff.Things you can't control are the right macro conditions, how receptive the funding market is, and how receptive the customer market is. So definitely over the last year, and I think this is pretty common, you see it in the data, but sales cycle lengths are just getting longer and longer. The win rates are dropping and budgets are decreasing. CFOs are signing off on fewer things. So every deal won becomes a little bit harder and what we can't control is whether that’s going to get better or get worse because from 2017 to 2019, SaaS was just booming and everyone was buying everything. Now everyone's churning out everything that they bought in those years.So that's out of our control. We can just do what we can. You plan for the worst and hope for the best essentially. We have always been quite conservative in our runway expenditure planning, but then we leave optionality to take advantage of opportunity when things are going well. For example, there were a lot of layoffs this year, so the talent market was great. You can hire really great people. Our plan right now, being really candid, is to keep doing what we’re doing. We’re not in a rush to raise a Series A, we really want to have a very compelling story when we raise that round and be certain that we’re building the right thing for our market. Early on when I started Workbounce, I spoke to a well-known VC who said “You can run as fast as you want in any direction, but make sure you’re not running into a brick wall…” and I think that’s really true. We could raise a series A and begin scaling our marketing and engineering but we could be running into a brick wall a year or two from now. We just want to be really confident that we’re not doing that and that comes from customer feedback, usage, and having customers that say ‘Please don’t take this product away from my team. That’s going to be a journey for maybe the next year or so.Neil: You touched on some of the larger macros trends from a broader technology standpoint—what are some of the trends you’ve found so far in your category?Adam: I love talking about this. I'll try and keep it really short. I'm gonna start broad like I'm telling you my life story, but it gets to the good point so stay with me. The last 30 years, as most knowledge workers transition to being digital first, we've had an explosion of information. And you see this now, a lot of people are burned out. Oversubscribed to meetings oversubscribed to to-do lists that just grow and never get done. You have Slack messages unread, emails unread. In the early 2000s, there was a lot of research around information overload because it was the first time computers and email on the internet really became a thing in the workplace. Then we sort of forgot about it, but it's just become exponentially worse since that point to the point like how many browser tabs do you have open? How many Slack messages? How many emails? I think I've got about 120 emails in my inbox right now that I haven't read yet. It's just like, how can anyone get to the point where they're on top of all that stuff?The lie that we've been telling ourselves is that we can, but we cannot. It is not possible for you to read the latest product announcements from your company and all your messages and answer your customer emails, you just can't do it. Up until this point, we've had no weapon to fight against it. But what AI is really, really good at is taking lots of information and extracting what’s really important, because that's what you do when you read your emails, right? 120 emails, five of which are important. The rest you don't need to reply to—they're not important or you don't even need to read them. But you don't know that until you've done it. So there's no barrier between your mind and the resource being wasted. What AI does really well is it reduces that information. And that for me is the most important trend going on right now. AI can do a lot of other cool stuff. We can create pictures, write stories, blog posts, alright, that's fine. That's creating more information. I'm not interested in that. What I'm interested in is whether AI can take my completely full workday and condense it down to an hour because I can get to the core information as fast as possible. If you build a product that does that, which is what we're trying to do, but any other, doing the same thing, that’s going to be a massive massive trend.Neil: What are some of the things you and the broader team keep in mind as you navigate growth?Adam: One of the things we watch out for we touched on earlier is people who say they love the solution or think it’s really cool, but especially with AI, do you really trust that it’s going to do it for you? You might like the idea in theory, but as soon as someone says here, this AI will respond to your emails, 9 out of 10 people aren't going to let it do it. They're going to say, 'Oh, no, I don't trust it. What if it says the wrong thing?' So I think those are the things to watch out for. Like just because you can technically do something, and it might be better, like, I'm pretty certain if I set out today you gave me a couple of months, I could build an email assistant, an AI email system that is capable of answering emails just like you would, in the correct way, but do you want me to do that for you? Because you probably don't—not yet anyway.After a few years when we’re used to this, then this will become normal but just because you can do something doesn't mean you should, or that it's worthwhile. And that's the thing that we look out for.The other thing that we look out for and it’s kind of a big belief of mine is that AI is going to replace a lot of people's jobs, and not in the way that we maybe think. The way to think about a job is it's a series of collected and related tasks that have feedback loops between them. What AI will do is become really good at some of those tasks and some of those jobs. Now some jobs, it might do 95% of the task and that job really then does get taken by AI completely. Customer support is a pretty obvious one, especially online support where you're chatting with a person. That can pretty much be done by an AI right now. You don't really need the customer support team anymore.So those jobs get taken away. What we try and do is say, well, yes, it's going to take away some of the things that we do, but it should take away the things that we do that we don't enjoy doing. Like I can't imagine people enjoying doing online support for SaaS vendors. I think those people would much rather be doing something else and obviously, I can't speak for everyone, but I'm assuming, right? I don't think we enjoy sitting in meetings all day long on Zoom. I don't think we enjoy endless or refined Slack messages. I think there are things that we want to do professionally, that are much more creative and human, but we don't have time for.The big belief of what we're trying to build is to free up the human aspects of work. So yes, AI is going to take away some things but it shouldn't take away our humanity. Especially in sales, which is where we focus. Sales, and customer success, are our wheelhouse—those are human-centric activities. The thing you do as a sales rep, where you're, face to face and humanly interacting with a person that's never gonna go away. That's always going to be what people want to engage with. And even buyers who say ‘I don't want to engage with a sales rep.’ They do. People buy from people. That's why most companies can't land big enterprises—you can't self-serve an enterprise deal. So the human aspects of what we do will always be around I think.Then there's a flip side of that, which is something that I really want to do when we have more resources, is upskilling and reskilling. If AI is going to take away these parts of our jobs, what new skills do you need to acquire to replace those? So that's something I'd like to do down the line and build out a content library and a Workbounce University or something for upskilling and reskilling especially for go-to-market jobs to get automated away.That's my belief. You want to try and amplify the humanity of work, not take it away with AI.

Drawing on her experience as the VP of People at a previous startup, Ioanna possessed a profound understanding of the pivotal role top performers play in organizations and the challenges associated with retaining these exceptional talents. This insight fueled the founding mission of Dextego: to provide top performers with precisely what they desire - personalized and meaningful personal development opportunities.

Ioanna: Prior to founding Dextego, my role as Vice President of People and Chief of Staff at a startup shed light on two critical aspects. Firstly, retaining high performers and nurturing the talent within the organization is paramount. Secondly, the existing learning and development tools fell short; they lacked personalization, making them unappealing to employees. These tools felt archaic, failing to engage individuals in meaningful ways. Employees viewed them as mere checkboxes rather than avenues for personal and professional growth.Traditional approaches to learning within organizations seemed outdated, especially in the face of rapid revolution across various sectors. Soft skills development, a crucial aspect of professional growth, proved particularly challenging online. Bridging the gap for employees scattered globally, trying to enhance communication and collaboration skills, became a pressing need shared by many companies.Recognizing this gap, I envisioned a solution. I realized that AI could play a pivotal role in transforming soft skills development. This realization fueled my determination to create Dextego. By harnessing the power of AI, Dextego aims to revolutionize and digitize soft skills development, making learning a personalized, engaging, and transformative experience for employees worldwide.Neil: I've heard a lot that the top performers, those A players that you're talking about here, are now seeing the overall value they can bring to the marketplace and rather than being contributors to large organizations, deciding to go out and try to do their own thing. Is that something that you found the companies you're working with are trying to mitigate against or navigate with?Ioanna: That’s a great point. I think in general we’ll see a rise of Solopreneurship. I don't think that it's necessarily only A-players. I think people are realizing that there is a lot they can leverage and they can just do their own thing without having a boss on top of their head. I definitely believe that this increases the need for soft skills development. For the A players in particular, I think there is more tension.The levers have typically been money and you can't think that the only way to negotiate someone staying within your company is monetary things.So of course this increased the need for talent development, for soft skill training, for L&D overall and especially, learning accountability, because now A players can see that a company is taking into account learning as a way to measure everyone and to push them to develop. So now they see that the company is trying and they see that there's potential. That's what an A-player wants to see. They want to see that the company has a future, that they keep developing, that they are giving them challenges they can solve. They want to feel like they're part of the decision making in the organization. So by instilling learning and accountability, I think a company has much more leverage and chance of retaining those top performers.Neil: What was the initial reception of your mission among your first customers? Is this something that they were able to kind of latch onto right away, or was there some education on that standpoint from your end?Ioanna: The concept has undergone significant evolution, expanding its scope while staying within the same framework. Initially, our focus was solely on enhancing leadership skills, and we envisioned a B2C product. However, through extensive testing and research, we discovered that our market potential was much broader in the B2B sphere. Restricting ourselves to leadership skills alone would have meant overlooking a vast demographic. Not everyone identifies as a leader, so limiting our offering to leadership development would alienate a substantial portion of potential users.My vision was to create a solution that catered to every employee, regardless of their self-perceived leadership capabilities. During my tenure as Vice President of People, I was driven by the desire to support all individuals within an organization, not just those aspiring to leadership positions. This realization guided our shift towards a more inclusive approach, ensuring that our product could benefit every employee, fostering growth and development across diverse roles and skill sets.So we resonated with the feedback from the market and then we opened it up to all sorts of soft skills, including communication, collaboration, conflict resolution to help everyone. In particular, what we saw is that salespeople loved it. They're handling objections, they really need it. But also technical people that saw the struggle when they talked to nontechnical people and they wanted to improve communication. So now we see different use cases for both client facing roles and technical roles. Lastly, I would say what we see today is that because a lot of human coaches’ focus on the C suite, leadership has a lot of access and training, but the rest of the organization doesn't.Where we fit in is helping first time managers, middle managers that are trying to get to the higher levels of the hierarchy but don't have access to sales tools. So that's kind of where we found our product market fit. Lastly, we've identified another big pain point that organizations have, which is that they have all these siloed learning tools and their main thing is a learning management system, but the learning management system itself is not personalized. It's just a course catalog. It has a lot of valuable information, but it's not being utilized as an admin wishes because when you're going through things at work, you don't have time to stop and say, ‘oh, let me go watch a two hour video on leadership and come back and see how I implement that with my team.’We are currently focused on addressing this gap by indexing Learning Management Systems (LMS) and utilizing AI to integrate them seamlessly into an employee's work context. Imagine having an AI coach tailored to each employee, possessing insights into your internal communications and your company's LMS. This integration not only boosts LMS usage but also guides employees within the context of their tasks.For instance, if you seek advice on handling a conflict and your HR department has specific frameworks in place, our system can instantly provide relevant information. No more waiting for a free day to sift through courses on conflict resolution. It's about delivering contextual, personalized, and on-demand learning experiences. This approach represents the future of learning—a future where every tool is tailored to individual needs, accessible when you need it, and focused on microlearning. Effective learning is about relevance, timeliness, and interactivity, enabling employees to actively engage with feedback and information rather than passively consuming it.

