A SaaS founder's reality check vs linkedin AI hype | Vlad Zhovtenko @ RedTrack

Speaker A: Foreign. Hey there. Welcome to another episode of Sounds Unbound. And with me today is Vlad from Redtrack. Welcome to the show.

Speaker B: Thank you for inviting me to the showtail. I hope I'll be able to share some insights with the audience and we'll see.

Speaker A: I'm sure you will. I remember in our previous conversation you told me that you've been on the Internet basically from 1997 or something like this. Yep, I was fine. So. So it would be awesome to start with your vision of what is going on right now because everyone is freaking out. I've been talking to a lot of people who don't know what happens to their job, don't know what happens to their tools, don't know if anything that used to work still works. How do you see that with everything you've seen already?

Speaker B: Basically that's I would say recently one of my favorite questions, to be honest, because when it was, I would say that 1997 was early in 10. There were already like even.com was already building up. But for me it was the first jump into the technology and where I was, it was still very novel. Most people didn't really believe it was serious. I was fine with that because for me it was just an opportunity to do things better. I had access to the information without going to libraries. I was working part time and I was a student. So for me, doing my part time job as a receptionist and having access to Internet allowed me to study without leaving the office. And that was an amazing resource. And then eventually everybody started to build Internet companies. But then time went past and Internet companies disappeared. You don't hear anybody say we're Internet company. My first business was an Internet company. But in essence we're doing websites, some consulting, some printing. What is happening today in my experience is with AI, it's exactly the same that was happening with Internet somewhere. It was 30 years ago. It's just that things are um, speeding up a bit. For example, took five years for the Internet, may take just one or two years for AI. So I would say that what we have, unlike with all due respect, the bitcoins, the NFTs or the other hypes, we reach the point where we have the new technology which changes the way people interact with information, how they get it, how they consume it, how they produce new knowledge, how they produce their work. And those things will stay our Internet will stay the final shape. I don't know, to be honest, because I'm part of the process, see how it evolves. But as long as you Start to treat it as a tool that gives you advantages in the work you do, you're fine. And then you decide, okay, do I need to use this tool? Do I want to learn it? Do I want to master it? At what level do I want to master it to get the benefits it provides? It gives a clear picture. And some people were not online 30 years ago, some people were not online 20 years ago. Was their life buried because of that? No. Maybe their job did not require the Internet to be efficient and that's totally fine. Same with the AI. Yes, it will remove some of the jobs, it will create new jobs. And seeing the evolution of the whole society and being a sociologist by education, I can see that AI will come. AI will change again the nature of the work. It will stay until the next cycle of the evolution and we'll see what that will be.

Speaker A: I know that you've got a daughter and I asked a couple of people here on the podcast who have a bit older kids, what the kids think about AI and they all said they hated. They were saying, what are you doing? There will be no jobs for us. I got a seven year old, he's still very excited. It's the coolest toy in the world for him. What does your daughter think about AI?

Speaker B: On the one hand, she uses it and interestingly I observed she actually used it as a tutor. Like she asks a question, gets the answer, but then asks clarifying questions why that is the right answer, Is there a complete answer? And if the answer has passed, all the things the, that she doesn't know, she actually digs into what it actually means. That's a very interesting interaction. But besides using it as a tool, she also says, hey, I don't really like it because it's not nature friendly, it burns tons of water, tons of resources, and they're not endless. For her, that's a, uh, different story. So she sees it as a tool and the tool has a price. So she tries to minimize the usage of the tool to minimize her, let's say, impact of that price.

Speaker A: Interesting. I think maybe one day I'll do an episode with somebody from Gen Alpha to figure out what they really think about AI. But before we jump into Redtrack, maybe a little bit of your background and what you actually do with it. Like how did you come up with the idea and who are the people that you're serving now?

