85% Said Yes to One Feature. 5% Said Yes to Everything Else.

Sean Weisbrot: Why do middle managers need to be disrupted by ai?

Dima Syrotkin: Because while it is the best solution we have right now, it's very far from perfect. I mean, if you look at any company that is on the stock market right now, it's like, unless it's a tech company or Elon Musk gets involved, they're all struggling to innovate and they're all stagnating. And I would claim that a lot of that is about coordination problems. Um, if you look at like, how big companies are right now, many, many people talk about like AI shrinking companies, where, you know, a lot of things will be automated, but I wonder if some companies would actually become bigger because coordination will not be as painful and people would not, you know, when you ask them like, where do you work? They would not be answering like, ah, big company. And there's this like, heavy sigh of like, yeah, you know, it's terrible. Um, so it, you know, I, I think, I think it doesn't have to be like that. Um, and I think, you know, what's interesting to me is once we fix that coordination, could companies actually be even more impactful and grow in, even bigger in size

Sean Weisbrot: than they are right now hey, business leaders and marketers. What if your brand could be featured right here? This ad spot could be yours. This channel is watched by a dedicated audience of ambitious founders, executives, and professionals who are actively looking for tools and services to help their business grow. If you wanna put your brand in front of this highly dedicated audience, that's difficult to reach. I'm currently looking for a few strategic partners for the channel. To learn more about sponsorship opportunities, click the link in the description. Let's go together. Why is the solution using AI though? Why not just fire people and, and let just there be less people? Why? Why do you have to apply AI to this solution?

Dima Syrotkin: I actually studied, um, i, I worked on a PhD on that exact topic, uh, what's called self-managing Organizations. One of the famous articles, um, that a one like management guru published on this topic was like, why, why don't we just fire all the managers? And I guess the answer is like, it's just not so easy. Um, like, uh, Zappos, maybe some people have heard of it. Mm. There was a cool book, I forget the name, by, um, the CEO, Tony Shea. Um, I think it's. Something that has happened is in the title, but it is basically a lot about like their customer success and how they really went, um, above and beyond there. So that company that sold to Amazon, they tried to. Become more self-managing and apply what's called holacracy. And long story short, it, it was a shit show and they, they ended up then like reverting back. Um, I, I think at the end of the day, like, you know, we looked at like many examples and it's just, it's just super hard. There's some like very rare occurrences when it does work. Like, there's a company in the Netherlands that has 18,000 employees, and they're all divided into teams of 12, and they actually don't have middle management. Um, but it's, it's like a nursing kind of service organization where every team works on their own, like small pocket of a city. So it's kind of easy to subdivide them and they don't have to collaborate that much. Um, so it might be a bit more of a, you know, um, of a, um, sort of exception. Not the rule. And if you look at ai, I actually think this is one of the things it's best at, it's actually best at, you know, becoming more of a glue. And one, um, one book I recommend everyone to read is called Reshuffle. I think it's like number one AI book right now, uh, on like Amazon. Um, that basically talks about the fact that like rather thinking of AI as like the super intelligence right now, we might be better off thinking of it as this like coordination mechanism that allows us to. You know, um, coordinate between people and, and, uh, serve as, as this like completely new mechanism of, of coordination.

Sean Weisbrot: Yeah. How do you get companies to trust you that this is a solution they should try out, especially when they're probably going to have to fire their humans in order to do it. And you might have to get buy-ins from humans to let it even start.

