Latest / The Collaborative Canvas Podcast / Market Research Is Changing: AI, Synthetic Data & What Comes Next | Vivek Bhaskaran

Transcript

Market Research Is Changing: AI, Synthetic Data & What Comes Next | Vivek Bhaskaran · The Collaborative Canvas Podcast

• speaker-0: Secret mantra behind Question Pro not going to the VCs. speaker-1: 2005 when I started the company, my goal was to just make a little bit of money so that I could, you know, live my life. SUA is really an analytical exercise. It's not a creative exercise. Anybody who says what's gonna happen in the next three months, six months, one year is probably bullshitting too. Nobody knows what's gonna happen. speaker-0: Where do you see Restec is head? What is that biggest misconception these guys carry with them? speaker-1: I think product market fit, really. I mean you could build an amazing tool, but does it solve a problem and are people ready to pay for it? So now if you say like I just want to do research with AI, like okay you can do it, but what is the ultimate goal? The ultimate goal is to increase revenue. A simple definition of product market fit is speaker-0: As an entrepreneur, should we really focus on just analysis, or should we actually see, you know, considering the current market conditions speaker-1: Earlier on in my career, I never bothered kind of building connections. I don't WhatsApp my employees. I always slack them. speaker-0: So what is that advice you would give to aspiring CEOs or founders in in in the tech industry? speaker-1: If you believe in something, you just have to grind and do it. Yeah. Yeah. speaker-0: Welcome back to the show. In an era where software categories get rewritten overnight and artificial intelligence dominates every board deck, the market research landscape is still facing a profound question. How do we separate real signals from automated noise? To unpack this, we have joined by someone who has navigated every major shift in the research and technology over the last two decades. He founded Question Pro in 2005, famously scaled it to into a global insights hub, an experienced powerhouse without institutional venture capital, and continues to think like both as a CEO and a software architect. Please welcome the founder and CEO of Question Pro, Vivek Bhaskran. Vivek, it's such a pleasure to have you on the show. speaker-1: Same here, same here. Thank you for having me. speaker-0: Vivek to start with, as usual, I mean I have something that I've o already learned from a few podcasts over a period of time that, know, ⁓ people that know us, ⁓ you know, within within our fraternity, maybe ⁓ very less, but there are a lot many people who really wanted to know us. So how about we start with your origin story starting from two thousand five? speaker-1: Yeah, so so I'm originally from India. ⁓ so I grew up in Calcutta, I'm a Tamilan. So I grew up in Calcutta though. ⁓ and after high school, ⁓ I really wanted to do computer science. This was back in ninety-three no ninety-five. No, back in ninety five. I really wanted to do computer science. ⁓ back in ninety five. I I gave all the IT exams. ⁓ didn't get into any of the ITs. I didn't get to any of the entrance exams. I I think I gave eight entrance exams at the time. ⁓ I didn't get into any one of them. So I was like, my life is f Right. I said, I'm dead and I left India and I went to Russia. ⁓ don't ask me what I did there. Probably what kind of woman. So then so I went there to study actually technically, but studying. So ⁓ after a year or so my mom came and visited me and she's like, Obviously you're wasting your life, so you need to get out of here. So I ended up ⁓ transferring to the States. I ended up Going to a Mormon university called Brigham Young University in Utah, actually. ⁓ finished off my undergrad there, did my computer science, ⁓ computer engineering degree there, finished it off in 2000, and then I moved to Seattle. ⁓ then I started Question Pro in about 2005, is when I started Question Pro ⁓ in my garage. Actually, back then there was no AWS, no no way to set up a server. So, you know, I just I I just bought a couple of servers, put it in my garage, got a team online. ⁓ that's how Question Program started, ⁓ in my garage in in Seattle. And ⁓ since then I've been ⁓ you know, that's kinda the the true origins of how how we kind of landed on this. speaker-0: I if I really have to tell you, 2005 is when I started into the market research ⁓ corporate world. Before that it was more of an interviewer role for three years. Right. But ⁓ considering that you started Question Pro on in 2005, you had any inclination or do you know anything about market research before that? And so how come this idea, you know, started in a garage? speaker-1: The real backstory there is that when I was at BYU, which is the ⁓ school, the college that I was going to in ⁓ in in Utah, there a market research professor cornered me. and he's a market research professor at the management school and he wanted to build a survey tool. And where do professors go if they want cheap labor? They go to the computer science lab where all the students are there. And so he came into the computer science lab, this was in ninety eight, and he said Hey, ⁓ I need somebody to build me some software. And being the poor student that I was, I was like, first thing I said, okay, how much are you gonna pay me? ⁓ And he said, $17 an hour. I was like, yeah, I'll do that, a lot more for that. So he and I partnered up and we started a company called SurveyPro. ⁓ this was back in the school. This was like a side hustle for me, frankly. Right. So he wanted something done, I'd do it. ⁓ that didn't go anywhere. ⁓ and then so we kind of ran it for about a few years. ⁓ he ran it actually. He I was just his you know effectively his software developer really. ⁓ and then we split apart in two thousand three, two thousand four is when we split. ⁓ he kept SurveyPro, I was saying I need to do something different, ⁓ or I wanted to split with him. So he kept SurveyPro, I kept Caution Pro, and that's how I started Caution Pro. speaker-0: What is that secret mantra behind Question Pro not going to the VCs as of now as yet? I mean of course you're like what what have you done initially? I mean a lot of people actually within the first or the second year look for VCs. speaker-1: Well, ⁓ I'd say that th the honest answer is ⁓ I didn't know I could go to VCs. Right. First of all, that's the the brutally honest answer. Because think about it, ⁓ when in two thousand five when I started the company, my goal was to just make a little bit of money so that I could, you know, live my life. That's it. I don't have this kind of like massive ambition of doing something pretty crazy, pretty big, honestly. ⁓ and then we started making a lot of money. Simple. ⁓ you know, online online research was hot. We got to number one for SEO for search. This is back in two thousand five twenty six really, right? ⁓ one of the things that we did pretty smartly was that like if you think about it, SEO at the time, people were putting that on under marketing teams. Okay, which is search engine optimization really, right? But SEO is really an analytical exercise. It's not a creative exercise. So it should be really under engineering, not under marketing. speaker-0: But it's it's still been ⁓ under the marketing route. speaker-1: Yeah, but in now today people put it under growth under marketing really, right? Not marketing kinda in terms of brand and things like that. Marketing under kinda analytical marketing really, right? So I'm an engineer. So it was pretty easy for me to get to number one on Google for the word survey, the word survey software, all this stuff really. So we had a ton of inbound leads. Once you have a ton of inbound leads, you just have to sell it to them, really, right? So the going back to your question, like why did we not raise money? Well, because we made money. Right. So speaker-0: And and when you say that, you know, y you started making yourself ranking number one on Google through SEOs, you know, fast forward now it is GEO. Are you still doing the same? speaker-1: They're pretty good. In fact, you in fact I I was talking to one of my buddies runs a company called Scalepost. He does this kind of like he's doing this AI based kind of thing and he's like, dude, you know, when when you ask Claude about survey platforms, you guys the answer is you guys are better than Caldrix. So so obviously we're doing something smart in that direction also. But that's but