Speaker B: That time, um, I went online. I was just casual browsing the websites like everybody sounds funny at the time. The company name.com, you usually end up on the website. When you found Yahoo, it was the portal and it was like years before Google actually or before any search engines proper one mostly like catalogs. And then for all those just browsing the websites I stumbled to click that it was already there. I think for me it was the first digital marketing publication. I read through a couple of articles people discussing like media buying, website building, digital marketing referral pro. Wow, this is the thing I want to do. And that's how I jumped into digital marketing. It was year 2000, I left my job at uh, Ancion for digital media startup. Everybody can see that it was a uh, stupid move. I guess at that time already like career all clear and going to uh, digital media. Never looked back, enjoyed the ride. And eventually it took me to the role of digital marketer, then head of digital marketing. And at that times we're running a lot of paid customer acquisition. We tried to entangle the black box of all the channels and we're doing this paid acquisition for SaaS company. And at that time it was considered even impossible to, to do paid customer exclusion for SaaS companies. And SaaS companies themselves were a new world. And we built our own analytical system. It crashed actually because we tried to add more and more data points and there was no technology to handle all this data we wanted to handle. So we built a ah, humongous amount of different Excel tables interconnected with all the reports analytics updating them daily. But that left an impact because okay, I was the consumer of the technology, I have some experience as a consumer of the technology. I think it can be better. And I think there are like a lot of media buyers who may benefit from having those solutions and without building on the sales, without having those granular years of building up layers of data and experience that coincided with the fact that the market trends in media buying was for me it was like screaming, the third party cookies are about to go into decline. And the first party cookies and the first party data will became really important. And that became basically the foundation of right track as a concept. So how can we allow media buyers to collect own and leverage their own first party data about their campaign performance to make efficient media buying decisions? And what started with ad tracking and conversion analytics, then attribution, then start to evolve into more complex systems as we were adding the other repetitive tasks that media buyers are doing. And so at the moment we are looking into actually remove the need for manual repetitive work from their media bias daily operation to make it creative again.

Speaker A: Okay, all right, how has it been Changing because we have been moving into this cookieless world for a while. Is accuracy and attribution still a thing?

Speaker B: It is still a thing, but what we need to consider what is actual attribution. What media buyers need is they need some consistent set of data to make the decisions against and evaluate the outcomes of the decisions against that consistent set of data. Attribution is not about knowing 100% accurate truth about each and every single conversion, each and every single touch point. No, it's about having the foundational level accepted as truth, measured consistently across the same principles, so all the decisions and their results may be evaluated against it. So, so you change something and on that foundation level you see positive outcomes like your ROI improves or cost per acquisition goes down or something else changes. Okay, that was a positive decision. Let's stick to it. If that measurement is consistent, then you can do the evaluation. What happened when the market shifted is that a lot of media buyers were relying on the data of their ad platforms. Meta, Google, later, TikTok or uh, something else. And with the third party cookies, those platforms had both sufficient access to the information to provide accurate reporting and accurate consistent reporting. And even the fact that they were taking credit for additional conversions wasn't a big issue because you can still see consistent sets of data. It works. You can make decisions. What happened with the rise of the first party data and basically after iOS 14.5 came to the media buying impacting a lot of operations across all the ad platforms, is that the whole process of binance shifted from programmatic, where platforms had all the information and media buyers, like humans were setting the rules to let's say algorithmic, where the rules probably are being set by ad platforms and the media buys are just setting the boundaries. Okay, this is my M target goals, this is my target audience and pretty much that's it. And to ensure this programmatic efficiency, they need to feed the information about conversions. And basically that's the type of the customers they want to attract back to ad platforms because pixels are not doing that right now. And to feed that information back to ad platforms, they need to capture, process, enrich, normalize and send back this conversion data, which is based on the first party data as performance data of their campaigns, clicks and conversions.

Speaker A: Yeah. Has AI also been making attribution harder in any way?