Dima Syrotkin: So I think it's, it's actually like, you know, it, it's gonna start as augmentation for sure. And even for myself, like it's not co completely clear, you know, will we actually end up firing a ton of people or will it just that they, you know, would end up going and doing more productive things. Was there time, um, because it's like, you know, when, when demand for electricity kind of went, went down, sorry, when, when supply supply of electricity kind of went up and, and the prices went down. The demand actually increased rather than like decreased and like the, the whole profits of, of, of theistic companies. So I think it might be something similar, uh, here as well. Um, I think what also helps us is, uh, just focusing on concrete problems. So we don't really come to companies and say, Hey, you know, we're gonna. Sort of augment your middle management. Here's a new way of leading your company. Now that's a bit too like, um, sci-fi ivory tower pitch. Um, what means that say is, look, you know, here is a digital transformation or change management program, or like AI adoption problem that you, you have, and the official statistic is that most of them fail. You're gonna spend billions of dollars and, and probably, you know, it's gonna take a lot longer. Um, why don't we try something new here because you know that it's not working. You have a contribu problem. What if we can help, help and, and, and solve, solve it for you there. Um, the last thing that I would say is we collaborate a lot with consulting firms and that really helps to build the trust because they, um. You know, they, they've built really long-term relationships and so when, and the consulting firms are somehow more prone to seeing the future, and maybe that's their job, I don't know, but they've been a lot more receptive. And, and then when they come to the client and when they come, not just like, you know, kind of like, oh, it's a weird startup, but they sort of pitch it during the golf session, you know, in their native language. Uh, and they've had a relationship for decades like that. That also, uh, changes the stance of the clients where they're like, okay, like maybe I'll try it.

Sean Weisbrot: So you're heavily reliant on Referers for sales?

Dima Syrotkin: Yes. I think in this industry, I mean, that's basically how it runs, right? Like in, in consulting it's like, you know, you, you call, you call your, you know, partner in McKinsey, Joe that you've known for like 20 years and you've been through, you know, hell and fire with, and maybe you've even worked in McKinsey yourself. And so you've, you know, um, have a lot of experience with them. And, and so I think like. If we want to disrupt the industry, that's sort of the best ways from the inside. It works so far really well for us because for them as well, there's a lot of incentives because they want to, um, be able to differentiate themselves. And AI is one, one way to differentiate themselves. Right now. They also want to, um, use our data to understand their clients better, to be able to serve them better, right? Because they can come back and say, look, you know, based on TRON data, we see that besides this thing you hired us for, there's also a problem X, Y, and z. Guess what? We have thousands of consultants that can help with that too. And then lastly, many people think that, you know, consultants would be afraid of us, like disrupting their industry. Um, but they actually, you know, if we can increase their margins, they're like all for it. Like, they're not in the business of selling people. They're in the business of making money effectively and like increasing their profits. And so if, um, you know, if they can start, sell, um, recurring sauce, like, you know, on only, only the better.

Sean Weisbrot: It's a really great model. Because I, I did this before. I had a consulting business a long time ago, and I was a hundred percent referral. And the commissions were so large that the people referring were very excited to send me as many clients as they could, as fast as they could, because the, the time to close was sometimes, you know, an hour, like just a single call. Yeah. All right. Let's do it. Yeah. Yeah. So because they knew I could deliver and because they knew the commissions were good and they knew that I sent them the commissions, as soon as I got the money, which was sometimes the same day as we signed the agreement, they were highly incentivized to keep going. So if you have something that makes them happy and as you said, possibly feeds them data so that they can do better, then it sounds like a no brainer. Is there a point at which this stops being scalable for not, not for the software, but for the referral model?

Dima Syrotkin: I think it's not so much about like the scalability, but more like maybe conflicts of interest, like some features we are planning for is that when we give the data to the csuite. And we say, Hey, you have certain problems. We wanna be able to have like a one button that says like, deploy a task force. And you know, they would press the button and we'll find the best people both within the organization and also outside the organization that could solve a particular problem. And then they could like confirm and then two weeks later they get a report on their table saying, Hey. Here's, you know, what this problem is about and here's what you should do. And, you know, here's a result of our investigation. Um, and that's something that they called their friend Joe and McKinsey about right now. And so I think at some point, you know, once we start kind of piling up those features that automate some of the consulting workflow, it becomes a question of like, okay, uh, do we become competitive right now? Neither us nor them really worry about that too much because we're super small. We're not a threat to them. At some point it could become. The case. And so we, we would need to think like, okay, at that point do we sort of then need to decouple and sort of sell directly, uh, only or you know, how to manage that or, or maybe they invest in us. So the incentives are aligned because they would then be like, okay, well if Panda John succeeds, then well, we own a piece of the company. So like, whatever, it doesn't really harm us. Um, so, so there might be ways to, to sort of bypass that. But that's, um, you know, and, and potentially we also wanna go into strategy, uh, uh, creation, which is like the biggest part of their business because if we have internal data, we could then, you know, also start, um, creating the strategy for them, not just, uh, working on the execution part. And that could be really disruptive.