Feedback as a growth catalyst

Neil: So from when you guys started to where you are now, was there anything that you previously thought to be true that you’ve found not to be the case?Ioanna: I realized as a founder, in general, that the client will see what you have and will always want something else no matter how good your product is. I've seen it with other startups and of course you can sell what you have now, but the client will never say they're fully satisfied. They will always ask for more, which is the beauty of innovation. They have the power to influence your roadmap, add new features and make it easier for themself. It's really never ending. Even if you say like this is our MVP, this is our bet, it never is. It's always in the process of being a better version.I don't know necessarily if there's one thing I thought of, but I would say I thought maybe for the B2C side that I mentioned earlier, we would see more of a need. But it turns out that B2B is where a tool like us really can be implemented at scale. B2C, you will have people that of course, are interested in personal development, but they are not the majority. If they were, the world would be better. Everybody would be striving for the next step. But there's a lot of people that are satisfied with where they are. There's nothing wrong with that outside that you can't persuade someone that thinks they have developed, that they need to keep developing.Neil: You said something there that I wanted to dive deeper on, about the customers going to always want more in those early stages.  How do you decide which customers are the ones you’re going to take the most feedback from?Ioanna: Absolutely, I often face this question from investors about how I handle the feedback we receive. Feedback, while invaluable, can be a double-edged sword. Many founders tend to get overly optimistic when they receive positive feedback, and that can be a trap. It's easy to bask in the praise and overlook the critical aspects. To me, the positive feedback is the low-hanging fruit—it's reassuring, but it doesn't necessarily propel us forward.
I prefer to first set aside the positive feedback and dig into the negative aspects. I categorize them into short-term and long-term challenges. The immediate issues are scrutinized to figure out what we can address today, tomorrow, or in the next month. I align these with our ongoing trajectory. Simultaneously, I consider our long-term vision. Some feedback, while well-intentioned, might not align with our core goals. It's like a process of elimination, where I filter out distractions and focus on what truly enhances our mission.In this fast-paced environment, where we encounter people and opinions daily, quick decision-making is crucial. The key is to process this information rapidly without feeling overwhelmed. It's about finding that balance between being persistent and adaptive. Positive feedback, while affirming, can also indicate complacency. The real value lies in using feedback to continually raise the bar, pushing ourselves to improve, innovate, and stay true to our vision.Neil: Not pushing the bar too far–can we unpack that a little bit?Ioanna: Yeah, you said this earlier as well–so many tools are coming out that are just helping us do certain things better. But there hasn't been enough thought in terms of strategy behind them, of how they will interact with other existing tools, how they are going to become part of a sustainable workflow. And so if someone sees that product and they say that's great, that positive feedback doesn't mean this company is going to make it big and they're going to change the game. So to me, at this stage, it's very hard to get people to see where you're heading because you have limited resources. So the positive feedback you get today is not the same feedback you'll get like five years from now.Feedback you get now, you have to be thinking about how you'll push the bar further down the road, but validate enough of what you have now to know you're heading in the right direction. So all I'm saying is that sometimes founders start with a vision that it's too hard to accomplish. So they're doing what they can to get there and then they get enough positive feedback and then they stop. I think as a founder, you have to have the mindset that it's never good enough. And you keep going and going. You are the ones showing people what they actually need and what they want because if they knew, they would have done it. And unfortunately, ourt of buyers are not the ones that are innovative and can create new products they are the ones that just consume them. So they will say they like it because they can't think of what's better or what you can do better with it so that's why I say don't take the positive feedback as what you're doing is enough.Neil: So it sounds like you have a pretty clear picture of a larger vision you’re running at, maybe 3,5,10 or 20 years down the line.  I’m curious to hear what that broader vision is and how you imagine the journey might look to get there.Ioanna: The vision for Dextego is vast, filled with possibilities. When I envision success for Dextego in the future, I see a world where every time you unlock your phone, there's a friendly Dextego coach there to greet you and support you throughout your day. This concept blends the roles of a mentor, a coach, and a friend into one powerful entity. In my interactions with numerous human coaches, I've discovered that people often approach them with work-related questions, only to find that these queries often lead to deeply personal discussions. People crave guidance on how to improve, communicate better, and enhance their overall psychological and interpersonal skills. These aspects, crucial to both work and personal life, are at the core of Dextego's mission.
I envision a future where Dextego is so integral to our daily lives that it comes pre-installed on every iPhone, becoming an essential tool for personal and professional growth. Beyond individual impact, I believe Dextego has the potential to revolutionize how we perceive learning and education. These skills—communication, emotional intelligence, and relationship building—are fundamental aspects of our development as humans and as a society. As artificial intelligence becomes more prevalent, it's imperative that we focus on honing our human skills. By emphasizing emotional intelligence, we can create a sustainable world, fostering understanding and cooperation.Traditionally, our focus has been on technical skills, but it's evident that there are numerous societal challenges we still need to address. I firmly believe that the first step towards overcoming these challenges is developing these fundamental human skills. Dextego, in its essence, represents this vital step towards a more empathetic, connected, and harmonious world.Neil: So what are some of the things that you and the broader team are looking out for as you navigate that vision?Ioanna: Accessibility is at the core of our mission. We are committed to ensuring that our product is accessible to everyone, regardless of their abilities or disabilities. This means designing a user experience that is intuitive and easy to navigate for all, including individuals with color blindness and other disabilities. Equally important is keeping our product affordable, ensuring that financial constraints don't hinder anyone from accessing our platform.
We recognize that people have diverse learning styles. Some prefer visual content, while others thrive on text-based information. To cater to these differences, our platform offers a range of learning materials, including videos and text, accommodating various preferences. Moreover, we are introducing a community feature, enabling users to connect based on their skills, fostering a culture of mentorship and mutual support. Unlike traditional mentorship models, this approach creates a dynamic, ongoing community where everyone can both teach and learn, fostering long-lasting relationships.Community-led learning is the cornerstone of our approach. We understand the importance of staying relevant, so we integrate emerging trends like micro learning and gamification. By making the learning experience engaging and rewarding, we motivate users to return to the app regularly. Positive reinforcement mechanisms are embedded within our platform, ensuring that users not only enjoy learning but also feel acknowledged and valued, encouraging them to continue their educational journey.Furthermore, we acknowledge the unique perspectives of different generations, especially Gen Z, who are vocal and innovative. Our platform is designed to bridge generational gaps, providing a space where individuals from diverse backgrounds and age groups can learn from one another. By fostering understanding and collaboration across generations, we aim to create a platform that reflects the real-world diversity of experiences and insights. We don't just focus on creating features for our app; we envision our impact extending far beyond our platform, bridging gaps and fostering understanding in the broader community. This perspective guides our continuous product development, ensuring that every feature we introduce has a meaningful impact both within and outside our app.

Mindful gamification

Neil: You mentioned some of the trends you’ve seen.  Can we double tap that? What are some of the trends you guys uncovered in the last year or so?Ioanna: Customization is key. People want something that they know is for them rather than just for your organization. It's something that's tailored to your current level of skills, your background, your interest, your role. So this I would say is huge. The other one would be on demand. I want it when I want it. That's for everything, right? From food delivery to learning. Now that's where we're going to… an on demand world. I would also say gamification is huge because people's attention span is low. There's way too much distraction and complexity in the world. Everyone is fighting for your attention. Gamification is a smart way, I think, of getting people to do something. But it has to be mindful. You can't do it in a manipulative way just to keep people on the app.

The athlete's approach to prep

Neil: What is the one thing about Dextego that you would want people to understand and take away?Ioanna:  I want people to grasp that working on their soft skills through Dextego isn't just a short-term endeavor; it's an enduring investment in their personal growth. Dextego isn't merely a tool for the present; it's a lifelong companion, accompanying you through various career transitions, job changes, and promotions.When individuals engage with Dextego, they are essentially investing in themselves. It's about desiring continuous improvement and having an AI coach as a supportive friend, guiding them through challenging moments. Dextego provides tailored micro challenges, enabling users to hone the skills they need over time. It offers a unique space for individuals to practice, think, and self-reflect before real-life situations unfold. Often, lack of preparation catches us off guard, leading to reactions we later regret. Dextego empowers users to plan, prepare, and navigate challenges with confidence because they have practiced, received feedback, and learned from similar situations beforehand. Like every single athlete will tell you, they prep in their head, they see the games before they happen. This is exactly the same thing but for life and for work.

Jing, the co-founder and CRO of 28Muses unfolds the journey from in-person to virtual experiential events amidst the pandemic. Highlighting the marriage of creativity and employee engagement, she sheds light on tackling workplace disengagement and redefining work-life balance.

Jing: We've been doing virtual events for teams since 2021. So originally 28muses, we were doing in person events in New York City as just like an experiential Immersive fun thing for New Yorkers. Then in 2021, we initially wanted a space, but then when the pandemic hit, and we were glad that we didn't have a space. When the pandemic happened, every team was just meeting on Zoom, and we realized, okay, some of these Immersive experiences that we did before, if we can adapt it for Zoom, it can be done for Immersive team building for offsite activities. So that's what we offer. Now, we mostly do team building, offsites, corporate retreats for New York City clients, as well as remote teams, and host it virtually on Zoom.I think it's a critical component of culture building and employee engagement. The new remote playbook where everyone is remote first.  We get together once a year or twice a year, or some companies every quarter, depending on how much money you want to spend. I would say like once or twice a year has been very standard. We recommend having a virtual team building once a month or once every other month. Sometimes, maybe if you don't have the budget, someone on the team can just host it for free, like a virtual little activity, a game for the team. Maybe hire someone like us, every other month to host something really highly engaging.We use science based facilitation to make sure we level the playing field between introverts and extroverts and get everyone to talk and create these activities that really get people to connect with just the right amount of vulnerability.  Sharing a little bit, but not over sharing. And for offsites and retreats like LEGO Serious Play is a great method for getting people to discuss an abstract or complex question, really peeling back the layers. Hosting in such a way so that there's no groupthink where everyone kind of piggybacks off of the same loud person who always talks first and who always dominates the conversation. LEGO Serious Play can really get a diversity of opinion. Japanese ink marbling happens to be one of our most popular activities.  It’s called sumi nagashi.  Once you make the pattern you’re making on floating water, you hand dip a coaster and then print.All of our activities are led by artists, all based on creative expression and with a belief that everyone is creative. Some people don't believe it, but everyone is creative, and these kinds of low barrier to entry exercises help boost creative confidence, help people actually connect over a meaningful activity and it's fun, right? So it works for building your culture, building your team, building long term bonds and trust.

Unlocking creative confidence

Neil: Having started during the pandemic, how has your overarching vision for the company shifted over time?Jing: I would say during the pandemic, we realized A, we had to just be really nimble on hosting, either in person or virtual or hybrid. There are more and more hybrid events. We just hosted off sites this year for New York Times and Capital One and DoorDash. Lots of big companies. We just have to be nimble and flexible. Meeting clients where they are. Some clients are 100% remote, some clients are not sure how people will show up. Some people might want to show up in person, some people might want to show up virtually. Just crafting this experience so that people feel like there's high engagement. What clients really care about is engagement, and when there's high engagement, then they tend to feel that there's a strong connection.But if engagement is low, and that's why we don't offer passive experiences like tastings–we just think that anything that's kind of passive where people are just listening to a lecture about wines or chocolates and then they get to taste but really they're not saying anything, they're not really participating. In my mind, it doesn't really build a direct, authentic connection. Or like cooking classes where everyone is step by step doing the exact same thing. You are active. It's not passive but if everyone is doing the same thing step by step, then there's no room for creativity, there's no room for expression.What ends up happening is you're just comparing how your dish looks like to the dish you see on the screen that a chef is making or what your painting is turning out like versus the artist on the screen. That comparison tends to lead to self criticism or judgment. So even though you're doing something active, a lot of people end up feeling bad about themselves and afterwards, I have so many people that voice, ‘oh, that was kind of fun, but mine turned out terrible.’ or ‘oh, it reminded me of art school art class and elementary school, where I felt like I'm a bad artist.’ So everything we design is intentional so that everyone's work turns out different, whether it's photography, animation, or the ink marbling.It boosts creative confidence because you’re not comparing how yours turned out with instructor for each other.Neil: It sounds like you cut across a couple of different categories for the Future of Work. Hybrid work solutions, employee engagement, collaboration.  How do you think about your category and how it will evolve over time?Jing: This year was tough. All those tech layoffs happened because during the pandemic, a lot of companies began to overhire because they were hiring to meet the demand. Then as soon as the demand went down, they had just too much fixed costs. I think for our category, for corporate events, virtual or in person, I think this year was tough because when the layoffs happen, one of the first things to be cut is fun things because they don't want the perception of, we're still having fun. It's such a somber environment because half your colleagues got laid off or this whole department got laid off or even it was like 10%, 20%.A bunch of companies, whether marketplaces or small businesses actually in our category, closed down this year or just decided to shut down. Even some bigger VC backed ones. So I think things will keep evolving. People are going back to doing more in person stuff companies always need. I think lack of connection and disengagement is one of the biggest problems right now. There's more and more news about just on an individual level,  about loneliness, and then at the corporate level, with disengagement with the rest of your team or with company values. So there's definitely still a need more than ever for services that help your employees develop that stronger connection to the company.That sense of belonging and for stronger relationships with colleagues that will lead to greater collaboration.  For us as a small business, we’re growing organically and we’re growing every year. SO we’re just working with people that we meet and referrals.Neil: How do you think about your growth for the next 36 months from here and what are some things you’re trying to be extra mindful of on that journey?Jing:  From a business development perspective this year we are experimenting with ads for the first time.  Everything has been totally organic until now and we will continue to grow organically but there are competitors out there and people are looking for team building or offsite activities through search.  So we just started advertising on Yelp, then soon Google Adwords, then there are a bunch of marketplaces like partners that we distribute through.  We’re not VC backed–I have 3 kids and my co-founder has twins.  For us, we’re living the future of work ourselves which is all about no longer being a slave to your company.  You really have to decide for yourself how you want to balance your work and life.We all have 24 hours in a day but how does that work for you? So personally, I homeschool my kids in the morning and then I work on 28Muses in the afternoon.  I love the future of work concept. It’s been around since before the pandemic but the pandemic brought it front and center and helped people understand what future work really means.  It used to mean robotics, AI, and how work was going to change but to me it now means how do we change the way we work both individually and collectively as a company.  That's why I created the future of work community on Upstream that we later migrated to Entra.So for 28Muses, the thing is, how do we grow while all of us have crazy juggling acts and grow organically while still having a great experience with our clients.  How do we grow organically and explore other ways of growth.

Drawing from her own personal experience of the challenges at hand, Cassie, along with her co-founders, started Archie with the mission to shape the future of the freelance economy by introducing essential financial infrastructure to level the playing field.

Cassie: Our initial inspiration for building Archie came from the realization that the future of work is shifting towards freelancers yet freelancers often miss out on the same financial opportunities as full-time employees because businesses lack the necessary tools to effectively manage contractors. As a result they tend to use outdated methods designed for working with B2B vendors, leading to issues like Net60 payment terms.From that, our mission was born; creating a more transparent, equitable, and improved working relationship that aligns with the evolving landscape of the future of work.Neil: So how did you and the team go about brining that initial vision to life?Cassie: We were addressing a major need in the market –streamlining payments between clients and contractors with a focus on transparency. We began with the realization that the problem for many businesses wasn't their financial capacity, but their inefficient workflow. Despite having the necessary funds, they often struggled to make timely payments.We started with a product designed to simplify this, empowering businesses to effortlessly onboard contractors, generate invoices, secure approvals, make payments, and consolidate all related information in one centralized platform. To kickstart our venture, we secured the support of key investors who not only provided funding but also leveraged their extensive network to help us acquire our first 10 customers.In this industry, trust and reputation play a key role, and having a product that genuinely addressed real needs was a major advantage.