Speaker B: Not yet, but it is coming because so far the biggest change of AI happened is that the a lot of search traffic now is uh, genai search. So people not googling for you, for your business, not getting to website from search results but getting to the website from the LLMs. So generative results. However, both of those traffic channels were attributed, but they're not managed directly by the users. They can create the content, they can create the third party publications linking the web properties, so there is no direct management. What can happen can be the impact if humans will start to use AI actively to make their conversion decisions like purchases, booking services, buying tickets, something else. Because those actions will difficult to connect to the actual ad impressions. Unlike now, because I saw the ad, I make a click, then I can be attributed some way. But if I see an ad and then I talk to whatever client as please buy me this thing I saw just online and he'll go and buy that stuff, then it will make attribution slightly different and difficult. The biggest challenge that may happen, and it's not yet happening by the way, is that the volume of human clicks will decrease and the volume of added clicks will decrease and the volume of actions done by AI agents will increase. And this will mean that they will get back to the state which we had so some 15, 20, 25 years ago with the black boxes where you run ads on one side and then you have the outcomes on the other side and then you try to make sense of what we did here, that our sales grew. Yes, this can happen, but because AI can process way more information than humans, we will use AI to match the actions to the outcomes and still make efficient decisions. So yes, it will change most likely the way people do media buying, but also people will adapt like they did in the past multiple times.

Speaker A: Do you think there is a way for companies to adapt now or start doing something so it's easier to navigate that change? And are you in a way trying to again balance and estimate what's coming

Speaker B: with Redtrack as a company? I think what everybody who is engaged in paydex and is not yet still collecting their own first party data, they need to stop, consider their options and start doing that now. And why is that? Because first party data is not only potent for conversion APIs that we mentioned, it's also important as a way to power up the decision making of the AI agents that are already available and most likely will be used a lot in the coming 12 months. And what means that it will not be the agent who has the better agent, but who has the quality data set for this agent to consume and make decisions. Because we all have access, for example, to Claude, we all can build agents there, workflows, et cetera. But if that agent will not have access to the data about the business performance. These raw records are not the derivatives that matter produced or Google produced based on the data they received, but actually the records of the business. Then those agents will help media buyers make efficient decisions. Right now is to start accumulating the data and not only clicks and conversions, but also the lack of changes that you're making towards all your paid campaigns. Because then six months down the line you can feel, okay, this is my performance, this is my actions. What do you think actually is going on? And maybe, I would say maybe highly likely you'll get amazing insights. And I'm telling that because we went through that experience a couple of months ago, just recently connected the agents to our historical data and start to uh, ask questions about the performance. And we got the insights and we were never expecting to get despite analyzing it on a like say daily, weekly, monthly basis. And those insights changed how we approach our go to market, our actions, our strategy, and they changed dramatically. And we see they already in the first two months of making those changes, we see the positive results.

Speaker A: All right, it's interesting what you just said. So you were still analyzing what you guys were doing, right? You had all the data sets available, so you were doing the work. It's not like you started it with AI and yet you then ran it through AI and it presented you with insights that you applied. And I feel like in a way sometimes we trust the insights and the uh, to do list of AI more than we trust our own gut feeling, our own history with the company or with the customer. I just wonder how to balance, okay, this is what we know, and We've known for 10 years of running our business and this is what AI gave us. Like when does it outsmart us?

Speaker B: It doesn't outsmart you. That's the thing, what it can do. Could I, uh, have found the same insights? Yes. But it probably took me two to three weeks of digging through all those records for the past three years. And I doubt that as the CEO I have those two to three weeks just to dig into the data sets. So what it does, it actually helped me to do. And we'll jump now on this topic of tons of processing work in minutes produced me a set of different insights and then I used my gardening experience. Say, hey, this is the line of thought we never considered. Let's explore it more. Another days of process work condensed into actual minutes confirms. Then we do a couple of cross reference points to confirm. Then we actually run it with the actual people and the actual data sets in our records to Say this, which. Yes. Wow. So can we produce the same results? Sure, of course. But should we spend all those, I would say even days to just digging through all those raw records, shifting them left and right to generate those insights? No, it's going back to that experience with me going online. My peers were taking public transport, stuck gems, get to libraries, request books from catalog, wait for somebody to get those books to them, then writing down the quotes, working in those paper formats, I would just find the same books online, do copy, paste of those quotes and type in the whole three. This saving me hours of work. AI does the same.