Sean Weisbrot: So wouldn't it make sense for you to focus more on building an alternative channel for sales, like direct sales now to prevent yourself from having this potential problem later? Because you could, for example, get to 50 million a RR, you know, if, if you keep building these referrals correctly, you might hit that in 18 months. Yep. And. By 12 months, they might already start being skeptical. And let's say you finish this round in March or April, and let's say September of 2027, you're what? Deep into spending that money deep into growing your revenue. But the companies are starting to cut you off. And while you may retain their clients, they may convince their clients to walk away from you. So if you don't have your own direct sales program that you're starting to incubate, like let's say summer of next year, um, summer 26, do you have a year to kind of figure out how to get your direct sales channel going using profit from these companies?

Dima Syrotkin: Uh, we already have it ongoing, so I think currently maybe 25% of sales comes direct. Um, and that is mostly through conferences. We sponsor, we speak. I think the limitation factor is the brand. But I do think that, you know, once we become a threat, that also automatically means that our brand has become quite big already. Because, you know, we're probably making quite a bit of revenue. We already have like, uh, a few big clients. Um, so I think like there it should be, you know, um, it should be kind of like. Possible to, to, to have that as a smooth transition. Because, because of the time when we do need to make it, uh, we also kind of in, in a way stronger, stronger position right now. Uh, we are definitely like investing there. Um, but it's just like, because we're so small, um, it's just so much easier for us to go through, through channel partners, um, than going direct. Um, yeah,

Sean Weisbrot: it's, I'd say it's 25% is a pretty good starting point. Like it's, I'd say it's pretty good already where you are. Um, if you said you had zero and you hadn't started, then I would be a little bit more worried for you. But that's good because if you have 25%, then it, it tells you that you already have something that's working. But you said that it, it mostly comes from conferences. So do you have a method to do like a, like paid ads or do you think that's not gonna be able to reach the right people from an organization?

Dima Syrotkin: We're definitely experimenting. Um, but it's, what we found out is that it's hard to go from even like sea level minus three because like they might buy, but then even if they're happy, then there's like so much kind of resistance they have to face when they go up and they have to, like, they cannot really upsell to sea level directly. They have to, you know, upsell to the level above them, and then that level has to upsell the level above them. So just like. It's hard, um, but we're definitely experimenting with it anyway. Um, I do think that the product itself also has to be very addictive for the users. And so then that backs the question of like, could there be some sort of product led growth approach where we provide, um, something for the, um, for the white collar employees. Um, where they, they also use that and, and enjoy the tool, and then they, they kind of come to their, to their managers to buy it. Um, but, um, still pretty, pretty early. So conferences work the best for us because there, we, we can meet actual sea level. Uh, so for example, like, uh, world Economic Forum in Davos and Switzerland, like we're going there in a few weeks. Um, that's been like a really good conference for us. Like brings a lot of sea level. Um. Um, but we're definitely experimenting with all, all kinds of different, different channels. I think now we are, we're fundraising, um, like our seed round. I think after that, like one of my early hires is gonna be the CMO and then we would definitely wanna like, um, look at more channels. Um, right now we're just not really kind of like, don't have enough resources to, to look into everything.

Sean Weisbrot: That makes sense. A really close friend of mine's had a SaaS business for 14 years, and they're on track to do 10 million a RR next year. They're fully, fully self-funded, so they've, they've never taken investor money and, and they got the majority of their clients from going to conferences, a number of conferences every year for many years, to the point that they now steal from their, um, their competitors because their product is better.

Dima Syrotkin: Yeah, it works in enterprise. Yeah, it's, it's magical. Like the, the, the power of face-to-face is crazy. Maybe because, you know, the decision makers are a bit like older, so like, it, it's a bit harder to get, uh, get them online and, and probably because everyone bombards them with everything already. So.