Navigating the ways we work

Neil: What were some of the challenges you guys ran into at launch? What did you learn from them?Cassie: There were several, primarily stemming from there not being one “best way to work”. Many businesses had their own unique way of doing things when it came to onboarding, approvals, and payment processing. When you think about accounts payable as a function, usually it's 100% of the financial team's responsibility, but a lot of the work is actually happening outside of the finance team with the contractors or the workers themselves — and then the people who are managing those relationships. We recognized that to overcome these hurdles, we had to develop software that catered not only to finance and operations teams but to everyone involved. We wanted to make it effortless for each individual to contribute their part. This inclusivity was key to unlocking genuine efficiency.To address these challenges, we engaged in ongoing communication and collaboration with our clients, from publicly listed companies like Soho House to small mom and pops which this allowed us to gather valuable feedback, understand their specific workflows and pain points based on their industry and roles, and ultimately develop a platform with the adaptability needed to serve a wide spectrum of businesses at scale.Neil: So what were some of the ideas you originally thought to be true that you ended up finding out not to be the case?Cassie: We assumed that businesses would easily recognize the challenges of managing numerous payments manually
and the associated lack of transparency. However, we learned that many organizations, particularly those dealing with over 20 payments, were unaware of these issues because decision makers were removed from the process. We dedicated considerable time and effort to training and education. We believed that the impact of delayed payments and the absence of transparent working relationships would be widely acknowledged, but we've had to actively raise awareness about the potential harm to a company's reputation within the contractor and freelance economy.Additionally, Archie’s launch coincided with a period of heavy venture capital investment in startups, emphasizing rapid scaling and funding rounds, but we've since redirected our focus toward building a sustainable, revenue-generating business that reduces
the need for ongoing fundraising.Neil: You mentioned rapid scaling--how are you guys currently thinking about getting Archie to scale?Cassie: We place a strong emphasis on achieving repeatability. We believe that, beyond just focusing on ROI, a product that enhances the quality of workflow and makes it feel better to run your brand is crucial. Our strategy involves a combination of brand-building, testing, and innovative go-to-market approaches. We also leverage our well-known and respected clientele, such as The Standard Hotel, Soho House, Public Records, and Elsewhere, to tell Archie's story and raise awareness. We see the importance of becoming a valuable resource to our customers, allowing them to appreciate our solutions. Our deep understanding of our customers, from CFOs to events managers and DJs, plays a pivotal role in delivering excellent products and customer service, creating a network of touch points that establish our reputation as the best product for addressing the specific challenges they face. This is central to our approach for scaling effectively.Neil: So what are the pitfalls you and the team are trying to be mindful of as you chart that journey?Cassie: Our primary focus is maintaining excellent customer service and personal relationships with our clients, ensuring they have a direct point of contact rather than relying on automated responses. We prioritize being customer-obsessed — creating a product
that unlocks the most value for all users, both businesses and contractors.Additionally, we recognize that hiring the right team members at the right time can significantly impact Archie’s overall dynamic and growth trajectory. Therefore, we put a lot of thought and vetting into when and whom to hire.Neil: That's interesting. So when you're looking to make those key hires, how soon does company culture become top of mind for you guys?Cassie: Our approach to shaping our company's culture actually started way before we officially launched Archie. Yunas, my co-founder, and I were best friends, bonding over shared interests like food and music. So, even when we launched the company, culture was always a big deal for us because we genuinely cared about each other and our personal relationship. We felt a
responsibility to figure out what a great working relationship could look like and what it meant for building a team. That's why, right from the get-go, we focused on creating a culture that people could vibe with and would want to be part of – it was a key factor in bringing the right people on board.Neil: What are some trends you uncovered about your category you weren't previously aware of?Cassie: Post-pandemic, there has been a surge in people trying to reconnect with their communities and experience joy. What we've noticed is that many businesses, particularly those drawing crowds through events and entertainment, haven't implemented effective scalable processes to collaborate with contractors in a way that benefits both sides. They're working with outdated tools and spreadsheet-based systems, which might work for low volumes, but prove impractical for the long term.Moreover, within our industry, we've observed a pattern, when organizations recognize the need for a better approach, they often try to adapt through their regular accounting, procurement, or full-time payroll software, unfortunately overlooking the nuances and complexities around the independent contractor relationship.Neil: Love that. Last question, where would you like to see Archie go over the next 36 months?Cassie: My vision for Archie over the next 36 months is to have it adopted across businesses in the entertainment and hospitality industry. I hope to see the product continually enhancing its efficiency, providing businesses with more time to focus on what truly matters, spending time with their families or excelling in their job roles. We aim to equip businesses and individuals with the right tools for success and growth. Additionally, we'll maintain our commitment to transparency, ensuring that contractors can work without the stress of uncertainty around payments. My goal is to create an environment where the contractor-business relationship is harmonious, akin to being part of the same team, ultimately making work easier, better, and less stressful for everyone involved.

Having started from a personal desire to regain connection with a lost mentor with whom he had great admiration for, Suman started Personal AI with his co-founder to give individuals access to an individual's knowledge that reflects they are and grows with them.

Suman: It was my innate desire to have an emotional slash intellectual conversation with a lost mentor and a previous co-founder of mine. Due to pancreatic cancer, he passed away, and I had this mantra of, ‘what would Larry do’. That element of trust, being able to access somebody's knowledge, somebody's mind in all the different things that aspects in your life, I think carries that emotional trust as well as intellectual honesty, where a general knowledge base wouldn't fulfill. So yeah, so our PCC is every person has their own unique individual thoughts and minds that have potential to offer to themselves and to a group of people around them. That group of people around them could be small and could be huge, but everybody has their own unique mind.We wanted to create a tool to basically scale who they are as an individual person by creating these models that essentially reflect who they are and create a digital version of them. One that is their style, their voice, their knowledge, and continues to grow alongside them. So it's not like one pre-trained model as such. It's a model that continuously evolves with an individual person's life. That potentially will live forever into the future. That's our genesis story.

PLMs vs. LLMs

Neil: So you’ve been doing Personal AI for close to four years now which precedes the launch of ChatGPT and AI going mainstream. Would love to hear about how you went about starting and building that personal LLM and how you feel about AI having gone mainstream?Suman: That’s a good question that people ask me a lot these days. We actually don't do LLMs. We build something called personal language models, PLM is we call them, and technologically the big difference is that LLMs by definition are large amounts of data aggregated to create a model. They’re pre-trained, so that many people can leverage that model to fulfill some specific use cases that are public knowledge or domain specific knowledge. Our motivation was to create unique models for every person. So large language models would not fulfill our goal because they are large. They're not cost effective. They're not meant to be grounded in one person's data, and they have attribution problems. Moreover, it's not environmentally friendly to create a fine-tune LLM for every person on the planet. It wouldn't work.So my CTO, who is my co-founder for the company, when we came together we raised a lot of different approaches, different solutions, and her having an AI background from four different startups, she offered a unique approach that is not dependent on the LLM. That's what the three years of work has been; to create a set of models that are scalable, that are grounded, that are factual, that are continuously being trained, that belong to you alongside with the data ownership. Post GPT-3, and pre-ChatGPT, we started experimenting our application out there in the market and specifically within the creator space because GPT-3 kind of unlocked some startups and some use cases around content generation in our specific layer. The most interesting thing was, our application layer turned out to be a chat interface as well. So I think the UX choices that we have made, weren't too far off from what a multibillion dollar business could do.So then ChatGPT came along and that created some confusion. Personal AI has chat interfaces, what is the difference? There was a bit of confusion in the market even to our investors, there was confusion. Like what is their method? From that point, I think it has taken off to a larger degree and there are startups who started building within the AI space. For us, the mission, the goal, we didn't want to change in any way if possible. The good news is there is a lot of AI awareness on the potential of AI. The downside is, now there is tons of confusion. There are different approaches. But the upside is the core principles and the core attributions and the core problems that we were trying to solve with the personal language model, is what is designed right now for personal use cases that are not or cannot be solved by large language models.It's very fascinating to be living in the moment sometimes, and questions like this will help you step out of your zone and kind of think through what just happened, right? It's kind of the wild wild west right now and the core propositions of Personal AI, with the element of, private individual models that replicates an individual person with data ownership, data attribution that is scalable, that is efficient, those are the qualities that we bring to the table. People want to create models for themselves. We want them to choose personal AI, because the large language model is not us. It does not fulfill the desires of what your use case is. It does not replicate who you are as a person. There are some implementations of fine-tuned LLM, such as taking famous people on the internet and then creating a character around it or creating a composition board around it. That works because their data is already indexed by large language models, and our goal is not the top 1%. Our goal is to take every person on the planet whose lives are more private and more personal than what’s on the internet, and give them a tool, give them a choice. This is your model that learns who you are as a person. That is our goal. We want to talk to the 80% of the people who don't have access to the so-called fine tuning algorithms.Neil: So being that artificial intelligence as a technology is still relatively new, how do you go about helping your audience understand that difference between an LLM vs. a PLM?Suman: So right now, I'm still thankful for the LLMs because LLMs to a large degree also complement PLMs. It's almost like, you study books and you form your own opinions, but there is also the internet, right? We study from general knowledge that exists out there, and sometimes when it is very specific knowledge, a specific focus, we pick up the phone call and ask our expert, ask our mentor, ask our professor, ask our coach. So in a way, that is meant to happen; large language models are solving this problem of accessing information much more effectively, much faster, instead of traditional search. However, access to you as a person currently is social media, maybe direct messages, personal websites. So there isn't a specific platform or tool that belongs to you, that will scale you for other people to be able to interact with you. So, LLM 's are good because when you are in the training process with your personal language model. If you do not have the memory of the data, then LLMs fill in the gap, if you will, to provide things like graphs and some ideas to personalize it. It's almost like you are learning from other people and forming your own opinions.. So it's a compromise, more or less. LLM versus PLM is more about LLM plus PLM.Neil: So upon launch, how did you guys go about finding those first few customers?Suman: We launched this product in April after the LLM hype and AI hype that happened. We also had issues with identity. Like what personal AI means to us is different from what big tech companies may call a personal AI assistant. For us, personal AI is ultra personal. It is yours. It is your model, it is your facts. It's your opinion, it's your everything. So when people talk to me about personal assistants, it's not about getting one task done. It's actually creating a digital living thing of yours, like a digital version of you who you are as a person, that people can communicate with but it knows all aspects of your life, not just a very specific spectrum of your life.So the ideal customers had turned out to be very interesting, because after we launched, we let people come in, train their models, and the people who stuck around and got greater value out of it turned out to be more like lifestyle brands–people who care about personal development, personal brands, small business owners, creators, others consultants and coaches. For them, their personal brand is their currency. So if you think you are personal, if your currency is your personal brand, then it was nice for you. You can leverage it to create your model, scale yourself, make extra money, monetize with it–but at the same time, use that to develop something that is almost like your twin that knows you so well beyond what any person or human can comprehend. I think that's a beautiful thing.Neil: A lot of the products I’ve seen that I’d say operate in a similar space are aimed to go directly to corporations to improve productivity or anything along those lines. My understanding here is that this is more so direct to the consumer. Am I understanding that right?Suman: Yeah, it's very interesting. I think the unfortunate reality is we don't fit in a bucket, because we're talking about people, right? So if you're asking about who the buyer is, the buyer could be a business, it could be a corporation, it could be an individual, it could be a consumer as well. The way to think about it is, right now formaking revenue for the company, there is a need for these brands to be able to deploy and incorporate AI into their lives. They have a willingness to pay because they want to automate their communication. They want to scale themselves even more. They want to ideate even more. But it's not just for them; every consumer on the planet can actually start creating their own personal AI and start adding their memory to themselves starting from the learning process. So how we think about this product is not by the current existing pain points and making you 10x more efficient with corporate copilots as much as this is a next generation tool. Students are going to start developing their models, continuing to add memory on what they're learning and build it over a period of time. In the future everybody will have their own second trainer, their own Jarvis, however you want to put it, that they trust and that belongs to them. One that is not biased by any other motivations except yours.So, yes, it is a b2c product, but small business owners who have personal brands, professional reasons, coaches, authors, executives, CMOs at large companies, they also build their personal AIs because for them, their personal brand is also important. For example, an executive may have their personal AI for new employees to come and talk to them to understand more about the company, the culture, the values and onboarding. This allows that to be done in a much more connected way. Otherwise, they're just looking at the CEO or CMO in this 1000 person company once in a while and in one year, you don't know who you're looking for. So I think there are different revenue streams that are B2B as well as B2C, but the product is very much individual focused.Neil: So from when you guys started, what were some of the challenges you guys ran through to get where you are now?Suman: Tons of r&d. Even if you think about ChatGPT, we saw through ChatGPT after eight years or so. So for us, it was more about, how do you take this unstructured data, put in the structure, and grow this memory continuously? How do we ingest that memory to the model to ensure that the model is continuously evolving? How does it replicate your style? How does it ground in your tracks? How does it generate ideas and content? So there's a lot that goes into the model development itself. We were like human AI labs first, and now the product. It's almost like, you had open AI, and then you had GPT-3, and then you had ChatGPT, right? So we are AI labs first and now we have personal AI and Personal AI chat applications, communication applications, agents and so forth.Challenges? Oh my god, many. I mean, just convincing people that this is needed. Convincing investors that we need Personal AI for everybody. Two years ago, it was a very hard sell. Even being able to convince people that this is the approach and method that needs to be done. I think it's one of those things where until you get there, people won't support it. But the beauty is, there are few that see it, and they support you on that. That's your advisors, that’s your investors, your customers, or people that you speak to. That's what founders will live for. We look for that. Hope there's a light at the end of the tunnel. And guess what? Personal AI, something that nobody would have believed in, is now its own category. When we originally pitched ‘Hey, this is going to be a category’ we often heard ‘Let's talk about it later, when it becomes a category’.