the point I'm trying make is that like, you know, some of these exercises are analytical in nature and not creative in nature. So it was easy for me to do it because I'm generally a you know, I'm an engineer, I'm analytical. By definition. And so that was easier for ⁓ for me. So right now, you know, we've invested in all this, all the GE also, and you know, making sure that you're you're available. And when somebody asks a question, we get a lot of leads right now already. When people ask us, like, you know, if you ask Cloud, hey, you know, who does Qualtrics compete with? I don't want to I don't want to renew with Qualtrics. Who do I go to? You know, go ask it, you know, ask publicity, and you probably get questioned for as one of the answers, really, right? So Going back to your point about like, hey, why raise money? Because we started making money and we were reasonably profitable. ⁓ and so we continued continued the journey, right? And every time, you know, we thought about raising money, we said like, okay, actually we have the money where we can continue going. And we've been profitable actually we've been profitable since two thousand five, to then say. I mean, like, look, I can't I have to be profitable in order to pay my bills. speaker-0: So Vivek, you've seen ⁓ cross ⁓ you know the things that have evolved over a period of time. You know, when you started nineteen ninety-eight. I mean I think ⁓ in terms of timelines, our story remains very much the same. It's just that you question pro I'm still here. So ⁓ considering that n you know when when you started with the professor, that's survey pro business and ⁓ when you started speaker-1: Here's the backstory. That survey pro business is Qualtrics. speaker-0: ⁓ yes. Now I I I can I was in fact thinking because I've heard SurveyPro long back, I think those speaker-1: So they rebranded to Caltrix. And I I I kept it Cushum Pro. I Cush Pro. speaker-0: Okay. So ⁓ you've seen, you know, ⁓ that you were creating a survey link for a particular survey. You started from there. Now we are sitting at a you know, it an evolving res tech market. I mean, considering that, you know, every other tech firm, ⁓ they either vanish in next two years or they built up something which through the help of VC to us to such a scale that, you know, they don't even have to worry about. But at some point of time there's a diminishing, you know, mark the on the on the graph that they say it's It's growth stage, now it's going down. And Restec has been emerging across. And you've seen every part of it. Based on your experience, the things which have evolved over time, where do you see that, you know, things would be at some point of time? You probably have to let go of things and you have to start thinking in a different way altogether. And then the next question that'll come across is, you know, creativity or analytics ⁓ thing. But before that I really wanted to understand, you know, where do you see ResTech is heading right now? It is emerging, I can see that, but I'll still feel very ⁓ you know, questionable when it comes to seeing how the market is right now versus the emerging technology overall. speaker-1: Well you think about it, first of all, anybody who says what's gonna happen in the next three months, six months, one year is probably bullshitting you. Nobody knows what's gonna happen. Yeah. Right. Look at the look at the underlying subsystems that are changing, the speed at which things are changing. All right. So to say that you know, I'm talking about general technology, really, right? You know, you know, one year ago, if you said that agents are gonna control, you know. With fair degree of certainty and reliability, people would be like, no, you know. And prior to that it was LLMs were controlling, and now today agents can control and behave with the level of sophistication and certainty that you and I can behave realistically. ⁓ so the speed at which things are moving, the the underlying technology is moving, is astronomically insane. Okay? Now Restec happens to be Research on technology, even CX and even EX. Those are three different platforms that we are usually involved in. These are all secondary subsets on top of a core underlying system. ⁓ So going back to your point about like, hey, where does this go? ⁓ you know, my opinion is that like it just accelerates research, changes research in many in many ways, where you know, research is conducted to make decisions, generally. Okay. Research is not conducted to for the sake of doing research to make decisions. So accelerating, you know, what what really it does is accelerate decision making. And how do you accelerate decision making? Either through, you know, we talked a little bit about synthetic data, we talked about digital twins, we talked about all these iterations are gonna come and some of these are gonna stick, some of these are not gonna stick. But what is amazing and interesting in the times that we live right now is that you can iterate fast, you can try new ideas, things that you could never have tried before. Right, that has just simply became impossible. Right. A simple example would be like, hey, look, if I told you I'll give you, you know, a thousand qualitative interviews, I want to do some deep research on all of them and analyze them and come up with signals out of those interviews. You know, three years ago, obviously you couldn't do it because you had to go through a thousand interviews, right? Today you could go to ten thousand interviews. Ten thousand say, like, hey, actually there a lot of signals in those ten thousand interviews, synthesize them, have a data pipeline and say, Okay, what do you want to, you I want to launch the new version of, you know, Axe Body Spray. All right, what are people looking for? Why are they looking for? All that stuff. You can do deep qualitative interviews with, you know, ten thousand people, for example, and actually analyze it and actually produce a hypothesis around it and actually come up with a point of view, which you know, which AI allows you to do today. Right. So so now maybe that becomes a product, maybe that becomes a solution, right? So my point is where does this go? Like ever all of us are gonna create different kinds of solutions based on AI. Right? And each of those solutions, some of those solutions are gonna work, some of those some solutions are not gonna work. And you know, it's it's the Wild West again. So everybody's gonna come up with different ideas and some of them are gonna stick, some of them are not gonna stick, and we'll see what happens. speaker-0: So in the current ⁓ or rather modern data collection world, what do you think the those enterprise founders or even you know the people who are starting up, what is the biggest misconception misconception they are carrying? Because when I speak to a lot of them, they say, I have panels, I'm, you know, integrating AI into it, I'm trying to scale things up. I generally feel that, you know, something is missing there. ⁓ because I'm not digging much deeper into them, but I still believe there's something is missing. What do you think? I mean, it's considering you've spanned the entire two decades into it, what is that biggest misconception these guys carry with them? speaker-1: I think product market fit, really. I mean you could build an amazing tool, but does it solve a problem and are people ready to pay for it? That's it, right? You know, so you could say like, hey, look, I got this AI and we can do A, B, C, and D. So now you gotta still sell it. You gotta somebody say it's like, okay, some some the ultimate the ultimate buyer is a brand, is a company that says, I wanna do something, I wanna increase revenue, I wanna build a new product or I wanna improve You know, improved sales of a particular product. That's the real ultimate goal. Research is in support of that goal, ultimate goal. So now if you say like I just want to do research with AI, like okay, you can do it, but what is the ultimate goal? The ultimate goal is to increase revenue. I'll give you an example, right? So you know, we did some research and said, ⁓ okay, you know, what are the what are the reasons for churn, for example? It's kind of useless. What is the reason? What are we gonna do about it? That's more important. Right, telling me like these are the reasons. Like, okay, what is actually more important is tell me the two things that we can do. Because unless we do something about it, it's kind of irrelevant. Right? Telling me like, ⁓ my God, these