Speaker A: Yeah, I think I used the wrong word though. Not outsmart, but out. Trust us in a way because I find myself in this situation quite often that I know I must know the answer or I have the experience and yet it's still there. Like, okay, let me just double check, triple check with questions, plot with ChatGPT. And I find that sometimes I still had to rely just on my own knowledge. And that's why I'm asking, okay, where is that balance? Yes, processing work, amazing. I think it's the way it can just distill all the knowledge of the Internet basically into three lines is amazing. But then how do we trust those three lines? And if your actual knowledge is slightly different, who is the winner here?

Speaker B: I'm m the winner. I make the decision. I'm not letting AI uh to make the decision. AI helps me make the decision by actually condensing all this processing work. But look, it's not an omnipotent entity that knows everything. It actually makes mistakes. The way it works, it's making very educated guesses.

Speaker A: This episode is sponsored by Reworkful. Looking for new ways to find customers for your SaaS business? Consider building an affiliate program. Rewardful is the easiest affiliate tracking platform to set up, manage and scale. For SaaS companies, building a successful affiliate program can be a little bit intimidating at first. And that's why Rewardful has taken what they've observed from their most successful customers, affiliate programs and distilled that, uh, into an exclusive online course. The exciting part part, their affiliate marketing course is absolutely free. Start the course at academy.rewardful.com academy.rewardful.com and turn your biggest fans into your best marketers.

Speaker B: Because what does it builds up those words into how it should most likely be connected one by one, one by one, without actually understanding the meaning of those words. For example, the intent is a huge, as you mentioned, like condense all the Internet, but there is a lot of incorrect facts on the Internet. So if this train of thought for this particular prompt will put it on the wrong path, it will give a wrong response. So that's why, by the way, it's very important to start building your own AI. LLMs Wikis Karparti recently showed this. It's just one of the examples how you can build your own custom repository for the knowledge that AI would not have to relearn when it does something for you every time from scratch, but actually can use your own past insight decides the past work done. This becomes important. But in the end, look, I don't think it's reasonable or will be reasonable to delegate the decisions unless I don't care about them. That's part of the job. Yes, it can do processing upgrade decisions should rely on humans. Yes, it can prepare the decision, maybe, but humans can still validate it. And otherwise. The processes where AI currently is great were called automation before, where the decisions actually known, like the boundaries for decisions. What is good or bad? What is good outcome? The bad outcome. It was done by automation. It was before the computers. There were mechanical systems that would automate the decisions in the past, like in the steam engines. There was the mechanical thing that would manage the pressure in the steam engine by basically making automated decisions. So AI can be good for even making those decisions if they are, um, discrete.

Speaker A: All right. I think it was just yesterday. I'm, um, going through a course and, you know, it's very black and white because it is about the body. So the skeletal system. Let's say I was writing a test about, uh, how all different bones and, uh, whatnot are cold. And I was like, okay, I just want to double check. I'm like, just to be sure. But I did it after answering the questions and ChatGPT gave me the wrong answer. And I was like, wait a second. It's skeletal system. It's just about like, everyone knows the bones. You enter any encyclopedia and it's there. Why are you making a mistake? And I was so shocked that I was still okay, maybe am I tripping? Am I like, no, he's making mistakes. It's so shocking sometimes. And yet we all see people say, hey, we are firing our whole, I don't know, engineering department, marketing department and whatnot to replace it with AI agents. And it's, I don't know, it's, uh, unnerving a little bit. You just said processing work. Yes. Huge. Yes. For AI. What is a huge.