Shifting behavior and nurturing adoption

Neil: So what are some things that you and your co-founder thought to be true at launch that you later found out not to be the case?Suman: This is an experience that we always debated. Whether or not people will talk to another person’s AI. Up until ChatGPT, that idea of talking to an AI was much more natural and seems to be accepted by almost everybody. Even technical people were like, ‘I don't know if I want to talk to a bot’.In our thought process, we thought people would talk to the AIs as well, but that process of adapting to that idea was so hard, because we know that shifting and the behavior is so hard. And moreover, with personal AI, the most interesting thing is, your model is grounded in who you are. In other words, if you ask me about biochemistry, I have no clue, right? So it is much more useful for people who are around me to get some context of who I am to benefit from it. Some context of myself, so it can draft messages for me on my behalf for any of the incoming messages or text messages or emails or anything of that sort that will come in. So instead of proceeding with human to AI interactions, we built this experience as human to human interactions with AI in the loop. So you always have AI in the loop, not the other way around where you're just talking to AI and getting the value out of it. No, you're actually talking to a human about that and your AI is assisting you and helping you know in the process.
If Neil comes to me and I trust him and he would have specific use cases to be able to talk to me and interview me for a podcast that you can actually do with my AI, then Sure, I would put myself in the autopilot mode where it's a automated, you know, Suman AI mode, which you can do right now, actually.Neil: What did you find was the biggest gap for those that aren’t early adopters of this technology?Suman: The biggest gap is awareness. Awareness in the way of identifying the truth, the reality of it. Ai is a hot market right now–the market is up for grabs, which is okay. Everybody would want to make money and that is fine, but having an agenda is not fine. You can put in capital and you can spread misinformation for a future promise, or a potential future promise, who knows? But I think right now the people I speak to are inundated with tons of different content, blogs, AI tools and Peronal AI is one of them now. So how do you really communicate the reality of it? At least people who pay attention to Personal AI understand the reality of it, so we are now back into not the need, but the attention grab, which is exactly what we’re trying to run away from. So the question is how do we get to a point where the truthful nature of Personal AI, to build an asset for yourself, is very well heard and out there, rather than conflating with every company having a personal AI assistant for a specific use case. The market is confused–who do I trust? Where is my data going? What is the difference between this persons AI and another persons AI? Every company has a personal AI product right now, but what company is working in favor of the people, that is more interoperable, that is trying to convert all these data aspects.There is no other digital asset that exists in the world right now. Even with all the AI hype, AI value and utility that has been created, it is still data grabs. There is no respect to what kind of techniques that you would apply to grab that data that belongs to you. We’re all about flipping that narrative, which is a hard thing to do.Neil: Do you think us as a society overcoming that is something that’s just a matter of time or is there something that needs to come from a more proactive approach from the people building products in the space?Suman: I think it’s a combination of both. If you think about what happened during the Linux open source moments, people wanted optionality, people wanted more control, choice to build. There was a fundamental shift in developing operating systems that are now fairly open, that are free, that people can develop and that eventually created value, but there is an element of always being able to build and grow good businesses within the walled gardens. You can be open about it and in a more democratic way of building things.So we can argue either or, but to your point, I think there are multiple different parties that are likely involved for the change that we would want to see within society and the most important thing is that people need to take control. I believe that when they do take control it is indeed beneficial to them in the long run.A good example is my kid daughter, she is six years old right now and shes already living with computers. It’s not rocket science for her to be able to talk to Google or Alexa. That medium ois already happening in our world now. Fast forward 20 or 30 years from now, do you want your kid to be talking to Alexa or mom and dad AI? Can you walk towards that future or are you going to let other people do the work? Where is that draw? Where is that connection that needs to be established?We all incorporate our phones to make calls to our parents and daughters and we incorporate the technology to connect. I think AI is a new medium that should eventually develop those deeper relationships and make ease of communication. Reduce the gap between people, which is the gap between hams and businesses. It’s a shift. It happens.Neil: What you just described is a very vivid picture of a future for us to march toward. As we try to make that shift as a collective, how do you and the team plan to position yourselves to be at the forefront of that?Suman: From a technological perspective, it has to be scalable. I’ll give an example–when I was 17-18 years old, I didn’t have access to a computer. There was no concept of computers at home. You would have to go access a mainframe computer that is likely sitting inside a walled garden and you have to rent five minutes of time to type and understand a document. It’s a shared resource. Over a period of time, the push has evolved into the idea of a personal computer–something to use for your personal needs.So what exists in our industry now are the mainframes–the language models shared by every person has to share and rent. What we’re building is the personal computer, Personal AI.Neil: You come from an engineering background yourself, your co-founder is the CTO, how are you guys thinking about your go-to-market and filling in any gaps you might have between the two of you?Suman: It’s about distribution–getting to the places people already are. The good news is these models will actually run on your computers and your phones and be accessible and cheaper for everyone. If it’s more economical, then it becomes more scalable and we can get it to the masses.So the current strategy is about focusing on the use cases that are economically viable today for the market and for that group of people to get immediate benefit from it. Over a period of time, penetration of the next group of people will start to grow. It will become more efficient and easier. The use cases will unlock themself.Neil: As you try to balance those things, what are some of the potential pitfalls you and the team are trying to be mindful of?Suman: It’s always a challenge because investors want to see founders and owners want to see the biggest impact. Finding a delicate balance on identifying the markets that make the most economic sense at this stage is a tricky part especially for a company like Personal AI where potentially every person could have this for themself. So we’ve done experiments for product market fit. We've seen the people who this resonates with, the personal brands. Finding those groups of people much faster is one of the things we’ll be mindful of for the future.Neil: What are some of the trends in the broader category that you’ve uncovered most recently?Suman: The trends are pretty fascinating. We’ve been developing the personal language model but there is a fundamental unit of this memory and when we designed it it was multimodal. It’s not just a text as much as it is something that will carry biometric information, visuals, things you see. So I think some of the macro things that are happening are that the multimodal approaches to AI interaction are going to converge.Even one year ago I remember talking about having a mind, which is the first language model, the voice which is like cloning, and you have avatars which model how you behave and act. You can bring all that to a digital presence. Metaverse is a good example of that technology.So I think over a period of time all these things will converge where you will have a virtual avatar living inside a metaverse that has your mind constantly self syncing with it and it speaks like you, looks like you, and lives in a virtual world. One that could potentially be projected into holograms and maybe robotics. So there are a lot of trends happening right now that will eventually merge into a single line to this inevitable future of human replicas that is like a living form of Westworld.Neil: Where do you see Personal AI going over the next 36 months?Suman: I guess 36 months is pretty forecastable. We obviously want to make the most impact so we’f say we’d want millions of Personal Ai’s out there in the hands of people.

Building assets not services

Neil: What is one thing you would want a reader to understand about Personal AI?Suman: Personal AI is not a service–it’s an asset that belongs to you. It’s like your home that has value that is occurring over a period of time.

Two brothers navigating a company pivot through a pandemic. Jon and his brother founded Scoop in 2015 in a world that is no longer, and revamped their company vision from scratch to solve new problems in the new future.

Jon: Our story is pretty unique when it comes to how we came to focus on the future of work. We originally started building Scoop in 2015 in a totally different world and weren’t thinking about the future of work, at least not in the way we talk about it today. The vision of the company for the first basically, five years was related to carpooling and commuting. We were building an enterprise carpooling marketplace. And the basic idea was that we felt that commuting was one of the biggest limitations on employee productivity and satisfaction and had resulted in an increasingly unbalanced cost equation for employees and their employers..We built the company around that commute focus from 2015-2020 and then COVID hit. Overnight, no one could go to the office anymore and our business ran into an enormous brick wall.It was from there that the inspiration to reorient and focus on the future of work emerged. We looked at how the world was going to change with respect to employee mobility and commute, and very early on, by mid-2020, we realized that the idea of people living in one place and commuting to another to go to the office was completely changed forever. We started hearing about it from our customers. We started talking to them about what buildings they were going to keep, how they foresaw using offices, and how they were beginning to think about what we now call hybrid work.And as we heard more, we said, you know what, we have expertise in the intersection of employees and their employers and how they think about achieving productivity and satisfaction in their daily jobs. Maybe we are in a unique position to build software that will help people be more successful as they transition into this era of flexible hybrid work.Neil: So you were four years into business when you made that pivot?Jon: Five and a half. We had raised our series C, and we were a 170-plus-person company as of January 2020. Then everything kind of stopped overnight. Our volume of carpools plummeted, not surprisingly. We navigated the first few months of 2020 not knowing what would happen with COVID like everybody else. And then it was in Q3, Q4 of 2020 that we said, you know, we're looking at a major trend change in the way that people operate and businesses operate with their employees. By December we basically fully pivoted into building software for hybrid work instead of commuting.Neil: Can you help me understand what that pivot looked like?
Jon: I mean, it was grueling, as you can imagine. Like many tech companies, we had to have a large riff in April of 2020 after COVID just to sort of get ourselves into a more defensive position. Then, by the time we got to the end of the year, we realized that we needed to basically restart the company from within. Basically, we started to set the clock back to being a seed stage company at that point. We were going back to the beginning of building a new core product--we weren't selling anything anymore.
We didn't have the need for much of the organization we had built up to support the existing business, and we reset our focus on understanding the challenges companies and their employees were having with the earliest stages of hybrid. Effectively, we started the company from within itself, but held on to our strong culture and values, some of our foundational technology, etc, to try to preserve the best parts of what the company had already created and accomplished.Neil: So, were there any customers who transitioned with you during that pivot?Jon: When we started with the future of work / hybrid focus, we essentially built our new customer base from scratch. The main reason was that our previous commute customers were predominantly major Fortune 500 companies. For them, creating a hybrid work solution was a complex and intricate process, deeply rooted in security and complex integrations. It wasn't an easy starting point for a new company. What did happen is we learned a great deal from them and their approach. It became evident to us that many large companies were developing their own solutions, which the broader market lacked. These solutions included office check-ins, capacity booking and management, and insights into workplace attendance trends and preferences. Companies like Amazon or Google had internal systems and engineering teams for these new needs. However, mid-market companies started actively seeking external solutions to manage those early stages of returning to the office.

Rethinking the go-to-market engine

Neil: What did the initial customer acquisition process look like for you when starting from scratch?Jon: For the first six months of 2021, we focused almost exclusively on product development without much go-to-market activity. Our initial touchpoints with potential customers were largely centered around user research, engaging with HR leaders and directors to understand their needs. During this period, COVID-19 was still a significant concern, and few were returning to the office. Interestingly, as 2021 progressed, the hybrid work landscape evolved dramatically, and our go-to-market strategy evolved alongside it. Initially, we aimed to create a solution primarily for HR, focusing on administrative aspects like desk booking, office capacity, attendance tracking, and vaccination status. But by the end of 2021, we underwent another evolution, recognizing that the hybrid model presented two distinct challenges: workplace management (what we’d been doing) and hybrid experience.Workplace management, which I mentioned earlier, is primarily an HR concern. Hybrid experience is a much more comprehensive look at the actual lived experience of hybrid employees and what goes into making hybrid work at an individual employee level. That means knowing when to go into the office, understanding work priorities, being aware of team members' locations, and managing their newfound flexibility to maintain (or better yet, maximize) productivity.. We eventually concluded that the workplace management arena was quickly becoming saturated and yet a comparatively smaller and less intriguing probably to small.. By the end of 2021, we shifted away from the traditional B2B motion/focus of workplace management tools. Although we had a strong core customer base in that domain, we decided to transition towards a product-led approach, focusing on solutions directly aimed at the end employee. Our goal was for that strategy to emphasize the software's popularity within the employee base and then lead to its adoption as the primary tool for managing hybrid work.Since early 2022, we've operated globally with a product-led, bottom-up approach, without engaging in direct sales. This strategy aligns better with solving end-user problems and has shaped our product’s development and growth throughout.Neil: Did you find that when you were building the solution, especially as a B2B product for end users, there was a lot of education needed about the problem initially, or did people understand it right away?Jon: It was definitely a moving target over the past two and a half years. Every time you think you know what’s going to happen, something changes – a new COVID variant, a vaccine rollout, or a new regulation. These changes caused attitudinal shifts within organizations constantly.In the last year and a half, there's been a lot of turmoil and tension in office return planning, specifically between employers and employees. We were constantly learning from both our HR contacts and their employees, trying to understand their evolving needs. It has felt like a whiplash effect – HR makes a decision, then employees counteract, and HR reconsiders or adjusts. We've seen companies change their return-to-office policies multiple times within 18 months due to competitive pressures, employee feedback, or government changes.Our role was to educate customers about the solution space. Many, particularly in the mid-market, had either given up on finding solutions or were overpaying for what they needed, often relying on spreadsheets. Our job was to help them understand what a viable solution could look like and why it was worth investing in.A common challenge was their nervousness about committing to something in a rapidly changing world. This uncertainty led us to focus on the end user. We were confident that a hybrid model was the future – not being in the office five days a week but not fully remote either. We aimed to create software that facilitated this balance. This approach helped us navigate and even moderate some of the fluctuations, providing a flexible solution whether employees were in the office two or four days a week, regardless of the company's requirements or location. This strategy proved to be a wise choice.

Anticipating uncertainty

Neil: What were some of the key learnings that were most impactful for you?Jon: It's kind of funny, but two major learnings come to mind that changed our trajectory. The first was at the end of '21. We realized that the workplace management software space was becoming heavily commoditized. There was little product differentiation – your desk booking tool, my desk booking tool, and others, they all seemed the same. This realization meant that such tools were becoming low value from a buyer's perspective.We quickly understood that these management tools would likely be integrated into other product suites. This has turned out to be true. Many companies entering this space have struggled, as everyone has some kind of solution. There are many options, which means downward pricing pressure. That was our first major learning (or aha moment).The second has evolved over the past year and a half, and it's about the ongoing uncertainty in how companies approach returning to work and the office. We've noticed that companies are often unsure about their operational plans.The funny thing is, this uncertainty itself was a key learning. We needed to create a solution that anticipated this uncertainty rather than trying to predict a final outcome. This mindset shift allowed us to build with the expectation of continuous change, affecting how we approach our product development.Neil: With the space becoming commoditized, how do you plan to differentiate yourselves moving forward?Jon: We caught on to this trend early and made a significant shift at the end of '21. We stopped developing traditional workplace management tools, abandoning efforts like vaccination tracking. We decided not to chase the ever-changing, highly specific company policy challenges. Instead, we shifted our focus to building solutions for the employee first. This approach is relatively unexplored in terms of tools for hybrid work and office planning. By focusing on the end user rather than administrative needs, we've managed to avoid commoditization and remain value-focused.Neil: When pivoting to this new approach, what were some assumptions you had that turned out to be untrue?Jon: Mid '21, I would've bet on workplace management software as a huge opportunity. That assumption was wrong; it became a standard feature rather quickly. Another point is our Flex Index, the largest global repository of companies’ hybrid work policies. We initially believed that the employee choice model would be the dominant form of hybrid work, but we found that more rigid models are common as well.We also thought that coordinating office attendance would be a major challenge for companies wanting to maintain employee choice. However, we've seen that most office-goers prefer certain days, leading us to evolve our solutions beyond just determining which day to go in. Now, it's more about coordinating who to meet or what to work on. This shift in understanding has significantly influenced our product development.