are the reasons for research, great. But you actually have to move that even further. So in my opinion, that you know, systems and tools that allow people to give, you know, real actions, contextual actions. And then there are all these constraints. ⁓ I cannot do okay, these are the problems, here's what you do about it. So ⁓ so coming up with tools and solutions that actually, you know, get to actions rather than just say, hey, this is this is the problem. Here's what you can do. You know, if you're a JK Tire, here's what you should be doing. It's almost getting to that point, right? So if you can get to that, then that's what people will pay for, going back to your person. If you build a tool or you know as a as a what are we what are people willing to pay for? A, you know, and that gives you a clear indication of whether you have product market fit or not. speaker-0: And I I think I've been what the thing which I have missed is of course, you know, I think that product market fit issue is there because when when they work on tons of data, they believe that they probably have a better control on that data and they know exactly what is missing. And based on that, what ⁓ what is missing, they create a product. Now when's once the product is being created, when they go it out into the market and say, you know, this is what we are solving, not necessarily it is for every single industry. speaker-1: Right. Or or every single use case, right? Yeah. Or you know, there and so my viewpoint is that like, you know, y if you get product ⁓ a simple definition of product market fit is are willing are people willing to pay for it? That's it, right? A simple definition. It's like a brass tax, right? Hey, look, I built something, you know, I have water over here, are you willing to pay for it? Now if you say no, no, that's it. You don't have product market fit, right? Till you get somebody to actually write a check, you can say like I raised money for it, that doesn't mean anything. You can raise money for it to build whatever you're building. And then you gotta still sell it. You gotta still get somebody to say, yes, I will write you a check for whatever you are produ and that's a clear exchange of value. speaker-0: And it is basically, you know, once you have a clear ⁓ indication that he's going to buy my product because it is now a product market fit, you can s think of speaker-1: You can scale it. Then you can scale it up, right? So you can say, like, hey, look, we want to do a bunch of pilots. And I think I I've seen a lot of stars that are like, hey, we do a bunch of pilots, right? And pilots are a double-edged sword, right? Pilots are easy to get deals done. But can you convert that pilot to an ongoing three year license? Product market fit is not when you finish up, you know, when you get a pilot. It's easy to get enamored by that, to be honest. Like, hey, we got these guys doing a pilot. Yeah. But will they convert that pilot after three months to a three year contract? speaker-0: So I'll probably take a skill set out of it ⁓ from you because you've we've been doing this for a while now. I I've seen a lot of people actually do the POCs initially and they don't be able to convert it. I mean, even if the POC is successful. What is that lacking behind? ⁓ is it client doesn't want it to do that or it was just speaker-1: Do not solve the problem. Simple. speaker-0: But they said, you know, it's an amazing POC that we have done, but I don't have enough work right now. speaker-1: That's bullshit, right? You c I say everything is amazing till I write you a check. I mean it's easy to say it's amazing, right? I mean I've I've I've seen so many of my friends speaker-0: So how would I know, I mean as an entrepreneur, how would I know that, you know, if they're saying amazing, I believe that they loved my product, but they're not giving me enough work. So what is not something that I cannot know exactly, right? speaker-1: You have to peel the onion. You have to peel the onion. They're saying amazing, but they're not saying if you if it's amaz truly amazing, then you should transact with me, correct? I can't say it's amazing and then say I don't want to buy. It's like if I'm trying to sell you something, you say, hey, look, I love this car, but I don't want to buy the car. Well, you don't love the car then. I'm just saying that I'm loving I love the car. This happens like in the more this happens all the time in the in the fundraising world, if you think about it. When you're trying to fundraise, nobody tells you. This is dull shit. They'll just say, like, yeah, everything is great, and they find some random reason to tell you no. Okay? Same thing happens in the enterprise world too, right? So like, okay, what is let's say you're pitching me something. There is no benefit for me to actually tell the TRF, is like, no, there's not solving the problem for me. Right? I and my my definition of this is like, look, if you cannot close a transaction where value is exchanged, that means it's not truly amazing. It's not truly you don't get enamored with what people are telling you. Focus on what people are doing. Right? At the end of the day you got you know you're selling a ticket for a conference. Yeah, I love the conference. Did you buy the ticket? Ultimately you still have to buy the ticket. You can't just tell me I love the conference. And and if you're not buying the ticket, why? speaker-0: probably I I still believe you know the POC works as as the as opening the gates for you. Yep. But not necessarily it would work the very classic case that I can tell you that, you know, when we speak with them, they said, yes, we do have it. But the next, you know, the peel that we do on the onion is that I don't have work right now. So I cannot tell you that I can give you more work. speaker-1: Yeah, that's one excuse. speaker-0: But how do I identify that what is missing in my product which is why they're not giving me? Is it because the market or I should I simply rely on the fact that the market is down? They don't have speaker-1: I don't know, right? I it's it's a case by case. The way I would look at it is I I will probe hard, I will try really hard to try to figure out. I have trying really hard like what is the real and drawing reason for you to say no. Obviously you're saying no at some point, right? And you're saying it in a nice way. Right? That's what it's actually practically happening, right? They're saying no in a nice way, saying that, well, I don't have the budget, I don't have this, I don't have that, really, right? They're just saying it in a nice way. So you have to figure out like what w the solution that you offered. is not fitting in into their into their ecosystem. And many times it'll be like other issues, like not even your product maybe. Right? They are doing other things really. And you can and then you just there are certain intractable issues at that point. Like you just can't control those variables. You control some variables. You don't control a bunch of other variables. So go into that knowing that really, right? And so if you don't you know you don't control all the variables. You control as much, you know, first of all you don't control many of the variables. So you try your best. And then and but the one thing that I've seen, like people get enamored by, ⁓ look, these guys are doing this, these guys are doing this, like in terms of, hey, you know, it's easy to say let's do a free pilot. Because there's nothing to lose. ⁓ And I hate free pilots. I hate it personally. As like we don't do it. No, because you're not there's no skin in the game. There is you know, this is a partnership. Pilot is truly a partnership, right? You know, we are trying to speaker-0: W was that ⁓ your reply when you were very s you know, new? Or now you can definitely say because now you're pretty set. speaker-1: No, y today today probably we have more leverage, you're right. ⁓ but even earlier on, because there was a real economic you know, you know, I've s I've told people where like, no, they wanted a free pilot, I no, we just don't do it. But even earlier on, you know, because we needed the money, frankly. It was the other way around. We really needed the money. I was like, Okay, you want to do a free pilot? No, I'm not gonna do it. Right? And yeah, you know and I'm glad that we did it, quite frankly. We'll otherwise we would waste a ton of time on something that people are not really committed towards. Right. I mean it's a commitment, that's the point. Right. And