Speaker B: No cognitive work. First of all, there is different types of work, but let's talk about us as a SaaS company. So most of what we do is either processing or cognitive. And processing is when you actually don't create any new information. You basically move that information from one place to another or you act on the pre existing routine or process where basically human judgment, let's say is not required. What is required is time and skill and how fast you can move this data from one place to another. And this is a great place to have AI. If you consider the workflows of the company, then identify those workflows and see which one have larger share of this processing work. That's the right place to start implementing the AI. But the trick here is it's not that, okay, we have five people throw them. Are you using AI? Told them I'm not using AI. No, it's like everybody is now using the AI and not just like for search for even prompting. No, it's rebuilding this particular workflow in such a way that is based on the AI, for example initiating it, doing the processing work and then showing the result to human to validate. If for example two years ago, um, and we started along the past year, probably choice could say hey AI is here, let's try using it. A year ago it was saying to her behave, just use AI. We'll buy you subscriptions, get experience, get those like habits of using it. This year we say hey, this is our AI manifesto. This is how we use AI. This is how we approach the transition of the company from AI super company. So from some aspirations to unstructured process, we evolved to have a very structured approach. I know we're not using everything okay, this is our main LLM. Everybody's using it. And in our case it was Claude. Not because it's bad or something, but because Claude makes connecting to the data sources all the slack notion, Gmail, et cetera so much more intuitive than other LLMs. Then we stick with Claude. And that for us was the only reason, the main reason. Okay, this is the LLM that we're using in a company. Can you use something else for our own needs? Sure, of course, fine. For some specific applications, some other jobs other people may do here. Of course we can have other tools. Like I use Gamma to build up the decks and actually don't like code design. Code design can build amazing decks, very beautiful visual. But for me editing Gamma, even if it's like very structured is so much easier. Then I'll stick to Gamma for example. But I'm not going back to Google Slides or PowerPoint.

Speaker A: Oh no, I don't think anybody. Anybody is. But yeah, it's interesting. How do you see like who is the real champion of AI basically in the company? Because eventually I think we're all going is AI. It's all the hype obviously, right. And every company wants AI there. And leveraging AI, empowering everybody in the company to use it is great as an idea, but let's face it, it's not getting cheaper and it will probably get a lot pricier with time. So how to find or how do you maybe already identify people who really should be using it, who really find value with the AI tools with their workflow and can bring value back?

Speaker B: Good question. Because it goes back to what you just said. It's not like who will use it, who will not use it, like everybody will use it. Like now everybody is using Internet for us as a business, it's just also the process of implementing it from let's say individual activities to a company policy. And so what we learned from our experience, from other sources, from amazing ways to learn online, how people do it, is that there are different levels of using AI. And let's start like basically using them the same as the Google like prompt. Write a prompt, find me something. Fine. Level number one. Level number two is actually intentional prompting. That's where people are very specific when they type in those prompt. And this prompt may be very huge. And yes, they may use another prompt to create that specific prompt, but it's already not like hey, who made that quote? Then you connect it to data, that's the level number three and it still covers the individual usage. And this is not moving company to be AI powered company. What happens is the next stage where you rebuild the workflows around AI and this moment is not like who's using it but which workflows in the company you will rebuild. First you can think about AI. Uh, it can either cut. By cut we mean not jobs, but actually time used to do the job, all these repetitive processing actions or AI can create or it can do both. So for most companies that are already existing, not just starting, not all those AI native companies seems the more evident path is to cut, for example time to complete the tasks and then they can identify the workflow flow within the company that has a large realm of processing work, support tickets, summaries of the phone calls, prospect research, something else and rebuild those processes using AI first and then humans as just the people who make the final judgment and approve the task. And this is where the company starts to transform and then those AI powered workflows can become agents performing work. And then eventually, not so distant future will have a mix of human and AI workforce working for the company and the future leadership. So the company will have to manage both. But now we can just say, okay, where we are, each of us are on these stages. Because for me, I'm not even in the workflow stage. I'm somewhere between intentional prompting and prompts connected to data depending on the tasks. That's my top. We have people in the company who like, our engineering team is way, way more advanced. They do miracles from my casual view, but for me, it works. So what we are doing right now is we're basically saying, okay, guys, in engineering, you are in your own world, but everybody else who are more, uh, down to earth, like without all those special skills, we pick up people, let's call them like champions, People who are just more active, have some reputation within their teams. And we actually say, hey, guys, help us. Because you know what the job you do on a daily basis, like in support, for example, and all the multiple processes, not multiple, several processes they have in support. Which release has the most processing work. Okay, now let's see how we can, uh, rebuild the process. So they're not rebuilding themselves. We help them, but they pick up the process and then they champion the transition. Let's rebuild this process so it's done by AI first. And now because you're spiral rebuilding the process and you are AI, champion your particular team, help your peers leverage the change. Because if somebody's not using it, that person becomes a roadblock. So everybody should use base unit, but it's not everything like, okay, you stop doing your work, you start using AI. No, it's okay. We rebuilt this one process, then the next one, then the next one, and we'll see how we'll get us.