Not thinking about scale to get to scale

Neil: How are you thinking about scaling Scoop from where you are today?Jon: Regarding the future of work, we're seeing some interesting tailwinds. It's a global problem, being experienced and learned about simultaneously worldwide, which is quite unusual. Most tech trends occur in geographic waves, but in this case, it's global from the start. Our platform is inherently global, already serving employees in over 40 countries.When thinking about scaling, the candid answer is that my focus isn't on scaling per se. Our focus is entirely on the strength of our core product value. We're not yet concerned about scaling in the typical sense. Our challenge is ensuring our product becomes indispensable for hybrid work. If we succeed, the mechanics of scaling – sales, organic growth, paid growth – are relatively straightforward. Our strategy revolves around solving an end-user problem in a unique way, leading to concentrated value and, subsequently, natural scale. Our growth today is steady and entirely organic, and we’re very happy with that as we make the product meaningfully better each week.Neil: Where do you see Scoop heading in the next 36 months?Jon: Our plan is to grow, but there's a difference between thinking about scale and actual growth. In the next six months, our priority is to continue refining our solutions for hybrid work. We’re focused now not just on the “where” but the “what” which is an exciting frontier, in terms of focusing employees on how to channel their flexibility into the most productive usage of their time. Once we feel we’re really delighting our users, we'll evolve into a phase more focused on scaling and identifying specific customer bases and sub-markets to target.However, if we’re being honest, I don't usually plan 36 months ahead. Our world changes too rapidly. I have a vision for our product that addresses our core problems over the long-term, but we're mostly focused on the next 12 months, adapting as needed.Neil: Focusing on the next twelve months, what pitfalls are you trying to avoid?Jon: Having previously built this company once, I'm aware of the pitfalls, both unique to our business model and common in startups. We've structured our company to concentrate our risk in one area: execution risk. This means we're confident in the market size and the problem's validity; our challenge lies in executing our vision effectively.We're certain there's a significant problem with hybrid work, especially around time management and the balance of flexibility and productivity. The question for us is how well we can execute our product roadmap, gather insights, and build the right product quickly with the right team. Our primary concern is excelling in execution. We've chosen to focus on this because if we fail, we want it to be due to execution, not because we built a product that wasn't needed.

Sail into storms

Neil: What's one key takeaway you'd want everyone to understand about Scoop and your journey so far?Jon: The most significant lesson from focusing on the future of work is the value of turning into the storm. In any major shift in how people operate, spend their time and money, or interact, there are always big challenges to solve. A company should aim towards these challenges and work to figure them out.A second point is that we didn't always know the exact problem we were solving, but we knew we were heading toward valuable territory full of unknowns and challenges due to the massive transformation people were experiencing. Reflecting on the future of work, I think the difficulty for most companies and people is adapting to such a large change in work dynamics. Many companies rushed to find solutions and consequently failed. They were too hasty with return-to-office policies, too strict in certain areas, or overreacted in others.Those who took a patient approach, acknowledging the magnitude of the change and refraining from hastily solving problems, tended to be more successful. We've tried to mirror this in our company, watching how things unfold rather than rushing to conclusions while the world is still in flux. Many companies sprint towards a perceived finish line, only to realize that the context they were aiming for has shifted. That's been a huge learning for us, and we've leaned heavily into this approach.

Started by solving his own problems he was experiencing running a marketing business, Vinay seized opportunity with Process street after learning that his problems weren't unique and there was an appetite for what he built beyond his network.

Vinay: It was really just solving my own problems. I scratch my own itch. I was running a marketing business. We were running a big portfolio of websites, a lot of repetitive work around advertising, content creation, website maintenance, and I was doing all that with spreadsheets and Word Docs and freelancers and virtual assistants. It was just very chaotic and messy and I wanted software to simply help manage that work which was very repetitive in nature. We weren't recreating the wheel.We weren't designing new products or concepts or trying to figure out product market fit. We're just doing things that we knew exactly how to do. We just needed to do them frequently and the right way but the right quality on time by deadline, very, very basic stuff. Why is this so difficult to manage? Part of it was before COVID. I've been running remote teams since 2009. We were using Slack. Just like before Slack started we were using Skype for communication and Google Docs and there wasn't really much else out there in terms of collaboration software.So there was definitely a need in the market for it. We were one of the people with the needs. So we first built it as an internal tool just for us because we knew that we wanted it. Started showing it around to some of my other entrepreneur friends and they were like, Oh, this is awesome. And so that was kind of the beginning.Neil: So beyond those first few friends that you shared it with, what else went into getting that initial launch off the ground and what did those first customer acquisition pieces really look like?Vinay: We kind of kept it in like a closed kind of stealthy beta for probably four to six months, getting the very basic stuff figured out. Just the core features and product to make it even usable. During that time, we really only showed it to very close friends because there were a lot of rough edges. When we did launch, we launched probably a bit about six months after we started writing the initial code. I think we did the initial launch on Beta List. I don't even know if that's around anymore but it was basically a site that kind of like before product time, so they would show up and come like products that had launched or were just launching them, that were still in beta.Basically that was the idea, and they would email out your product and we didn't even have the ability to reset your password. I remember when we launched, we went on Beta List and it was just like, it was silent up until then, we had five friends playing with it or whatever and then that day we just got hundreds and hundreds of signups and our phones were just pinging every couple of minutes with signups and then support tickets started coming in and were like, yeah, I can't reset my password.Neil: Other than the resetting password issue, what other challenges did you guys run into when launching that first iteration of the product?Vinay: The very first feature was basically repeatable checklists with content. That was essentially what the product was. So you had a checklist, like this is a set of tasks that need to be done to complete this process and complete this job. Like an SAP document, these are the 10 steps, but we broke out each step in the document into a task that you could track whether it was completed or not. Then inside each task, you could add content. So I think we could just add text and images as the content, but we kind of essentially broke out a document into a trackable, completable kind of step list.That was the first thing that we launched. So essentially, you could add tasks and you had inside tasks, you could add images and text and that was it. Then what we quickly found out was that there were some very basic business processes that are just a straight list of tasks with, you know, a minimal amount of instructions that are needed. But there's this whole world of way more complicated processes where there's actually a lot more value in those processes, because they're not that simple. The fact that they are complex means that there's more value attached to them.That wouldn't work with that structure of just a very simple you know, A to Z like basic content kind of UI. The next thing was what are my other processes in here but I can't because they have branches or they have loops or they have forms or they need video or they need these other tools put into them or whatever. So basically, we started to then try to solve more and more process use cases and kind of like going after increasingly complex processes. That's kind of still going on today. Many many years later, you know, we've gone from now we can handle any SMB process, no worries.Now we're dealing with very complex, 1000 steps, hundreds of people, multiple departments, legacy systems, AI automation like processes right in enterprises. It's just the complexity that just keeps going. It's like how do you solve increasingly complex processes that are branching out from individual teams to multinational organizations to even cross organizational processes, things like that?

Continual adaptation to stay competitive

Neil: So what is your vision for how this will affect the everyday worker?Vinay: From a larger paradigm or technologic shift, I think we're still on like the continuum where over time, businesses need to adopt new technology to become more efficient. That efficiency drives productivity and competitiveness. If you were a bank today, but you didn't use computers, you had no app, you had no web and internet banking, you had no way to manage anything or get your statements online or anything. You just wouldn't be very competitive against the banks that do adopt technology.That's from a product experience perspective. Then you could say the same thing if you were on the back end of that bank, if you were doing all your stock trades with a paper ticker, and calling in your trades, or mailing them in or something, and you're trying to compete against like an algorithmic AI hedge fund bot, that can make decisions in like fractions of a second. The one that's using technology is going to out compete and outperform you. You're going to need less people to do the work, and you're going to be able to move faster and have faster data flows and communication and decision making.So, basically as new forms of life, efficiency come out, businesses need to continuously adopt them or else they fall behind competitors that don't. Technology continues to come out as new layers and platforms develop. The computer unlocked a whole bunch of technology, then the internet unlocked a whole bunch of technology. Then the phone unlocked a whole bunch of technology. JavaScript apps and API's and AI, right, they're all just different platform unlocks that unlock a whole new layer of efficiency, potential, that people build products on or utilize.Then companies need to adopt that to stay competitive. If they don't, they're going to fall behind their competitors. So I just see all these new kinds of waves of automation, just us continuing to basically push that envelope. How efficient can we get? How much value can we deliver without using as much human labor or human resources on this particular task so that humans can go off and do the next thing that is not solved or is not able to be completely automated away. So as the future of what platforms, no-code products, AI, API driven tools, whatever order they're coming in, they're just part of that equilibrium continuum that continue to push the envelope on how we drive automation, efficiency and productivity for humans.Where we're at today with the technology that's available, we're seeing that lots of companies still have a lot of enterprises, but lots of companies are now using largely a stack that has open API's. So now when all your tools are using open API's, you're able to connect them together and run all sorts of interesting automations against them that drive efficiency. Now we're seeing this AI layer come in whether you're an AI native tool, or your normal SaaS product, that's layering AI like functionality, which is what we've done, and which is where I actually see most usage of AI coming from. It's not actually from people using AI through ChatGPTs chatbot.I think most people are gonna be using AI in Gmail, in Word and Excel, in Slack just as part of the tools you're literally in all day, every day already. Whatever you're using all day already. Integrating that into the tools you're already using. That's what actually I think, the most usage is. It's behind the scenes somewhere. It's like, when you're looking at the TikTok algorithm or you're doing a Google search or something, it's just using AI in the background to drive more efficiency in those workflows.It's just another layer of life. AI is unlocking a new layer of automation. Some are gonna have larger impacts than others, some platforms are more impactful than other platforms. It's hard to tell exactly which ones will be until times past a bit, but they also need to work on top of each other. So they need to work with each other to get the full value. If you look at something like Zapier's AI feature, Zapier is using the computer, the internet, the browser, single page JavaScript apps, API's and AI.If they didn't have all of that stacked, they wouldn't actually be able to provide the value they're providing. If they were just an AI tool, but there wasn't an open API ecosystem, they wouldn't be able to provide the same value if that kind of API ecosystem wasn't created before. So I think we just keep stacking these technologies on top of each other and then those continue to drive more and more productivity. It's gonna keep going, as new platforms come out, then new ways of driving efficiency from those platforms will be discovered.Neil: That's an interesting take because there is definitely an influx of new tools entering the marketplace now but you're saying most of the transformation we'll see will be on the back end of the tools we're already using everyday.Vinay: There'll be both, right? There will be new tools that come out that are doing things that you previously couldn't do in a totally different way, like thinking of some data categorization tools or data insight tools that maybe you can now get these interesting data insights by running AI across these different data sets. There's a new data Insights tool that you're now using that you weren't using before because that has a new kind of model of data that you can access that you couldn't have accessed before. That's an example.So maybe you'll switch up and play a bit with this new AI tool that's just got a better data profile than you had before. That's an example of maybe you will adopt a new tool, but I don't think that's where the majority of people are going to be spending their day to day when they're using AI. They're going to be in Word. They're going to be in Gmail or Salesforce, where they are all day already. But there's going to be AI features incorporated into those tools. If you go out and say I'm going to try to build an AI CRM, I can start as a new startup trying to build an AI CRM. The functionality that you need to catch up to Salesforce, your gap is so much larger than the gap that Salesforce has to catch up to your functionality as an AI CRM. Salesforce's mobile app has hundreds of engineers on just their iOS app. You're so far away as a new startup from catching up to that to be able to really sell your CRM to 300,000 users at CitiGroup. It's much easier for Salesforce to add AI features into that tool than it is to build the entire business of something that is able to sell a deal like that.Neil: The Play Bigger consulting group talks a lot about this. It being hard to catch up playing in a game dominated by what they call a "category king".Vinay: It doesn't mean you can't build a successful business by building an AI CRM, you can build a business that gets 10s of millions in revenue and SalesForce will even buy you, right? But you're probably not going to build a product that is used by most people in the world, using AI. Probably Salesforce will be that CRM that still has the most amount of users that are also using AI all day. Say in five years, where the sales reps you engage with AI the most in the world, it's probably going to be Salesforce. There'll be plenty of smaller CRM products around doing stuff here and there, but they just won't have the market share.Neil: Fighting over the bottom 24% of any market.Vinay:. And that's fine. For most people that still could be life changing money. You go from being employed to never working again. That's a pretty good outcome. I'm not saying I think it's a bad idea to start businesses in those spaces, it's just the way that I'm looking at it, it's where is it gonna be the most usage. I think it's still gonna be that.Neil: I'm sure you've seen a lot of companies pop up that will probably sit in your same category around workflow automation. I'm curious how much attention you pay to those new entrants and how you think about competing over the next 5, 10 years, whatever the life cycle, you would like to be for this?Vinay: I definitely notice a lot of competitors come in with tangential products or direct copycats. What we've found is that they tend to be going after the verticals that we've moved on from. So without sharing too much of what our strategy is, we started as a very SMB product. We were initially selling to very small businesses doing pretty basic processes. Not a lot of security, not a lot of regulation. Marketing agencies are a very easy to enter, low barrier businesses to get into.Over time, as our products matured, and we've invested 10s of millions of dollars in building out the product and enabling all these more complex use cases, getting our security way more mature, SOC2, HIPAA, these types of things, we now go after a much different customer base. What we've seen from a lot of competitors is that they come in after us and they've gone after our old customer base while we're already thinking about the next customer base. I think that's pretty much how we will continue to compete. Us figuring out where we can add value in new market segments that we are not yet extremely competitive.We can offer a lot of value and focus on those segments to get growth while our company's got to go off to the segments that we're already moving on from. I think it's gonna be like that for a while. There's still so much like greenfield space in taking companies from legacy environments to modern, API first, AI driven workflows. There's so many that are still running. So many companies are run on mainframes, giant spreadsheets, or even still on pen and paper. Things get written down, get scanned and go in.So it's gonna take decades and decades for all that to completely move off and the savvy tech forward SMBs move faster, but there's a lot of larger, slower organizations that are still going to take decades to get everything figured out. We're working on identifying where those opportunities are and going after them.