you know, ⁓ and we've tried to you know, we'd have to go through that process. speaker-0: I mean considering that we tried our best, ⁓ but I think something that I'm taking from this conversation is that I think probing would definitely help me out understand, you know, how deep I can go within that product fit ⁓ thing. Because if if somebody's saying that, you know, your product is good but I don't have the money or I don't have more projects for you, then there's definitely something which is missing and we really need to know and then if there's nothing then we can probably move on to find somebody else. I think that's where we should do that. But do you think ⁓ you know, since it is more of a B2B environment, I mean even you sell it to B2B to cater to B2C. ⁓ and and I think I've been by far been almost like two and a half years now that I've seen a questionnaire, but I still believe, you know, people do a 30-minute questionnaire. To do something like this and probe harder, if I create a because I I've tried doing it, I wanted to minimize it to like 10 questions. Yeah. People were not even filling those. In a modern documentation or sorry, in a modern ⁓ data collection era, are questionnaires still being the go to strategy for people to know exactly where whether their product is fit for the market? So what is what what should speaker-1: ⁓ The way I think about it is like, you know, the whole if you think if you f rewind fifty years, forty years, surveys were designed to extract signals. Yep. Okay. But here's the problem. The the survey instrument was designed to extract it was actually started in social psychology, the core the core survey. Like I want to understand how you think. Okay. So back fifty years ago, you could give me like eighteen or hundred signals. Each each individual could give 100 signals. But there was no way to process all those hundred signals from you know, let's say 100 people giving a hundred different signals. Just too much information. We could not process it. So the survey instrument was designed to say, we will capture signals at source, which means I'm gonna ask you a few questions and you can only answer these questions. I don't wanna know about other things. I'm gonna ask the same thing to 500 other people. It's actually if you think about it, we are for limiting the amount of data that's coming in so that we can process it and we can make actions on it. Two parts, right? Process it and then and if you tell me a whole bunch of other random things, I I don't care about that. Right? So that's the original construct of a survey, right? For you know, if you think of fit for purpose, when you're trying to explore, you don't want to stop all the signals, right? So so running a survey when you're do when you're in exploratory mode makes no sense. But running a survey where you just wanna know look, I am gonna make one of these three decisions. Even if you said anything you want, I can only make these three things. So I don't want you to talk about a whole bunch of other random things. Right? So we're doing this podcast, for example, right? You wanna know like, hey, look, is the you know if I if you ask me for feedback on this, there are only two or three things you are gonna do. You're not gonna if even if I gave you random feedback, you're like, I'm not gonna do any of those things. I'm set on my ways. Yeah. So you just ask me, hey, look, you know, are you good or bad? That's it. I don't care. So you the surveys are very good for those kind of purposes where you are trying to measure a system. You're not trying to explore. If you're and that's the you know generally the the the spread between qual and quant, right? So when you come out to you know you know you're going back to your point about like, hey, why do we have you know forty, fifty question surveys? Well, they've taken a different use case and plugged it into that because they thought they could do it. Okay? So now with LLMs, you don't need to do a fifty question survey. What you could do is an interview five thousand times. See the point? You could not do an interview five thousand times two years ago, let's put it that way. Right? If I to if I gave you a stack of, you know, hundred thousand interviews, like actual real interviews, one-on-one interviews, ⁓ you know, a hundred thousand, you know, ⁓ Podcasts, you'd be I can't do shit with it. It's too much, you know, this is nothing I don't have a system to process it. Yeah. But today you have one. So if you're doing exploratory kind of data, so you could actually process it. So the surveys, so long surveys were like, okay, we that's why they were created. So now my opinion is that all the l all the lengthy questionnaires, the use cases for lengthy questionnaires are just gonna get converted to lengthy interviews. Because they're already lengthy anyway to begin with, right? I don't know. speaker-0: No wonder why I'm seeing a lot of qualitative ⁓ approaches being taken versus the quantitative. speaker-1: So the qual yeah, the the the standard or the standard conversations are on qual at scale and all that, right? So it's effectively means if you're gonna do a 30-minute survey, you might as well do a 30-minute interview. God damn it. Why? Why are you structuring a 30-minute survey? You can if you're gonna spend 30 minutes trying to probe, you might as well do a 30-minute interview. And then and then process the interview using an LM. And honestly, you could you know you can come up with pretty good signals, pretty good data out of it. speaker-0: Yeah. Correct. And we'll come back to that chatbot while I think. But ⁓ going back to that ⁓ creativity and analytical mind, I mean considering it that you've been an engineer by profession. ⁓ you've created the entire question pro as an ⁓ entrepreneur as well. Yes. ⁓ what we have seen a lot many times when we start an organization, we really need to be analytical, plus we really need to be creative as well. Which one comes first? Question one. Question two, ⁓ As an entrepreneur, should we really focus on just analysis or should we actually see, you know, considering the current market conditions, how can we be on those top I mean, to give you a context, ⁓ I've been building this podcast for almost like three years and everybody tells me that, you know, the first three seconds matter. You cannot be analytical in that case because now the analysis is saying that, you know, you have to be three, now you really have to be critical or you have to be creative. Yep, yep. I'm asking a question to the Vivek Phaskaran who created the question pro whether a analytical mind should come first or a creative mind, considering the current situation, not you know, going back. speaker-1: I'd probably say I'm a ⁓ creative first. speaker-0: And why? speaker-1: For the exact same reason. You n you need to catch you need to catch people's attention. Right? You need to in any any equation, right? The first three seconds matter the f you know, you know, how do you how do you present yourself the first time, you know? ⁓ because we're all inundated with so much of information. We have to sort it out. Right? Think about it, right? In the first five seconds, first, you know, your LinkedIn profile, somebody's gonna look at it. If you if you ping me on LinkedIn, I'm gonna take exactly two seconds to look at your LinkedIn profile and make it spot decision about whether I'm gonna respond or what am I gonna do. And it's true for everybody, frankly, right? And so to me, like, at least given cut the current situation, I'd say like, you know, being creative about how to get connected, how to think, ⁓ is probably more important than being kind of like, you know, supremely analytical, especially if you're in a startup, right? Like because you are trying to get people's attention. If you're a startup, you're trying to build your brand, you're trying to get people's attention. And ⁓ I believe this. I believe you know y y you have to do something that other people have not done. If you're trying to get people's attention, fundamentally you have to do things that people have not done. ⁓ That's kinda that that's the definition of catching people's speaker-0: Generalize this, we we're in an AI world. To catch people's attention, you can just give a prompt to chat GPT and it will give you. Yeah. But that doesn't work. speaker-1: You can It may or may not work. It depends upon how creative the output is, how your prompt is. You know, if you say like, hey, you know, how can we come up with a new idea? It's