Speaker A: You were. Because you mentioned agents, right? And I just had another podcast with the founder and what he said is, I can manage three people. Managing each agent is like adding another person to your pool of people that you manage. And before we decided to record this episode, we were talking about how not everyone is a manager or a people manager. Do you think everyone can be an agent manager?

Speaker B: No, I don't think everybody needs to be a manager.

Speaker A: Okay.

Speaker B: Because in the end, like I said, what we need to be able to do in different roles is to do our job best to deliver the results that are supposed to be done by our jobs using the best tools we have. Using existing workflow that was built, which is powered by AI and managing agent other different things. Creating that agent is even more complex thing. And I'll give an example about, uh, should everybody be a manager? And this will take us back to me being in the reception in 1997 of Ernst Young being in receptionist. When I joined the reception team, we need to print out addresses on the envelopes. And we had a typewriter, a very fancy typewriter which we actually clean up. Like it has this special function to fix typos and mistakes, like untype things. But then I said, I'm not good at typewriting. I don't know how to use typewriter. So what I did, I built the template where you can actually type in the address and just feed in envelopes into the printer. It would print out everything better, faster. The team that was before me in the reception and I think everybody else was changing pretty fast. I was very consistent last day. Nobody was using this process I built with the, uh, word file and using print instead of typewriter because it was alien to them. So what I made sure when I was training the next team for the reception because I was even receptionist, but with more experience. I never told them about typewriter. I just, hey, this is how we print addresses on the envelopes. And everybody was happy. I replaced one tool with the other. Was any of us the manager or the typewriter or the manager of the printer? No, we're just using it. Same with AI. Should you be the manager of the agent? No, not everybody should be the manager of the agent. But should you be able to use the processes and the AI that supports this process efficiently and the two. Sure, you have to. But using and managing, I guess, uh, were different things. And people always try to put on those facing titles. Hey, we're managing, we're managing this, we're managing that. We are the managers. No, we are just using it. It's to call the driver the manager of the car.

Speaker A: Honestly, thank you for sharing this because currently you go to LinkedIn and everyone has an org with 64 different agents that they magically manage and everyone's firing and replacing people with AI. It's interesting actually, uh, just a couple of minutes before we started the recording, I saw a post that said in China it's now illegal to fire people to replace them with AI. The question is how they're tracking it, but it's a different kind of question. But I think in this whole hype and anxiety creating social media world, this is very down to earth. And yeah, uh, thank you for that. I haven't heard this in a while.