Sean Weisbrot: Well they do, uh, they serve ho like hotels, but they're serving more boutique hotels. So it might be a single owner of like a 10 unit or a 15, 20 unit hotel. Um, some of them go up to a hundred units, but um, a lot of them, they just went to the hotels, knocked on the door and said, hi, I am here. What's up? Um, a lot of it was, that

Dima Syrotkin: makes

Sean Weisbrot: sense.

Dima Syrotkin: I mean, hotels seem to be like this, very physical businesses, right? Like you're literally hosting people in this, you know, building like. It makes sense that you would also like, get them face to face more than you would get them online.

Sean Weisbrot: Yeah. Uh, but now they can have, you know, they have other people that go to the conferences and do those sales and they have, you know, advertising and other sources that make it easier, um, on the founders to not have to do a lot of that hard work. So what has been one of the most challenging things for you about doing this business?

Dima Syrotkin: In the beginning, it took us a while to really build that, uh, you know, that demand generation, right? Like going to the first conferences, getting the first interest, like when you have zero, right? Uh, 'cause then you, you go to this like giant corporations and then, uh, getting our first reseller initially, uh, like a small. 20 million Revenue consulting firm. Um, and then they brought us some big clients like Panasonic Mitsubishi. And then that brought us enough credibility to then get our first like mid-tier reseller. And then that brought us the first like 8 million a year, uh, dollar deal in the pipeline now with Marriott. So there, you know, that, that is, um, that, that all took time and, and we had to kind of build, build up to it. Uh, so if you show up to, to enterprise, you know, it's, um. Uh, it, it's hard unless you really have like strong connections. And then the other thing I would say is on the product, which is also similar, I think enterprise is also challenging in terms of the product because the expectations are very high. You know, everything has to work perfectly. Um, especially like we work internationally and for example, Japanese have like crazy high standards. Uh, like they hate any kinds of bugs or anything like that. It's like, you know, it's just unacceptable, you know, here. Here, here is the, the katana go performs a pku. If, if there was a bug, this is totally, totally unacceptable. Um, and then, um, joking, of course, but still, um, and then also like security requirements, right? Like, you know, you absolutely cannot sell unless you have like really strong security and that ends up being like. One third of your like tech team basic working on like privacy and security and that That's like a huge expense that like if we were just a consumer startup or we were sending to SMB, we could have like put those resources somewhere else.

Sean Weisbrot: So when you're talking to these prospects, are they telling you, we need you to have this thing or this thing? 'cause I, I built a, a SaaS by myself. If I coded a SaaS, that was a financial service. And the only API that we really needed was from QuickBooks. QuickBooks required us to have this. There's this four page, oh, this's just me there. There was a four page document with all of their security requirements for them to allow us to allow me to use a production API. And so I had to say to Claude, Hey, hey, Claude, review this document. Tell me what we have, what we don't have, and then implemented for me. But then I had to go back several times and make sure that it actually did the things it said that it was going to do because it didn't. And I caught it lying to me several times. Um. But, but once I got it all done, then I was able to do it. So do you, do you have any things like that where the client's like, look, I wanna work with you, but I need to see that you've implemented this specific security thing first.

Dima Syrotkin: I think like likely the good thing is that in security there are standards. So once we got certifications, um, things got a lot easier. So like ISO SOC two, those are more or less like the golden standards. And then like, you know, GDPR if you are like working in Europe. Um, and, but, but like to your point, you said four pages. Um, we've, we've gotten like, hit with like, uh, 300 questions, like, I don't know how many. Uh, you know, pages that was, and it's not uncommon like in, in corporate setting. It's, it's, but then the good thing is that if you have a certification like that probably covers like 290 of those 300 questions. Uh, but yeah, our like, trust page is like insanely long with all the, all the documents. Like, I don't, I don't know. It must be more than a thousand pages, uh, which is not uncommon. Um, but, uh, yeah.

Sean Weisbrot: Um. Can't you take that document and put it into like Claude, for example, and say, look at my code base, look at this document and help me answer all the questions that you can answer. And if you can't answer it, tell me. So I know you're not just like bullshitting.