actually, you know, it's actually, I would actually argue LLMs are pretty fucking creative. You know, it all depends upon what temperature setting you run and what the prompt you run, and say, I want two ideas. Right? I want two I I want two ideas about how to market my podcast. What should I do? You'll come up with two ideas. You'll ⁓ it's not that creative, it's not that creative. Yes, no. But you can change the temperature setting. You can do a bunch of things. You can improve the prompt by saying, Hey, I wanna do this, I wanna do that. And you can be creative. And it'll it'll be creative, actually. The system will be creative. And any of the AI tools are gonna be. It all depends upon, you know, the temperature setting. You can say, like, hey, do you wanna be creative or do you wanna be an article? You can do you can do both. speaker-0: I'll take you back again to the same question. ⁓ not the same question, but another story behind this question. So there are people who really believe that, you know, the data gives a lot of signal and then you when you train the model, it would give you the right answer as to what creativity should go out or what creative post should go out. Now ⁓ you really need to collect that data. So if I'm going back to an earlier question about being marked product market fit, I really need to know whether this product Would be appreciated by potential customers or not. And I am not doing a cold call, but I'm still doing a cold email and I'm just writing a message onto LinkedIn and just sending it to them that you know, hey, this is my new product and this is solving this problem. Are you interested? That's it. It's not helping me actually get that ⁓ yes as the next reply in the same chat bot, but people are still saying that you know you should have actually, rather than doing this, first speaker. Talk to ⁓ say like ten, fifteen people. And that exactly the bigger problem for a lot of ⁓ SaaS companies these days that you know, even if what they are creating, they really need somebody to back them up first in in the whole network ecosystem. That if I know Vivek, Vivek, can you connect me with this guy who can you know buy my product and I can at least get 10 minute window where I can tell him. So that that creativity w is already parallel working or probably you know being executed but not working, but this this thing is working. Mm. So in in this current scenario, where does creativity actually or probably does it stick there or not? I mean the point is that, you know, y or you or or or probably you would be saying that, you know, do everything that you can do within your pace or, you know, jitna bitung kar sakteo kind of that, you know, whatever you can do by yourself, you know, do everything that can help you turn this game out. If you speaker-1: You are not I mean what LLMs are giving you today is like if you yourself are n are running out of ideas or not creative enough, you can ask LLMs to be creative for you. Right? It's pretty easy actually. Right? If you give it the right prompt, you will get and to different LLMs, you'll get different answers. So if you say like, hey, how do I ⁓ how do I do this? I've I've come up with an idea for me marketing this. Yeah, I mean LLMs will come up with like, you know, ten ideas. You may not like all of them, you may not like like that's a secondary issue. But it will come up with a bunch of ideas. speaker-0: But my point is that, you know, even if I get all those ideas and stri try you know, let's I I did ten I got ten ideas from Chat GPT. I didn't like e any of them, but the eleventh idea sounded like, you know, I should do something. I did that, but it didn't work on on getting the right client. But then I spoke to Vivek and Vivek connected me to someone else. That worked. So mm the business idea that now I have is to con contact people like Vivek rather than going there. Yep. speaker-1: Yep, yep. I mean, I think this is a function of more the inundation of slop, really, right? So everybody's gonna I mean there are two different problems here. One is how do you get through the noise to somebody? As an entrepreneur, you want to if let's say you want to get through the noise, right? And you're being creative to to cut through the noise, right? And I would argue that you know, you know, you you can either use LLMs to be creative and then cut through the noise. Right. If they don't work, they don't work. That's it. ⁓ and if you just say like, hey, look, I can connect with people, I mean, human connection helps you cut through the noise very easily. Because if I introduce you to somebody, they'll have to, you know, it's it's a it it's a it's a vote of confidence, it's a vote of trust. Trust, really. Yeah, you're right. And so so that helps. You know, as entrepreneurs, like, you know, okay, but if you're starting off and you don't have that many connections, you don't have that many connections, right? The one thing I will tell you, like I never bothered earlier on in my career, I never bothered kind of building connections. The first five years, I didn't even give a shit, honestly. I literally do not give a shit. Because we were getting so much of inbound leads, right? It was a system, right? We were getting inbound leads, we don't care about it. Right. It's only towards the latter part of my career where I started building connections. as frankly as we w went up market ⁓ in the enterprise. ⁓ so the one thing I would say is ⁓ you know if I were to rethink what I was doing earlier on, I should have invested more of my energy and time building building human connections, right? Because they they always almost pan out. speaker-0: Well you're you're an engineer, you're a technical guy, you knew that you know, using technology can help you scale rather than building connections. So I think the first five years may not be about building connection, but you used what if somebody like me who doesn't even have a any technical bent of mind, I don't know how to do SEO and now I think SEO has already died, it's all about AEO or GEO. Right. And I really wanted to, you know, rank one. What should I do? I am only relying on, you know, knowing Vivek and then getting that connection. speaker-1: Or you have a have a technical co founder maybe. That's why I know that that's the reason why a lot of you know a lot of times almost all the all the all the, you know, venture studios want you know, want a technical co founder and a business, you know, a marketing guy and a technical guy, you know. So they get both sides really. Right. So you have a you have a technical co founder that solves some of these analytical problems, and then you have, you know, a red brained guy on the other side who's like who's who's who's more talking to everybody. speaker-0: Are we saying that, you know, analysis and creativity goes hand in hand? speaker-1: Yeah, yeah. Yeah, it's been it's been true for many, many years. And that's why the the co-founder model works really, right? I mean even for me, when I started Question Pro, I actually, you know, I had a co-founder. I bought his I I bought his stake out ⁓ back in ⁓ back in 2005, Kevin and I started started Question Pro together. And he was he was you know ⁓ he was with me and you know I managed the the code base and he was the the sales and marketing guy. And then at some point I started doing sales too. ⁓ that was a shift. It was a it was a four shift for me. It was like I I had not done sales before and speaker-0: And when did you start doing sales? speaker-1: ⁓ it was like a matter of like I think two thousand seven, two si six, seven, I like, Shit, I gotta do it 'cause I gotta close the deal. So and I've never done it before. So picked up the phone and said, Hey, are you gonna pay me? Really, right? I'll tell you a story, right? So I was a very typical like engineer kind of like mindset, right? Like look, my job is to build some software, I'll just build some software and move on with my life. And then I was talking to a customer of ours, I think it was US Bank, and and he was some S V P US Bank and he's like, You he's like, Dude, you're you're pretty smart, you're solving this problem for me, right? And we were brainstorming on the call. And that's when I realized like, actually I know more about this shit than this guy. And he's the SVP of marketing for US Bank. So I'm like, I can do it. So that gave me a lot of confidence, right? Personally. Personally, it gave me a lot of confidence. Like, look, a lot of confidence and all these things are in our own