Speaker B: I believe that the boundaries between all these social platforms are um, dissolving because, uh, LinkedIn used to be a place where people would actually share their insights. Now LinkedIn becoming a place where people sharing the best posts that will be picked up by the algorithm to get them exposure. And because everybody's reading how everybody is removing their teams because of the AI, this becomes popular and people start to produce more of it. I can see those trends again across other social media. Than consume something becomes popular, all the key creators will start to do that specific type of content within their niche. Everybody. And you can say, okay, now this is popular because everybody's doing that right now. Same with LinkedIn. Everybody's writing about not what's actually happening, but what other people want to read.

Speaker A: Yeah.

Speaker B: And that's why it's so difficult for me to post on LinkedIn consistently because to produce a quality context time and actually it falls between all the algorithms because I don't want to write things that algorithm want to pick up. Not writing, but I'll pick it up.

Speaker A: Yeah. All right, so I just have one more question for you because you've told me that after raising your first funding, you've actually gotten to this very sustainable workflow and now you're not sure if whether you need another round of funding or not because you've got, you're in a great position basically. And that's not what I hear from most founders because everyone is now, even bootstrappers thinking vc, uh, funding is the next thing for everybody because with AI, even though you don't allegedly need to hire people, it's becoming so expensive and AI visibility, discoverability, defensibility and whatnot becomes so expensive that they need VC funding. So is there a hack or maybe just something that people have to understand to come to that stage and to feel comfortable right now? 2026.

Speaker B: Yeah, I would say we are not looking to raise right now because we have our Runway and it's fairly large. So we have sufficient time to uh, evaluate when and how and why we'll approach the fundraising. But I think the primary thing is that first have with everybody who is going for the fundraising need to have start with the vision for their business. What are ah, they building? Why? That's the key question. And then if they have this answer, and this should be a very honest answer, it's very difficult to come up to. It may evolve with time. So it's something a question that probably each founder should, or we as a founding team, we ask this question to ourselves every six months. Is it the same has something changed. And some things stay like the general direction, but on the tactical level things change and we adjust what we're doing. But then if you have this vision, then uh, you can answer the question why do you need the money for? And how you plan to use it to get to that vision. And with those questions now you can decide where to go for the money. Should it be vc, should it be growth equity, should it be somebody else? Because different, let's say types of investors have different investment criteria. Somebody looking for hyper growth, somebody looking for the profitability, somebody is a sector agnostic, somebody looking for specific focus on a specific industry, niche product sector. Until you have first two questions covered. What are you building and how the money will help you get to your goals? You don't even start fundraising.

Speaker A: Not a good reason to raise.

Speaker B: Again, it might be very subjective opinion. It's a technology, yes, it's groundbreaking technology. Yes, it will impact us in ways we can't even foresee. But it's not the first technology that appearing in human society and I hope not the last technology we have to adapt and it should not like.com.com probably killed quite a lot of businesses that were not using Internet and then the business that were built just being a website get the place towards the SaaS companies like something else but doesn't mean that's end for all the other businesses. I'll give you maybe not the best example, but look, uh, Coca Cola, it was there before the Internet. After that with the AI, people will still be buying soft drinks and it's always a question what are you building, why you need the money for? And then go get that money if you actually need them.

Speaker A: Yeah. All right, thank you so much for the answer and for sharing your opinions here. Like I said, it's been very down to earth and actually to quote chatgpt, very refreshing. Uh, so thank you for being here for your time and hopefully we get to do it again sometime to see where you're at.

Speaker B: Sure, it will be my pleasure. Thank you for the opportunity and enjoy

Speaker A: the thank you too. Take care. Thanks for listening. SaaS Unbound is brought to you by SaaS Group. We're a long term home for a great B2B SaaS. We buy, keep the team and brand DNA and help with the boring stuff like hiring and finance so founders can truly focus on building great products. If you're a founder who'd like to be featured or explore an acquisition, reach out through the form on our website or email me at Anasas Group.