Maturing the business alongside the product

Neil: I'm curious how you handle that split of those companies that get it, that want to be adapt to those new technologies really quickly, versus the ones that are a little slower. Do you take an approach of, hey, how do we educate you on how important this is? Or do we just dump them and focus on them at a later date?Vinay: The companies that are not as fast at naturally adopting require more hand holding. So part of that strategy is how do we make education and adoption much more palatable for those sectors or segments? That's been doing things like building out a partner ecosystem so that we have partners that are able to work much more closely with those customers, that understand those customers at a deep level and understand their businesses and industries. It actually moves into more of this kind of people business than pure technology and that's kind of the argument actually around why it's so hard to displace someone like Salesforce. Even if you build a better product, there's this whole other side to the business that's very, very difficult to clone. You can't clone it with engineers, you need tons of capital and brand and trust and lots of people with all sorts of different skills, to be able to actually not only physically, close a big deal in the enterprise, but even just to give the enterprise confidence that you can service them in the first place.Then assume that you have a big team of partners and technical specialists, implementation specialists and support people and sales reps. Companies will think, How are you even going to service us? You don't even have the people that you need to get this deal done or to make this deal successful. So a lot of it is maturing the business to be able to service these types of deals. And a lot of that requires more kind of business maturity versus product maturity, right. So it's a lot of work.Neil: You mentioned being able to instill the confidence that you can actually get the problem solved for them. Is that something that you instill proactively as a part of a sales process or really just in the makeup of the company?Vinay: It's both because during the sales process, they investigate the makeup of the company, right? So they're assessing that as well as the technology, because they want to know that you're going to have the people and the resources or the expertise to implement and drive change management and deal with the political entities that need to be convinced to move things forward. So yeah, through the sales process, you want to do that by introducing all the people they're going to be working with both through the sales process and through implementation and ongoing.You want them to win over the customer on a role by role basis. So you know, that your decision makers are the implementers, the technical experts, and the legal experts are all kind of like dealing with the right people in your organization. Everybody's having a good experience. Everybody's gaining trust through that process. You know that your partners are skilled and qualified and able to do the work and they have a good working relationship. So through the sales process, you're doing that and then you're also trying to do other things to show business maturity, like telling them about the other customers that you have.Telling them about other projects similar to them that have been successful is harder at the beginning when you're just trying to get these initial customers for the first time. When you're trying to get those customers for the first time, you don't have reference customers to show, you're trying to close your biggest customer so you have no one bigger that you can show them. Then a lot of that comes down to hustle. You just got to show that you are more reactive, you are going to move faster.You are going to deliver higher quality work, you are going to put in the extra effort. You are going to really do what needs to be done to make this work. Whenever we're dealing with really large customers, we always try to fly and meet them in person and put in the extra effort. Sometimes a customer will say we don't want to be your biggest customer because we don't want to be the person that you're learning on. But sometimes you'll find customers that are like, a 10,000 person or a 15,000 person organization, somewhere in that range, and for a lot of startups, that could be your biggest customer.Against really large enterprises that's still pretty small. Against an Accenture that has 750,000 people or something, a 15,000 person organization is actually still pretty small. So sometimes they'll say we don't want to be the customer that you're learning from. But sometimes they'll say, we're still pretty far down the list of priority customers for your Salesforce or whatever when they've got CitiGroup and 50 customers with more than 300,000 employees. We're down here with 15,000 employees. They might actually like that you're putting in all this extra effort, that they know that they're going to be your number one customer.They know that they're going to be in a Slack channel with the CEO, with the CTO, with people that can really move quickly, make decisions and move that project forward. Prioritize that project over every other project because they know that they're the biggest customer. Whereas if you're a 10,000 person organization, you're buying from Salesforce, you're never talking to Benioff, you're never talking to the CTO, you're never talking to the heads of product. You're many many layers down talking to sales reps somewhere in the organization still.So yeah, you got to find those ones that are willing to give you a shot and have a big enough need, that really want that white glove service directly from the founders and that the leadership team can make that project successful. Then once you get those, you can then use them as reference customers for your next deals. It requires some hustle in the beginning.Neil: How involved are you still in some of these sales processes for some of these larger clients you guys are trying to win?Vinay: I always get involved in the top three to five deals of the quarter. I just drove down to Philly on Monday right now to try to meet one of our clients who are in the quarter. If it's the top batch of deals then I tell my reps if you need me to do anything to get these done, I'm in, let's go.

The product-led growth wall

Neil: What were some of these things that you had originally thought to be true when you were launching Process Street you later found out to not be the case?Vinay: That we could continue to grow as a product led company without needing more sales, customer success, sales engineering, partners, solutions consulting, etc. We could get this product deployed into complex workflows and use cases without needing people to do it. That it could just self-serve quickly.As we tried to tackle more complicated processes in bigger companies it was very clear that there's a lot more money to be made, and it's easier to sell, to activate , to retain , to expand customers, when you actually layer on a lot of these people resources versus letting them just run on their own.As part of that realization, one of the other mistakes I made was believing that we could get maximum value out of the product through a self-serve pricing mechanism. Pricing is closely attached to PLG. If you want your product to be self-serve, you have to have self-serve pricing, right? Which means your product has to have some type of value metric that's closely attached to some cell in your database.Stripe as an example is after payments. It actually makes a lot of sense, the main value that they're providing is a transaction that's going through, and as their transactions continue to grow, Stripe's revenue continues to grow with that client. That model works great as a self-serve model. There's many many products where you can't attach the value of the product to some cell in the database and where that is common is when you're dealing with really high value, low volume workflows or jobs to be done.Healthcare is a good example, when maybe you're only doing a few surgeries every year or something and you had some software that's helping someone do this surgery. If you were to attach it to just the number of surgeries that surgeons do, you might have some surgeries that are very infrequent or really high value, like a cancer surgery or a heart surgery or something. You might have surgeries that are way more frequent, but lower value like tonsillitis. If your thing was the number of surgeries performed, and that's the number in your database that you're attaching it to, you're gonna undercharge the heart surgeons and overcharge the tonsillitis surgeons because one is doing something more frequent, but it's not of value. Suddenly you're charging this tonsillitis person, you know, 50 weeks worth of things and then you're charging the heart surgeon three weeks worth of things, yet the heart surgeon saved three lives. I don't know if that's the best example but there are lots of things that happen in the world that you can't directly associate the value against this number in the database. And that's where things get confusing.For us, a good example is we have wealth management processes. So we have some of the biggest university endowments as customers. They are small teams of 20 or 30 people that make hundreds and hundreds of millions of dollars a year, and they are only doing a few investments a year. They're not doing a lot and their activity is very low. Small team, low activity. Think Warren Buffett and Berkshire. He makes like two decisions a year and makes $20 billion or something, whereas then we might have a call center in the Philippines where their volume is really really high. They have lots of activities, they're running, lots of workflows, and lots of tasks. Lots of people, lot's of data, right?So if I was to look in the database anywhere, users, activity, runs, integrations, tasks, just any metric that I have in the database, the Philippine call center is going to be way, way higher. This user is way more active. If I was to try to attach pricing against one of those metrics, I would end up charging this call center that maybe has a million dollars in revenue total per year, hundreds times more than I would charge this fund that makes hundreds of millions of dollars a year.So the pricing doesn't always work. Based on your product, it doesn't always work to attach price to something that you have in the database. Maybe it needs to be more attached to the actual value that the customer is receiving, which may not be something that you capture in the database. That's what happened to us and that made our self serve pricing not work where we were overcharging people who are doing these high volume, low value processes and under charging people that were doing these low volume high value processes. For us, there were a number of reasons that PLG didn't work, even though we really wanted to.Neil: So with where you guys are at now, how do you envision your journey through scale over the next 36 months? And what are some of the pitfalls you guys are trying to be mindful of?Vinay: We're figuring out segments. Going after those segments that we think there's good opportunity in. Largely, it is going after larger companies, larger deals, more enterprise transactions, and scaling that is going to require unlocking new channels that didn't get us to where we are today. When we were more of a PLG company, it was a lot of the wide net marketing stuff, like ads and viral hooks and those types of stuff. As we move more on the market, it's more traditional. It's more Account Based Marketing. It's more channels, like going through a partner channel. It's more enterprise sales teams. So really refining those processes and getting them efficient around the new channels and getting the CAC down and the scale up is what we need to figure out how to crack to continue growing.

After having been a people leader himself, Josh and his co-founder saw a problem with how performance reviews were done and went all in on building a solution better suited to the new world of work.

Josh: I spent six years at a company called Carta. I was the chief product officer there. I was a people manager of course, and suffered all the pain of performance management, like writing the performance reviews and having them be reviewed. It's the office politics that come with being reviewed. It's the calibration sessions where all the managers get into a room and decide on the final ratings together, which is really, it's supposed to increase fairness, but it's really just kind of a Game of Thrones-style political event. When my co-founder and I got together and decided just to start a company, he had all those pain points as well. He was the CTO at Doctor.com for a similar period of time. He had all those pain points too and we said, you know, maybe we can actually make a contribution to this and help to fix it.

Layovers in Anchorage

Neil: How did you go about bringing your initial vision to life and finding those first customers?Josh: We had a little bit of a weird founding story. So my co-founder and I both live in the San Francisco Bay area, and we live in the South Bay. This was December of 2019. So he lived in Mountain View and I lived in San Jose. We decided when we started this company that we needed to be where the action was. We needed to be in San Francisco because that's where everything was happening. We kind of left our lives behind and rented a corporate apartment and we moved in together to start this company.We were really excited about it. If you read a lot of the Paul Graham essays he talks about, co-founders should live together because that's the fastest way to compress that iteration time. What ended up happening was we moved to San Francisco, I think in January of 2020, and then, we were building a product around reference checking. That was our first product. Then we raised a seed round that closed in March of 2020.Then, of course, what else happened in March of 2020. The world shut down. It was the two of us sitting in this high-rise apartment in San Francisco looking down on the empty streets every day. Building a product--we had money in the bank. We had like 20 years of runway if you calculated it. We're building this product around making reference checking more efficient in a time when the US economy was losing a million jobs a week so no one was gonna buy reference checking. It was just like the whole world just turned upside down.We went through this whole exploration process of what do we really want to build? What will the market actually reward us for building, and we probably spent five or six months trying to figure that out. It wasn't until probably August of that year that we actually got our first customer, it was Carta. We went back to them, and the CEO had an idea and he said, there's this thing I really want. I want to understand the skills and behaviors that we have in our company. If I want to find somebody with a series seven license who speaks Mandarin, I want to be able to do that right away, but I don't have a way to do it. My internal teams are too busy to build this but if you build it, I will buy it.That ended up being the first thing that we built and the first company that we built it for. It's not what we do today, but it just is this kind of reminder that, building a startup is like flying from LA to New York with a layover in Anchorage. That was our Anchorage. In this way, way out of the way stop that we had to make it to get to where we wanted to go.Neil: So I've found that companies typically fall into one of two buckets in regards to how 2020 unfolded for them. There are those who were uniquely positioned to serve with what they've already built and those who had to drastically change how they were doing business. Being that it sounds like you were in the latter group, I'd love to dive deeper into what that pivot looked like for you.Josh: The pivot for us was we had built this product that was kind of this internal resume. So you could create a project that you've worked on with some coworkers, add it to your internal resume, and then you'd be able to tag the skills that you used and talk about your actual contributions and comment on other people's contributions. What we were trying to do was build up this skill and activity bank. Who is actually doing what in this company?We released the product and we had some good adoption. We could probably reach 12-15% of the company organically but it wasn't exploding. If everyone in the company had a profile on our platform, it would be amazing because then you could search for the skills that you want, and you would know that you've searched everyone's skills, versus if you've got 12 or 15% of the company, that means that 80% of the time 85% of the time that you go to look for a skill or project, it's not going to be there.So we thought: if we could get 100% coverage, we could create this sort of searching activity, and then there would be this flywheel effect that would take off. What's the one thing that could get everybody to go in and write in and enter what they've accomplished? That's the performance review. That's the performance cycle. That's one of the things that you do; typically talking about what you accomplished. We went back to the customer, and we had a great relationship with them. We said, hey, we know that you have a performance review coming up. How would you feel about doing it on Confirm? We'll build the software for you.We had an amazing champion there. She took a really big swing with us and said, You know what, you can do it, but we're opening up a performance review in six weeks. And by the way, we need two weeks for training managers, so you have four weeks to build this software. Now we had no code, we had no mockups. We had no specs. We had nothing but we were gonna run a performance review for an 800 person company that wanted 360 reviews, they want to manage ratings, they want to calibrate, they wanted a really complete platform.My co-founder in particular who was the more technical of the two of us by far, worked 23 and a half hours a day for about four weeks. Got this software up and running, and it was down to the wire. We were building in tandem with the four phases of the review cycle.In other words, like the first phase would be everybody, the second phase would be managers, the third phase would be HR, and we were literally shipping the software in phases, meaning that while everybody was doing their inputs, he was building the software that was going to be used the next week for managers. Then during that week, while managers were doing it, he was building the software that was going to be used by HR. So it was literally like if he had gotten sick and missed a day, it wouldn't have happened. It is like laying the rails while the train is barreling down the tracks.But that was the first performance review that we ever conducted, and that actually turned out to be the core of the product that we sell today. Turned out that we were successful in getting everybody to create these profiles and these resumes and all their skills and then nobody cared. We ended up kind of throwing that product away or at least putting it on the back burner. What we realized was everyone was really leaning into the performance review data that we were able to generate, which was new.We had the CEO logging in, three or four times today to read performance reviews, and we didn't expect that at all. That was actually when we realized we might have something here but it's not the thing we thought we had.Neil: Performance reviews themselves as a concept aren’t new in nature. How much education did you have to do with new customers about how what you guys were doing was different, and the value in it, versus how many people really just got it off the bat immediately?Josh: Really good question. So in the first performance review that we did, that customer champion there, she came to us with a couple of problems. One of those problems she said was 'we've had a couple of instances where we've promoted people and then learned something about them that was really bad, and then actually had to manage them out even just a month or two later. Then we also have this other problem where sometimes we don't promote the right people because we don't have the data that tells us that they're actually really really good. Can you help us solve these problems in the performance review?' So we said Yeah, sure, and we took a crack at it. We returned to the science called organizational network analysis, which essentially says that anyone in the organization can review anybody else, and it can be a positive review or it can be a critical one but the point is, that's the way we're actually getting things done today. If you look at traditional performance reviews that are manager-driven, they're 100 years old. I mean, they were built for a time when we were all sitting together in an office, we didn't have computers, we didn't even have telephones.The world of work today looks completely different and a lot of things happen outside of the managers' visibility, but we still have to make those decisions: Who do I promote, or maybe who is struggling? So, we turn to organizational network analysis as a way of getting more data. The customer loved that. That was the part of the product that they leaned into. So when I talk about the CEO logging in three times a day to read performance reviews, that's what he was reading. He was able to see: why do 12 people think that Mary is just doing outstanding work? Or why have five people raised concerns about Bob and said that he needs some help?That was data that not just the CEO but the organization hadn't ever seen before. We pretty quickly back-burnered the rest of the product and just focused on bringing organizational network analysis into performance reviews. Nowadays, that's the foundation of what we do. Like we are the ONA company. We make performance reviews faster and fairer using ONA. There is an educational piece of that. I think that for every successful startup, you can sort of break the success into two things. One is the company's ability to execute and then of course, what are they actually executing on and things like that. What's their go-to-market? But the other one is this larger trend, think like Tesla and electric cars, it's not just that the time was right for Tesla, the time was right for electric vehicles. The battery density and lithium-ion batteries had actually gotten to a point where you could make a viable car and Tesla decided to do that. So they propelled this trend, but the trend also propelled them.The trend we're banking on is: can we make the case that organizational network analysis is actually the right way to review people today to actually measure their work in a time where work is more distributed and networked and collaborative than it's ever been before? We're all trying to do more with less. We're trying to grow, you know, revenue, but not grow headcount, which means you've got to get really good at understanding where those payroll dollars are going, where they're making an impact, and also where they're not.Neil: I see a lot of different tools that are being built to support that collaboration and remote first work environments. I'm curious as to how any of those trends play into what you're building and whether or not you found it more difficult with distributed systems, to get that quality kind of review in place for employers.Josh: The idea that we're all gonna go back to a five-day work week in the US is a fantasy--that's not gonna happen. What I would say is that the trend of remote and hybrid work is a tailwind for us. Two things are kind of at play. One is that because we work in networks like Slack, I can go on to Slack and message anybody in the organization and they can direct message me. If you weren't using a tool like that before the pandemic, you're using it now.The second trend is, because of that, managers just don't have the visibility that they used to have. When I was in college, they taught me to manage by walking around. Nobody is showing me how to do that on Zoom. I just can't be in every DM. So the employee touchpoints have exploded and manager visibility has just gone down and down and down. What that fundamentally means is we need a new way to measure people that actually reflects the way we're really working today in these networks. So that's where we think, ONA, or organizational network analysis, has a role to play.Neil: So what were some of the challenges that you guys had in the initial run towards where you guys find yourself now?Josh: It's actually probably a shorter list to talk about the things that we weren't challenged with. So one is my co-founder and I both come from R&D. We're both product builders. We're both very good at building products, and we never really had any doubt from the get-go that we would build the product right. The question was more like, were we going to build the right product? And that really is a go-to-market challenge. And that's something that almost every startup struggles with. Who is our customer? What do they want?Can we actually meet those needs, et cetera, et cetera? Is it a big enough market? And the list of problems just goes on and on and on. I think that having been through the startup spin cycle a couple of times now, we were able to do a much better job just getting that initial product out the door, recruiting great talent, getting some great investors behind us and ultimately then landing some great customers. But every step of the way is a challenge and almost none of our challenges are unique.