heads. Yes. Right? It's it's the one part of our head saying, True, bro, don't do this or do this, this, that, that, all the stuff, right? It's not like somebody else on the outside is telling me I'm I'm I'm good, bad, or dumb, or yes, no. So I really believe there's a lot of people like constrain themselves in their own head. And if you can unlock that, then you don't give a shit, right? Then you're like you you can unlock a lot of things. speaker-0: May I tell you something? I mean, ⁓ that was a US bank. speaker-1: Yeah, that was my customer at Usbank, way back when that was in eight, nine. Yeah. All right. speaker-0: Six seven. ⁓ fast forward twenty thirteen. ⁓ my first sales ⁓ con conversation or the pitch was with ICICA Bank. speaker-1: And speaker-0: With seven, eight years of me, you know, building panels, understanding panels, executions, project management, everything was the f the day when I I poured out everything to him that, you know, this is how we can solve your per problem. So I was not technical, but this is again, you know, very technical, right? Operational. I get it. And I was like, dude, I can actually sell. Yeah. I mean, I this is what I I know, and I'm answering that question to that say ⁓ that ⁓ research guy of ICIC Bank, I can I can definitely sell. And ever since then I've never looked back. I mean, considering that those seven years stayed with me now, things have changed because, you know, ⁓ I think a lot has changed over a period of time. We're talking about synthetic data, we're talking about digital twins. ⁓ for me, when we are building Accurate SI, we know that, you know, these are basics that you definitely need. Now the entire purview of creating the Accurate SI logo or the brand was was based on the panel data only, or the based on the data that we have con ⁓ collected from humans. So ⁓ I think now it is not that challenging because over a period of time I had learned basic tricks of talking to someone and telling them what what works out. So for me, of course, you know, analysis definitely works, but at the same time creativity also goes hand in hand. But my question to you that, you know, you're sitting in ⁓ you're you're based out of US and you know, things happen very early in the US than it comes to a any other country. How's the synthetic data market there? speaker-1: It's up for debate, honestly. Right? I mean, I don't think, you know, I don't I have not seen people, you know, there's a lot of experimentation, a lot of hype, right? I don't see people pouring money into synthetic data yet. Okay? So everybody's it's it's it's almost like a wait and watch game. So how people are using digital twins and synthetic data with their own data, with other people's data, with panel data, with community data. So far what we've seen at least is that You know, there's a lot of interest and a lot of experimentation, but as a you know, frankly, as a vendor, not a lot of revenue. Right? So it's not like, you know, people are not interested in it, but I have not, at least I'll speak to myself, right? ⁓ I have not been able to convert a lot of that interest to revenue. speaker-0: Well, honestly, I don't really wanted to convert interest on on synthetic data. I mean, one for the fact that, you know, now I'm talking to you, I can ⁓ it's more of a qualitative interview for me right now asking questions. I know exactly, you Vivek is sitting right in front of me. Even if I create a ⁓ you know, ⁓ Vivek ⁓ digital twin, though I'm selling digital twin, but I'll still say you know, if I create a digital twin, I can only talk get to know Vivek about eighty percent or ninety percent. Not exactly that same. And that could not be generally The same. I mean, I think I'm also part of that debate when I'm actually making something like that. My whole point is that if if there there's still a wait and watch thing on synthetic data, how ⁓ and and and then you know the digital twin also came in in in the play as well, which is being backed by the data again. Where are we actually heading in terms of the research data collection world? Because a lot many people are still struggling right now and we see the market is not coping with that. speaker-1: I think like the way I think about it, like a lot of people say like this is gonna replace. ⁓ you know, my head of kind of innovation research, a guy called Chris Robson, he's a friend of mine, he ran a you know, micro research company in Seattle for, you know, twenty plus years. ⁓ I've been trying to recruit him for twenty years, I finally did it. ⁓ anyway, so he's he came up with this idea around like, look, I wish I was mine, but it's actually his. ⁓ you know, i it's he calls it, you know, three D research where we you have different modalities, right? You think about like the same argument we've been having Even pri priorities, you know, qual versus quant. All the qual guys think the quant guys are idiots, and all the quant guys think the qual guys are idiots, and you you're just arguing it for like, you 30 plus years. The synthetic again the same kind of model. So you almost kind of use synthetic for like exploratory, kind of like, you know, exploratory research, and then you can use qual and quant. So the way the way he at least, ⁓ the way we think about it now is that if you start saying that these are gonna replace each other, it's actually not true. It's like saying me coming to you and saying, Slack will kill email. You'll be like, bro, first of all, no. You know, instant messaging does not kill email. Instant messaging adds to the communication platform. Yes. Right? So if you say if I told you, you know, SMS, Slack, text messaging, WhatsApp, all of these things are gonna kill email. Well, it has taken a bite out of email, correct? But email's not dead either. Right? So what it has what is what is what is practically what happens? The total volume goes up. The total volume of messages have gone up by 100x. Right? Email still is going up. By the way, the number of emails being sent is still going up. But people have started using email and Slack and WhatsApp and Telegram. You know, whereas if you go back, you know, ten years ago, it was only email. Let's assume, right? Like you ten, fifteen years ago, it's only email. Right? So so it's i it's convenient to say, like, hey, look, you know, instant messaging will kill email. It's actually not true, first of all. We c we we've seen this. We've seen this literally the same playbook I've seen, right? When Slack came out, you know. speaker-0: Yeah. speaker-1: WhatsApp came out like, ⁓ well, we're never gonna use email again. Well actually, you know, obviously, you know, we're still using email, right? But we're but we are using Slack too, right? For internal comps and we are using, you know, WhatsApp for external commun communication. So so my hypothesis is that a very similar play play will apply, really, right? So it's not like, hey, synthetic data or you know, digital twins are gonna kill qual and quant. It's just gonna expand the expand the pie and say, like, hey, it's for a certain fit for purpose. And I always keep going to fit for purpose, certain use cases, actually synthetic data. You don't wanna do a full-lone research study. I wanna I just wanna know whether, you know, I wanna change the shade of gray from green to blue. It's not a huge, you know, it's not a huge impact. Okay, we can use synthetic for that. Well, I'm gonna come create a completely new product. Well, it's a big impact. It's a hundred million dollar investment. Okay, do a do a deeper research on that. You know, you'll do both a qual and a quant study for speaker-0: So basically you're saying that synthetic data can can be a complementary methodology. speaker-1: Yeah, it's it's it's one of the tools. Yeah. One of the one of the tools in your tool chest, really, right? So everybody's arguing that ⁓ yeah, it it's gotta replace this. No, it doesn't have to replace it. It can argue. It's just like you arguing that speaker-0: But when you say it's a wait and watch thing, do you still see that people are still using synthetic data by any speaker-1: Trying to find the use cases that it ⁓ applies to. That's the point. Right? So not all use cases are apply. So the same way we it's taken a long, w long time for all of us to understand which use cases we apply for qual and quant. Think about it, right? So the same way it'll take some time to figure like, okay, these use cases are great for synthetic, but these use cases are not. speaker-0: The way technology is moving right