Dima Syrotkin: I mean, we do that to some extent. I mean, luckily, like sometimes it's like easier for our CTO just to answer directly. Like he doesn't even need to ask Claude 'cause he just like knows the answer. Right. Um, and, and then the good thing is that with like certifications, you basically like, you know, 95% is already answered. So that makes it like very, very easy. And then, uh, nowadays some clients don't even ask anything. They're just like, okay, you have like ISO and SOC two. Like, that's fine, you're good. You know? So, so, you know, but, but the getting, there was a bunch of work and you have to, of course, like keep it updated and, you know, so I, as I said, like I think one third of our team is on like security and privacy. Some of it is security, but then also the other part is privacy, right? With ai. So they, they wanna know that like, um, we, for example, have a module that like. Uh, removes personal identifiable information before passing it to the LLMs. Um, so that's like one, one example of like something we had to do that was like custom. Um,

Sean Weisbrot: was it complicated to get those certifications and like, did you need to have clients before the certifiers would recognize you? Or like, did you have a chicken and egg problem at all?

Dima Syrotkin: Oh, no, no. Um, then you know, it, it's all about the product, so you don't have to have any clients before getting certifications. To be frank, we though we did already have clients and, and we like, saw their demands and we were answering this 300 question documents and we're like, okay, we really need to get certified. 'cause like, this is, this is not, not working. This is clearly not the way you should do it. Um, but then we were very lucky. I mean, I can plug them here. Uh, we were really, um, happy with the provider called One Lead. So it's kind of like, I would say it's like a Vanta competitor. Like I think everyone knows Vanta by now. Um, but they're like a ton more expensive. I haven't worked with Vanta, so they must be really good. Um, but like one lead basically was like a ton cheaper. Um, it's, you know, kind of you could say a startup that is trying to like, make it. A lot more affordable, a lot better. And they were great, kind of like every step of the way. So honestly it was pretty smooth sailing. And I mean we also got like quite good people on security 'cause we knew that it was important so we reinvested there as well. But all in all, I would say the challenge was just more recognizing that that's a need and then actually then like picking the right partner and then like consciously investing in that, uh, on the team side as well. Uh, but it's all doable.

Sean Weisbrot: So this company consulted you through the application process to get certified?

Dima Syrotkin: Exactly. They're kind of consulting and they also have a platform. So for example, our trust center, like they host it, um, so they have an actual product as well that helps with every step of the way, uh, with some AI features as well there, et cetera. It's like very, very good product. Honestly, I think they raised their series A recently, and I, I understand why.

Sean Weisbrot: So in those earlier days. You were getting these giant documents, but did any of them just say, Hey, or are you certified?

Dima Syrotkin: Yeah. All of them asked, uh, are you certified? And we were like, uh, no, but we are definitely compliant. You know, we can answer all your questions. And then in the back of our minds we're like, shit, we, we definitely need to get certified SAP. Um, so yeah.

Sean Weisbrot: So how long did it actually take to get certified?

Dima Syrotkin: I would say, um, half a year. And I mean that depends on your, on the size of your product. So we already had quite a bit built, like if, if your product is like, if you're just starting up, right? Like you could probably do it in like one or two months. I mean, it's pretty, like pretty fast. But like in theory you can't, if you don't have much, right? So there's not much to certify and it's very easy to like, you know, plug all the holes, et cetera. Um, but because our product does have like quite a bit, I think there's like. Already more than 15, like diverse services that like work together. Um, it's, um, yeah, it's a heavier lift.

Sean Weisbrot: Hmm. And what was the approximate cost of being able to get certified? Just for the people out there that are thinking about this?

Dima Syrotkin: Uh, it was 20 K, which I like, expected a lot more. Um, but I think maybe because of, you know, this company again is like up and coming. And I think their general logic is like they're trying to make these things affordable. Um, but yeah, it was basically 20 K for ISO SOC two and GDPR, so it was kind of like a whole package.

Sean Weisbrot: What are you looking forward to for 2026?