Finding a seat at the table for HR

Neil: What were some of the things going into this that you and your co-founder thought to be true that ended up not being the case?Josh: One thing that surprised me is that I think I thought that HR had more of a seat at the table than it really does. What I mean by that is when you think of an engineering organization and the buying of version control software, and they're gonna buy GitHub or GitLab or Bitbucket, they don't have to consult the rest of the company to make that decision. They just pick the right tool for the job. That's it. They buy it, and then they implement it. When you buy performance review software, it's a totally different story.You would think that HR would kind of be the final decision-maker on this and that would be it. What we were surprised by was how many other people in the organization, especially at the executive level, have opinions, and I really will call them opinions, because they're often not informed by anything. Opinions on how performance should be measured and how that complicates the buying process for HR. So it becomes this thing where HR doesn't have the ability anymore to say, You know what, we think this is the right approach to measure performance here.It becomes this group decision where it's like passing the federal budget or something. It's like you got to get all these people bought in to make this one decision. That was a little bit of a surprise and I think it maybe speaks to a larger sort of problem with HR today, which is that HR doesn't quite have the seat at the table that I thought they did. You know, when you look for example, at how people are promoted into the role of CEO, where do those people come from?Very common to see somebody go from being a Chief Operating Officer to being a chief executive officer. You'll see Chief Product officers move that up, you'll see chief marketing officers. How often do you ever see the Chief Human Resources Officer be promoted to CEO? We say that people are the most important asset, HR has never been promoted to CEO. I think there's a lot more to unpack there for another podcast, but that was one thing that surprised me.Neil: I never thought about that--very true. So with the importance that you guys clearly place on people in general and especially around what you're building here, how soon do you begin thinking about culture for the organization and how do you balance that with trying to find the best talent?Josh: The first thing that I would say about culture is that you have it whether you think about it or not. I mean, it's like a diet. You have a diet, whether you think about your diet or not. You have a diet. You have a culture, whether you think about it or not. It starts from day one. If I were a single founder, I would have a culture of one. With with my co-founder, we had a culture of two and with two people we had a shared set of values and practices.I think in our case in particular, we started thinking about culture much earlier than maybe we would have early in our careers. So for example, when I was at Carta, and I was the second employee at Carta, I got to see the culture from a very, very early stage and then see how it scaled up to five or 600 people when I left. We had a really strong, very idiosyncratic culture at Carta. In fact I like to say you can't spell culture without cult--and that was a little bit of what it felt like and that's frankly true for a lot of startups with strong cultures if they have this kind of culty feeling.What we didn't think about was, how do we shape that, and replenish it, and sustain it as we grow? We spent a lot of time thinking about that. I think we've done much more of that with Confirm, so by point of comparison, I think Carta was probably 400 employees before we wrote down our values. At Confirm we did it when we were I think 12 or 15. So we just think about it. We thought about it really, really early, because that's the fun of going to work.If you go to work in a shitty culture, that's just no fun. Given that no matter what happens to Confirm good or bad, my co-founder and I, we're going to be the last people there. We're going to be the ones who are most consistently on this journey from start to finish for this company and have to show up every day and if you accept that reality as part of your job, it really motivates you to want to build a culture that makes you want to get up in the morning every day and go to work. So yeah, we were deliberate about it.Neil: What was coming up for me while you were talking was how much of culture you're proactive about, and how the people you hire end up molding that. As you guys continue to hire to those values that you set out, have they changed over time, or are they still the standard that you guys try to rein everyone inside of?Josh: I think they're still the same. What I will say is that it hasn't been about protecting the culture, it's about growing the culture like we have. Think about it this way, between myself and Dave, my co-founder, there are two relationships. Mine to him, and his to mine. The moment we hired our first employee, and the number of relationships in our company, the number of connections in our company actually began to grow nonlinearly. You hire your first employee, now you have three people. It's not three relationships, it's actually six relationships because we each have a relationship with each other.That's where culture comes from. And so when you continue scaling that out, the opportunity to grow the culture gets really big. There are so many things that we do culturally today that Dave and I would never have thought of, and at this point, it just kind of feels like even at our size, we could be much much bigger. It already feels like we're not in control of it per se, but we can do things to guide it. We can do things to model it. If you talk to CEOs of companies that have 1000s of employees, oftentimes their number one fear is having to change culture. Because it's so hard to do.Neil: Shifting gears a bit, given where you guys are now, can you help me understand how you guys are currently thinking about making that march towards scale?Josh: It's a good question. I would say that I mentioned earlier that a lot of our customers are in this position of doing more with less; we want to grow revenue, but we don't want to grow headcount or we don't want to grow expenses. We're kind of in that same bucket. I wouldn't say it's an abnormally hard fundraising environment--this is probably normal compared to maybe 2018 and 2019. But compared to two years ago, it's way more difficult and we want to be mindful of that, and grow as much as we can by spending as little as possible. So that's kind of our scale plan. How can we keep reaching customers, keep making them happy, etc., while not burning more than we need to?Neil: What comes up for a lot of people when I ask this question is that in order to do more with less it is requiring a lot more focus on certain segments that they might be tackling. I'm curious, have you guys lined that out in terms of different areas you guys are tackling for your GTM strategies?Josh: We've defined the four different ideal customer profiles. Everything that we do around lead scoring, is mapped against those ICPs. Everything we measure around pipeline, conversion, and close rates, is also mapped against those ICPs. So what that means in practice is that when we see an ICP 1 come in, which is like our bread and butter, we get very excited and when we see an ICP four come in, we go well, you know, we can make it work. That we’re pretty clear on. If you ask almost anyone in the company, whether it's someone on the go-to-market side, or whether it's engineering or QA, they'll tell you what our ICP is.Neil: It's a team activity, not just just for sales, right? Every department is involved.Josh: Well, it affects everything, right? I mean, we made the decision about a year ago to focus on enterprise and we had been working with SMBs and decided, you know, we really want to move up-market, we want to do it quickly. We're going to focus on enterprise and it's totally different, it's a totally different sales motion. It's a different marketing motion. The product that you build just can't break. The way that SMBs are typically much more forgiving, enterprises are not. They have a lot at stake. Everything around data security and privacy becomes different. So yeah, it's a change that permeates everything that we do.Neil: If you're open to sharing, I'm curious to hear a little bit about what that lead scoring really looks like in terms of how many layers deep are you going into the ICP to build out that lead scoring.Josh: We do some lead scoring in HubSpot around engagement. So if somebody has downloaded our white paper or if they've opened our emails versus if those emails haven't been opened, things like that. We have a series of activities that we can attribute to a prospect and essentially give them points as they do those activities, which results in a higher score or higher lead score.Neil: How involved are you and your co-founders in any of the sales that are coming through the door for you guys?Josh: We're pretty involved. I would say we rely on the sales team to tell us when and how to be involved. I would say that as a founder, you get really good at telling the story of your company, the story of your business, and the problem that you're solving in the world and why what you're doing is unique. You could hire some incredible salespeople, but you probably will, as a co-founder, you probably will always be the best at telling that story.It's simply because you will have done it more than anybody else for a period of years and years and years and you'll see what works, you'll see what doesn't, what hits, what kind of falls flat. You will find that and so our sales team today gets me and my co-founder involved mostly kind of when they need us to come in and like to tell that story. And so they bring us in when they need us. Generally we do stay pretty involved.Neil: So what are some of the things as you guys continue on this journey that you're trying to be mindful of?Josh: For me and my co-founder, it's been a great learning experience for us because, as I mentioned, we're R&D people. We come from product and from engineering and we have a really good, I don't want to dehumanize our company by using this word, but we have a really good machine going. Our team is fantastic, they know what to do. Our attention has really been on the go-to-market side. It's really just been on how we create massive leverage in marketing with a relatively small budget compared to our more established competitors? How do we close these deals? That's really been our full-time focus. That go-to-market motion

Leave room for mistakes and ruffle some feathers

Neil: I'm assuming that you've already thrown some layers of thought and action upon that. I'm curious to hear what some of those steps you've taken have looked like?Josh: Well, one step is just honestly allowing ourselves to make mistakes and to have false starts and to try a campaign that doesn't work or try messaging that falls flat. We've done lots and lots of that. I think one of our biggest lessons learned in that respect, was that you know, it's a noisy world--especially in HR because they just get sold non-stop. If you want to break through the noise, you really have to not only offer something new, but you need to do it with a message that's bold and compelling. We've really stepped outside of our comfort zones to be able to do that.It was a little bit of a magic moment for us when the first person to follow one of our advertisements said, this offends me or it's wrong or whatever. That was a little bit of a magic moment because it was like, oh, okay, now we're actually putting a message out there that has some teeth, that is actually drawing this dividing line between people who like us and who probably will like us a lot more because we took a hard stance. I think if I were to give advice to my younger self, it would be, don't be afraid to ruffle some feathers and take a risk and man, maybe even piss some people off.Neil: How are you and Dave thinking about the next 36 months and where do you guys like to see yourselves?Josh: Obviously we want to grow as fast as possible, as responsibly as we can, which means doing more with less. What I would say to a lot of startups our size and stage, and what I remind myself is, Paul Graham talks about how the distribution of startup outcomes is nonlinear. It's a power law distribution, meaning that most startups are going to die. Some startups are going to achieve outcomes that are good, and maybe in some cases, life-changing, but not stellar.Then a very small percentage of startups are going to really become rocket ships and just explode. His point in talking about that distribution is it's not so much that you have to try to be this stellar 1% startup. It's more that like, you just kind of have to not die. If you don't die, if you don't end up with that long tail of startup failures, the odds are pretty high that you're going to end up with a good outcome.That's kind of how my co-founder and I think about it. Just don't die. Keep going. If you don't have the runway, the resources, whatever it is to fight another day, try another experiment, get another customer, try a new market, whatever. If we can do those things, the odds are in our favor that we'll get to a really good outcome. That's how we think about it.

Taking inspiration and learnings from his experience under one of the most prolific entrepreneurs of our time, Nicholas set sail on building his company through treacherous waters.