now, things are changing very drastically. I mean, for for the big data, I remember five years ago we were only talking about big data, and big data was a bigger hype than anything else. And now ⁓ within a year or two, ⁓ or probably like post-2022, you know, the the GPTs of the world have arrived, right? Or the LLMs are there. ⁓ the fast-moving data collection has taken its own shape, synthetic data has come up, and now digital twins and whatnot is already there. Again. Over a period of time, where are we actually heading to? I mean the what what would left for us to actually understand the humans better than speaker-1: These are all tools that go into the same understanding. I I don't think that that we are trying to find some nirvana. Like we we are trying to understand we're trying these are different tools that we'll use. All these tools are come, you know, along with that, some will some will be less in certain use cases, some you'll just move the speaker-0: I mean at the end of the day I think it the use cases will only matter. Not necessarily that you can use all the tools, but if you speaker-1: The argument is invalid, right? Like that that's why I give the email argument, right? Email is great for certain use cases, right? And WhatsApp is great for other use cases. And Slack is great for the third type of use cases. If I'm doing internal communication, I don't s I don't WhatsApp my employees. I always Slack them because that's an internal tool, right? I WhatsApp with my friends and I email, you know, if I'm sending out an RFP and this and that, I email business to business comm is on email. So there are different use cases. All three are communication. There are different use cases and you know, and each of them are equally valid in that in that use case. speaker-0: I I did ask this question earlier between ⁓ your creativity and your analytical mind. I actually had this one specific question again, the builder as a CEO mindset. You've always been an engineer and you've you've architected code entirely. And CEO responsibility comes with a lot of challenge. speaker-1: Okay. speaker-0: So after building I a question pro for so long and managing, you know and I I saw the recent post of your when you were in Bangladesh, you were actually talking to the tech guys and you're also meeting the tech guys in in the coming weeks. How how are as a CEO you're you're more technically inclined towards your tech mindset than a CO mindset? speaker-1: What is a zero mindset? I don't know. No, I mean I'm I'm just running the business. That's it, right? So ⁓ no, I love doing what I do. Simple. Right. I I enjoy creating, I enjoy building. ⁓ so I don't I I I think ⁓ there's no reason to believe that, you know, a technically minded person cannot run a business. No. speaker-0: My question is that you're you're actually doing both. But then there are different challenges. speaker-1: Yeah, I've I'm not sure I'm doing the other part. I don't know what it you know, so ⁓ no, I li the point I'm trying to make is that we've grown and organically grown into where we are. Yeah. Right. ⁓ and the one thing that I've always kind of challenged everybody on, like I don't just take ⁓ standardized rules and follow them. I create my own rules. Right? I mean I you know, I I've never If I did that I wouldn't be a follower speaker-0: Do you let people create their own rules or do you let them follow your rules? speaker-1: ⁓ good point. ⁓ I have a set of rules that I don't break and I don't let anybody change that. And and I we have a you know, a graded model where you say like we have certain principles that we don't compromise on. But then there are a bunch of other things. You can do whatever you want. and and to your point about like, you know, you know, the the technology and the technical aspects versus kind of the operation and the management aspects. ⁓ I have an amazing team. I love working with the team, right? My Scoro is my COO and artists, you know, rare on sales for North America, Asimbrant sales for Europe. ⁓ so we have a a a an amazing team that I love working with them. I'll tell you simple. Right? I like working with them. Hopefully they like working with me too, but you should ask them. ⁓ and that's been my kind of like motivation all these years. Like I I just want to have fun and work and make money and that's it. That that's about it. speaker-0: I think your goals are very set. So you don't even worry about whether you're technically minded or you're COA minded. Because at the end of the day, I I believe you CEO comes with its own challenge. I mean, when you're running a business, ⁓ the core reality of your business is that it is it is a code that you keep on you know building better and better and better. So I think your engineering degree stays there, but now you become you you you're a businessman, you you really also have to look at you know how you can make money. So you're more inclined towards that and then then balancing both the act together, it takes a lot of I guess time and courage. Yeah, yeah. speaker-1: It's fifty fifty. You know, I w one day I do this, one day I do the other one. Probably I drive probably a s a few people crazy. It's life, really, right? ⁓ but ⁓ you know, I I I I I don't even think about it in those terms, to be honest. Right? I just think about look, I you know and on the technology side I would even argue I'm not I'm not that super plugged in. I'm you know, we have a we have a team that manages the code base, I'm not in it. Although but two months ago I started I went back to writing code. About two month in all of May and June, I was like at lock myself in a room. Just to get into AI coding, ⁓ and went back. ⁓ it was super fun. I gotta tell you, it was super, super, super fun. ⁓ because like, you know, programming has completely changed in the last ⁓ you know, between in the last, you know, one year really. ⁓ just I wanted to get into it. Honestly, I wanted to get into it. ⁓ going back to your point about there are certain things that quite frankly we can do ⁓ that if I was funded or if I had then I could not do. Like there are a lot of th that it gives me a lot of flexibility. Like these are things like, you know, I doubt a board would agree, like, okay, look, you know what, the CEO takes off for a month, you know, locks himself in a room so that he can do AI coding. ⁓ I doubt it, but maybe they're they're they're good boards and they're you know, you know, but ⁓ so that I'm I'm blessed and I'm happy that way. So I'm like, okay, I can do what I want to do. And it also gives me kind of going back to the creative aspect, it gives me creative ⁓ latitude also. Like there are lots of lots of things that We've done. I'll give you a good example of creativity. One of the my best marketing hacks I've done is literally the best marketing hat I've done. ⁓ so we run conferences. So we have a conference throughout you know every speaker-0: Let me tell you that Bajaj Auto speaker-1: Sir. Sir. I know that. I know that. It's it's it's only ⁓ I think it took me, it cost me $2,000 to buy it. I think the rate was 75 rupees or something like that. It cost me another two thousand dollars to ship it. So I've had more juice out of that four thousand dollar investment than it's probably than any billboard that I could have taken off, really, right? And it, you know, fits the brand and everything else. So it's it's it's a fun exercise. So it's like, you And partly I'm also very competitive and create in in the sense I s a bunch of guys said like, No, it cannot be done. So the minute somebody tells me it cannot be fucking done, I wanna do it. speaker-0: I agree a w on the on that point. I mean, you know, when you actually get a board on your head then you really have to be more like a puppet than just speaker-1: It's more sterile. Yeah. I'd say like a little bit more sterile, honestly, right? Like because you're all everybody's worried. Everybody's like everybody wants to kinda do the least common denominator, right? speaker-0: You're still making money without even their indulgence or their their involvement or they're not even there. So you know exactly how you really wanted to do. One quick question that I really wanted to understand from your standpoint, you know, you've s you started a company at a very early stage and now you're you're already, you know, you're already there where people actually know you. So you've you made a lot of changes in your own ⁓ you know, life with respect to moving into doing sales and then later earlier this