Dima Syrotkin: So we are, you know, super close to hitting a million a RR, which is an exciting, um, exciting milestone. We also have a couple of super good people joining the team that have been chasing for a while to join the team. Um, one of them, uh, he sold his, uh, design agency. I think they had like. 900 designers. He sold it to Capgemini, um, I believe it was a $300 million exit. Um, so quite, quite a large exit. Um, and, you know, he was Angel investing and supporting startups, and I convinced him that what we have is, you know. Worth, worth, um, investing in and supporting, and he doesn't wanna miss it. And, uh, he will bring also his top designer from that firm called Adian that they sold to Capgemini. So they're gonna join us in January. Um, that's super exciting. Um, I'm, you know, always so happy when like, I kind of manage to secure some really, like top talent. Um, I also, um, the guy who basically built Google Cloud, like he was the COO of Google Cloud. Um, I also, he will join us as an advisor, so he will be kind of my coach. Uh, he invested himself and, uh, he's gonna join. So that's, um, that's pretty exciting. He's not, you know. Not gonna be full-time or anything, but like, super experienced guy. And, uh, we'll see. Maybe I'll potentially recruit him to, uh, to join full-time. Um, but yeah, uh, we had really good chemistry and like he's done some really good stuff. Uh, he was like part of Google, but then also part of like Palo Alto Networks. Um, yeah. Um, very experienced. And then besides that. A lot of stuff on the product side that I'm excited about. Uh, like we want to build, um, a ton of features, um, that we, you know, e every day I interview people and they're like, yeah, this thing is missing. And I'm like, yes, I understand. And we are already building this and, you know, I'm hoping to have this soon, uh, one example that I didn't mention yet. For example, uh, as you know, um, as an illustration. We would like to connect the right people in the organization. When you have a giant organization, many people feel like they're like, I don't know what other people are even working on. So one, one feature we're thinking about is, um, to say like, Hey, you know, there's other 40 people working on the same problem as you in your company. Do you wanna connect with them? And then we would open like an anonymized group chat and they could kind of diagnose the problem together or find out what other people have done before that actually worked for them, et cetera. And when I was doing like a user research, uh, some features, like I was asking people, well, if you get an insight from our AI agent, do you wanna share it on LinkedIn? And like 5% said yes. So pretty small. For this feature, it was like 85%. I was like, wow, okay, we might have something here. Uh, and many people like, yeah, I would love to have that because like often I hear afterwards like, oh, someone solved this problem already like three years ago, but I had no idea. Um, basically, yeah.

Sean Weisbrot: What's the most important thing you've learned in doing this business?

Dima Syrotkin: Um, honestly, I feel like the most important is maybe more on the personal side. I feel like I, um, I probably as many entrepreneurs, I want to, being very ambitious, I want to learn faster, and I recognize that the thing that actually prevents me from learning faster is the fact that. Maybe at times too stubborn and don't listen enough and, um, get kind of defensive. And I think I still do it to some extent, but now I've managed to recognize it a more, a lot more frequently where someone gives me feedback and then I like, sort of have this impulse to be like, well, they're probably wrong. Or like, you know, I've thought about this before. And so instead I'm like, kind of, I, I pause and like, okay, but like. What if they actually right, like, can they be right? Like, is there a world where they actually correct and I'm wrong? Uh, and, and what does that mean? And that probably means that I need to do a lot of work to change the things that, you know. Um, I thought were correct, but maybe they aren't. Um, so that, that has been huge for me as the CEOI think. Um, and particularly what helped with that was one of our angel investors joined full-time because sometimes as a, as a founder, you have all this advisors who kind of give you advice, but they're so like, you know, so removed and they've only. Heard you for an hour, and then they feel like they have the best idea ever. And you're like, well, I've been, you know, working in this for a while, but then when it's a guy who works with you full time as well, and he sees everything you do, then when he gives feedback, like it's just like a lot tougher for me to push back. And I'm like, oh, damn. Like he, he's right. Yeah. Like he notices the like, um, you know, uh, the worst things. And then he doesn't like bend down. And then like, even if I was getting defensive, he was like, no, you're wrong. I'm like, okay. Maybe you're right. I like I give up so that, that's been one of the massive things for me.

Sean Weisbrot: Thanks for watching. If you liked this insight, I've handpicked another video for you right here on the screen. For more actionable strategies that get you real results, hit subscribe.