Neil: What was your original inspiration for building Workstream?Nicholas: Many conversations I have start this way. ‘Future of work’ companies often stem from founders' problems and pain points. I started my career early at Tesla, on the analytics team, supporting the Tesla Roadster and Model S engineering, manufacturing, and sales teams. I experienced workflow breakdowns in collaboration and teamwork. After business school, in New York, I was running operations at a growth-stage company called BetterCloud. I faced similar challenges from a different perspective. The data and analytics technologies at Better Cloud were more advanced than what I had access to at Tesla, yet the workflow challenges persisted.Neil: Interesting.Nicholas: Despite advanced core data technologies, analytics teams' workflows in supporting and empowering data usage felt broken. This led to discoverability and efficiency problems, creating repetitive work. That's why I started the company – to use technology to solve these issues.Neil: Got it. So, after developing this vision, how did you go about acquiring your first customers?Nicholas: The challenge was going from zero to one, and we're still seeking product-market fit, to be honest. In summer 2019, before the pandemic, I quit my job and took some time off. In January 2020, my co-founder Chris and I began conducting numerous interviews with potential users, initially focusing on any analysts and understanding their challenges. We developed an initial product and launched an alpha version that summer. The first customers were those I had interviewed earlier, around late summer, fall of 2023. Our product and customer understanding have evolved significantly since then, leading to multiple pivots.Neil: That's impressive. You mentioned still searching for product-market fit. How do you define it, and how will you know when you've reached it?Nicholas: Great question. In B2B software, which I'm familiar with, product-market fit means you're solving a problem for enough customers who are willing to buy your solution. It's an inflection point where your product is validated by a critical mass of customers, and the market starts pulling for your product. It's about the product and market fitting together. Achieving product-market fit can vary – it might happen quickly or take years. Once achieved, the company's focus shifts from proving the product's value to scaling it efficiently.Neil: Makes sense.Nicholas: Post product-market fit, the challenge becomes product-market-go-to-market fit. It's about scaling the business in a cost-efficient way. For instance, to prove product-market fit in the current market, you might need around 100 customers paying, say, $10,000 each. The strategies for acquiring these customers can differ significantly from those used to scale from one to ten to a hundred customers.Securing product-market fit is crucial before scaling. It confirms that you're solving a specific problem for a defined customer group. After achieving this, the challenge is to efficiently distribute the product. As a venture-backed business, your focus shifts based on the existential problem at hand, which changes from validating your product in the market to finding an efficient distribution method.Neil: Absolutely.

A step from either glory or failure

Nicholas: Every good startup is always a step away from either glory or failure. You're constantly facing existential crises. If you're not, you're probably not pushing hard enough. Every successful startup balances on this edge throughout its journey.Nicholas: I had a conversation with a founder friend about optimizing for runway. He pointed out that if you're only focused on optimizing for runway, you're essentially standing still. You need to make bets that make you uncomfortable, right? Runway is important, of course, but if that's your only focus, you might miss hiring talented people who could help solve existential problems. For example, when I was at Tesla, Elon Musk, a brilliant yet unconventional entrepreneur, would push us to take huge risks, always charting a course close to the sun. He wouldn't have achieved what he did without being willing to risk failure repeatedly.He wouldn’t have created all that he has without being willing to be on the brink of failure over and over.Neil: It’s the equivalent to, “risk it to get the biscuit.”Nicholas: Exactly. There are many examples from my time there Take security at the Fremont factory. Traditional advice was to have a large security team, but Elon opted for minimal security to reduce overhead, despite potential risks. This risk-taking attitude is essential for founders.Neil: Absolutely.Nicholas: On the topic of gaining our first hundred customers, the challenge is immense, especially now when many are hesitant to buy new enterprise software. Our focus is on honing our value proposition and delivering 'painkiller' features. It's about refining our ideal customer profile and solving specific pain points effectively.Neil: So, what are the biggest barriers to generating the demand you need for those first 100 customers?Nicholas: The main challenge is finding people who recognize the urgent need for our solution. If the majority don't see the problem as a priority, achieving product-market fit is difficult. We aim for those who immediately understand and ask about the next steps.

Be your own bullhorn

Neil: How are you approaching the market to gain traction?Nicholas: Our approach involves being the 'bullhorn' for our company, creating awareness through various channels like LinkedIn and our own events. It's about engaging potential customers at different funnel stages, from awareness to interest, and eventually to sales. We're increasingly using LinkedIn for its targeting capabilities in all parts of our funnel.Neil: That makes sense. Developing a media-like approach seems to be the new wave in go-to-market strategies.Managing deliverability, for things like e-mail just to get into the inbox is crucial. Then you're competing with everyone else who made it in there. Developing a media company around your organization can really assist in that demand generation, allowing salespeople to focus more on demand capture. Chris Walker at Refine Labs talks about this, developing content strategies around demand gen and capture for new go-to-market motions. Does that sound familiar?Nicholas: I haven't seen his stuff, as I'm heads down at work and don't have much time for extensive reading. But I agree, controlling your narrative and message is crucial. As a founder, you need to be the spokesperson; if you don't do it, nobody else will have the authority or vision. It's essential, especially for certain types of businesses.In enterprise, particularly super infrastructure, your traction might depend more on something like GitHub stars. But for more business-oriented products, finding ways to create awareness yourself is key.Neil: So, what challenges have you and your team faced over the past year that caught you off guard?Nicholas: We've discussed quite a few of them. The headwinds in the current market have been a major challenge. I didn't anticipate how tough it would be, not just in the capital markets but also in the buying market. People are less willing to buy, and when they do, they spend less. There's a compound effect with more diligence from investors and a decrease in product valuation. We need to do much more to achieve what the market deems as product-market fit, especially if you're venture-backed.Neil: Absolutely–and where do you see yourselves going over the next 36 months?Nicholas: Our focus is to accelerate through product-market fit. We want to foster deep customer love for our product, as that indicates a significant impact. It's all about the impact our product has – the time it saves, the security it brings. That’s the ultimate payoff.

From sports to spreadsheets, Nic shares insights on evolving enterprise automation, AI's burgeoning role in business, and strategic approaches to reshaping the SaaS landscape.

Neil: So, What inspired you initially to build Frame?Nic: Sure. Back in 2014-16, I had a startup in the vertical sports business, a SaaS company. I noticed that as soon as a company grows to around ten employees, managing the numerous applications and SaaS becomes overwhelming. You almost need a Google Sheet just to keep track of all the different SaaS products and who has access to what. Since 2016-17, the idea of SaaS orchestration has been on my mind as a problem I wanted to solve. I observed various attempts at addressing this issue, mostly through different versions of Chrome and various access methods to applications. However, these didn't tackle the core problem: reducing the number of apps and rationalizing the proliferation of SaaS.Additionally, I worked in the enterprise sector, where I saw companies like Swift relying heavily on the Microsoft ecosystem. This often felt like a disjointed patchwork of software – a combination of OneNote, Teams, OneDrive, and SharePoint. It seemed more like a suite of software playing catch-up with current technology, resulting in a fragmented user experience. Moreover, I noticed business departments coordinating tasks with IT using Jira, a complex tool for non-IT personnel. This led me to think: Can we create a simple task management system to facilitate coordination across various departments?That's where the idea of Frame came from. It's a minimalistic business operating system comprising a suite of apps that are simple, interoperable, and built for teams to co-create documents in real-time.Neil: Wow, that's fascinating. How did you bring this vision to life, and more importantly, how did you find your first customers?Nic: One of the hardest steps as a founder is turning an abstract idea into reality. My advice is to first understand the market space and how others are approaching the same problem. Then, based on your personal experiences, start sketching out a solution. Working with a designer, or if you're a designer yourself, quickly move to creating mockups. This helps in getting faster and more accurate feedback from potential investors and clients. Having something tangible like a mockup makes the idea less abstract and more compelling.As for finding the first customers, it's about launching as soon as possible and seeing what resonates in the market. Then, based on that feedback, you can start to define your Ideal Customer Profile (ICP) and improve the product accordingly. It's essential to have analytics early on to understand user interactions with your product.Neil: Regarding the ICP, can you elaborate on the level of granularity you're using to target your ICP at this stage?Nic: Our approach is unique. We focus on the type of business and the role of the user. For example, are they a freelancer, VC, software development agency, pre-seed, or series B business? We also consider growth dimensions like the user's role – are they a product manager, an operations manager? We ask users why they joined Frame, such as to reduce the number of apps, speed up work, or enjoy a better work environment. This data helps us analyze our KPIs based on different criteria.We're currently targeting early-stage businesses that are more likely to adopt new solutions like Frame. We consider where these businesses gather both online and offline, like Reddit, podcasts, or various Slack communities.Neil: That's really insightful. What were some of the challenges you faced at launch, and what did you learn from them?Nic: Our approach was somewhat antithetical to typical startup strategies. We were essentially launching multiple startups simultaneously – a task app, a whiteboard app, etc. The common advice is to be somewhat embarrassed by your first product version, suggesting a timely launch. However, in our field, users have high expectations for software quality and user experience. So, we focused on launching quickly but with a limited scope, emphasizing quality, speed, and accessibility.Our strategy was to build a magical experience within a fixed scope, especially considering our limited funding compared to some competitors. The challenge was to be extremely efficient with capital to match user expectations.

The evolution towards fully automated enterprises

Neil: Considering the competitive landscape and the evolution of technology, what do you envision for your category, say, in 2034?"Nic: In the long term, technology evolves rapidly. Currently, there's an excess of B2B SaaS solutions offering narrow, specific solutions. We're entering a phase where companies are looking to rationalize their SaaS spending, leading to a trend in SaaS consolidation. In the future, having a unified metadata system like Frame's could be crucial for AI-powered insights and for AI agents to effectively collaborate with human employees.Regarding interfaces, while voice and augmented reality are emerging, I believe 2D will remain a crucial format for human comprehension. Our brains are wired to process 2D representations more efficiently than 3D. So, even in virtual or augmented environments, projecting 2D interfaces will likely remain important.As for data, centralizing and rationalizing it makes it more powerful and accessible. Think of AI as an internal McKinsey team in a company. The quality of insights depends heavily on the organization and clarity of the data. Finally, there's the question of how AI and human intelligence will collaborate in the future of work.We're definitely heading towards a fully automated enterprise era. There was a time of manual enterprises, followed by semi-automated ones. Now, we're moving towards full automation, where data APIs connect and manipulate data through algorithmic agents. This shift brings up philosophical and social questions about the future of work and how we'll compete with these agents. By 2030 or 2050, these issues will be at the forefront of discussion.Neil: That's a really intriguing insight. I've been thinking about how fully automated enterprises could change the entrepreneurial landscape. More people might have the opportunity to build things, but it also means increased competition. I wonder whether the market will become more fragmented, focusing on vertical-specific solutions, rather than serving a wide range of industries.Nic: Exactly. It's about the competition between humans and, in a sense, capitalism. The accessibility of technology and the reduced capital needed to start a business has already shifted the landscape. With technologies like AWS and GitHub, people can start with minimal capital. As a result, simpler ventures have become saturated, and now, more complex, costly solutions are needed to gain market share. With AI, we might see a similar pattern, leading to more specialized AI applications like AI lawyers or AI vets. Eventually, building software could become a commodity, shifting the focus to data access rather than the interface itself. Vertical SaaS companies, like toast.com with its wealth of restaurant data, may have an advantage in this scenario. However, I believe both horizontal and vertical businesses can leverage AI effectively.Neil: Shifting focus a bit, what were some beliefs or assumptions you had going into this venture that you later found to be untrue?Nic: "When I first pitched to VCs, I believed that if we created a beautiful software experience, data import wouldn't be as critical – people would naturally start using the new software. However, I've learned this isn't entirely true, especially in B2B contexts. Organizations are entrenched in their data ecosystems, and without robust data import tools, it's challenging to persuade them to switch to a new platform like Frame. Another area where my perspective has shifted is regarding fundraising. There's a misconception that knowing the right people might lead to easy funding. While network and luck play a role, successful fundraising is more about being systematic and determined. It involves methodically connecting with investors, enduring rejections, and persisting until you secure the necessary funding."Neil: "That's a great perspective. As you navigate the next stages of growth, what are some key considerations for you and your team?"Nic: Our primary focus is on product quality. We initially aimed to build a lot, but now it's about refining and perfecting each aspect step by step – starting with quality, then speed, AI functionalities, and data import. In terms of growth, our focus is on retention. Many people equate growth with user sign-ups, but in SaaS, retention is the real game. From day one, we're concentrated on improving our week twelve retention metrics, as this is more indicative of long-term success than just sign-ups or even ARR.Neil: Regarding retention, how do you manage potential churn that might result from changes in the product?Nic: "Given our broad market scope, like Notion's, the end market is virtually infinite. After launching a product, you observe what resonates, structure your findings, and then focus on a specific ICP. Optimizing for this ICP is crucial, and it's important to have an intuitive understanding of your target market from the start. Unless there's a major pivot, continuous customer engagement and impact measurement should see retention rates improve. Our vision at Frame hasn't fundamentally changed since our initial mockups, which I believe indicates we were on the right track from the beginning

The user-generated flywheel

Neil: Okay, I like that. Looking ahead, considering where you are now, what are some of your goals for the next 12, 24, and 36 months? Where do you see yourselves?Nic: In the next twelve months, our goal is to secure a substantial seed round and achieve high retention at week twelve. Looking further, say 36 months from now, we aim to reach a million and a half in ARR and secure a series A funding round.Neil: In relation to your go-to-market strategies, could you share some of the approaches you're taking to reach these targets?Nic: Certainly. Our focus is heavily on retention, which means improving onboarding, the notification system, and overall product experience. We believe creating a great product can naturally lead users to become paid subscribers. As for acquisition channels, we're exploring all avenues, including inbound and outbound strategies, paid ads, which can be scaled even with a small budget, influencer marketing, and user-generated content.Neil: That sounds intriguing, especially the user-generated content part. Could you elaborate on what that involves?Nic: User-generated content varies depending on your brand identity. For more serious businesses like Google for Work, you wouldn't expect TikTok-style content. It can range from influencer reviews, which often involve paid advertising, to short video formats popular on platforms like TikTok and Instagram. We're exploring these options, considering how they align with our brand's seriousness and identity.Neil: Great. Now for some lighter questions. What books, articles, or videos have most influenced your entrepreneurial journey?Nic: Paul Graham's essays have been incredibly influential. They're concise and rich in learning, drawing from a wide spectrum of startup experiences. 'Zero to One' by Peter Thiel is also a must-read, offering valuable insights into developing unique business ideas. Guy Kawasaki's 'The Art of the Start' is excellent for those new to entrepreneurship. And Y Combinator's 'How to Start a Startup' series, particularly the 2014 Stanford class, remains very relevant.Neil: Those are great resources. Lastly, in one or two sentences, what would you want people to understand about Frame and the problem it solves?Nic: Frame is about bringing back joy and simplicity to your day-to-day work. We're moving away from a fragmented, overwhelming experience to a beautiful, unified software suite. It's like having a macOS look and feel across all your productivity apps, built for teams. AI plays a key role in Frame, acting as an accelerator to bring AI-powered enhancements to this unified experience.