year you started writing codes as well. What is that specific advice you would probably give to, you know, aspiring founders who are getting into this tech business? Because you know, it is like it's very shaky right now. I mean, what I see from outside, I don't know exactly what is happening because when I see a lot of layoffs been happening. I remember twin 2008 when there was a financial crisis, a lot of people actually started opening up their own companies. And now they're they're where they really wanted to be. With this financial this is not a financial crisis either. I mean, it is still that the economy is downsizing at some point of time. What is that you really wanted to give that advice? Because n you no nobody actually trained you to become a CEO and you you became what you really are right now. Correct. So what is that advice you would give to aspiring CEOs or founders in in in the tech industry? speaker-1: I mean say I I am like a huge fan of founders, right? I mean I I support founders. Even people, you know, even in the company who have left Question Pro to start companies, I've invested in them, I've supported them, really, right? Fund fundamentally I believe like the act of creation is a very noble act. It's a good act. Like, you know, so I'll support it one way or the other. So going back to your question around what you know, what advice. I mean like look, if you believe in something, you just have to grind and do it, right? And and don't take no for an answer. Don't take no that you cannot sell. Don't take no that it's in all these no's are what stop us, really. Right? ⁓ if you want to do something, you have to do it. And you find a way to do it. That's why I kinda and that's why I was talking about the rigshaw. Like you have to do it. Like you know, we it almost like almost like a challenge. Right? How are you gonna get this thing done? Right? And you're gonna get, yes, as a founder, you're gonna get punched in the face. Hundred percent. There's not a single founder who's not been punched in the truth. Right? You put all your effort into something, everybody says yes, last minute, f off. Yeah. Right? Yeah. Last minute. And it happens. You have to deal with it. You have to have a thick skin. First of all, you have to have an extremely thick skin. Right? And it's emotional because you've put your heart and soul into some project and then it just dies. Right? So I mean having a thick skin, it's easy to say, but you know, that's what I would say right now. I wish I had big thicker skin. I used to get, you know, depressed and happy and depressed and happy and depressed and happy. Now I've kind of like leveled up a little bit, right? Because I know like these things are powerful, the course. Right? Somebody's gonna come and punch you in the face. It's it's it's gonna happen, right? And so as a founder, like the only thing I can I can tell you is like, you know, get ready for the ride, right? And have a thick skin. And that's it. There's, you know, and then and keep grinding, right? And you have to grind. If you can't grind, then if you you know, most I think most of the people I've seen that who are who walked away, you know, wish that they had not walked away. Right? ⁓ you know, you gotta you gotta you gotta push it a little bit harder. And those are I mean, I know they're not specific answers, but you know, it that that that's what's worked for me at least. Like I I had no option. I was like, I just gotta do it. Yeah. And then and it worked. speaker-0: And one last question I really wanted to un again understand from somebody who has flown all the way from the US and being attending a lot of conferences and knowing exactly where everything is heading to. ⁓ I know, you know, ⁓ the people that I speak with mostly are brands or agency and they're part of some or other forum, whether it's on WhatsApp, on any other social media ⁓ site or probably conferences, we see a there's always a shift in ⁓ understanding consumer in in one way or the other. And brands focus entirely on using different methodology and we spoke about almost everything. ⁓ what do you think in next five years brands should specifically focus on a particular big shift that can change the entire course of market research? speaker-1: I think one of the super interesting things I've heard recently is ⁓ agentic shopping. So people have agents shop for them. All right. It's a pretty n you know I don't know whether it's gonna fly or not. But if that takes off, now you have to create a new environment altogether. It's not even like the idea behind shopping was like purchasing. Yeah. It was like I you know, I'm trying to figure out what you wanna buy realistic, if I was trying to sell you some you know, if you're a consumer, right? But if you've outsourced that job to an agent which is gonna go look for the best shirt you want, knows everything about you, and agent is looking around for a shirt for you, it creates an entirely new ecosystem. All right? Because I don't care. Now I have to optimize for the fucking agent, not for you, because you outsource your shopping to your agent. Right? So it's kind of a completely different kind of game plan. So if you ask me, well, like if five years down the road, How should if I am Nike, I'm trying to sell to you. speaker-0: But aren't we there already? I mean, considering that we give ⁓ instructions to Alexa that, you know, create my shopping list. And then we put everything into that. speaker-1: Somewhere that it's not it's not operationalized at a s at scale. And which is what like Shopify, you know, Toby, the Shopify CEO is working on something like that. ⁓ where they they re everybody realize, you know, Yogesh, ⁓ he talks a lot about it, ⁓ Yogesh Chava, like agent take shopping. Right? So if you are Nike, like now you're like, Okay, do I go after you know, do I go after humans or do I go because humans have outsourced the shopping to an agent. Now it's not fully hundred percent obviously this is like speaker-0: Again, there's a lot of questions. speaker-1: Yes. Yeah, it's you asked me for a five year plan, right? Like okay. My opinion is this this thing, if that does happen, it's pretty crazy, right? Pretty pretty out there where like speaker-0: So something that I'm taking from this conversation is that, you know, in case it happens, people will I mean the brand will more focus on the agents than the consumer. speaker-1: In case of Absolutely. Because the agent now now now you now you have to figure out how the agent is thinking, right? Because the agent's gonna buy your shirt for you and just ship it straight up. Like and you just say like, okay, look, the agent knows everything about me. Yeah. And like I'm like, you know, this this shirt doesn't look good on you. speaker-0: Then the whole ⁓ concept of keeping the shelf eye brands would go fade away entirely because now I have the entire grocery list embedded into my Alexa and every ⁓ month that grocery is actually getting speaker-1: Now you have to optimize for a completely different use case, right? Yeah. If you're a s if you are in the other side of this equation, you're ⁓ you're you're so far optimizing for the person, now you have to optimize for the agent. speaker-0: I mean I think my the the way I'm ending this is more on ⁓ a different, refreshing perspective altogether. I mean, this is something I've never thought of. Agentic I mean I I do follow Yogesh, but I never actually probably would have missed this part. But now I'm I'm more going back and looking into this agentic shopping thing. But this is this is ⁓ this is amazing. I mean thank you. speaker-1: Yeah, you know, Yogesh and I have talked a lot about this. Like he we that's why I bring it up. So Yor Yogesh is like in the forefront of this thing, and he and I have talked a lot about, you know, and there are lots of ways to kind of start thinking about it. And and there are companies who are actually investing into it. Because they know that they are, you know, the forward leading companies are thinking like, okay, today we are doing this, but you know, we need to start preparing laying the groundwork for if this happens, what do we do? Right? So we need to understand how agents are speaker-0: But but I'm sure Vivek, I mean, by the time this would land in India it would be something would have else. ⁓ speaker-1: I disagree with that, man. I disagree. I I think this idea, like there's a l the the speed at which things are moving, like you know, information architect like information is generally free, right? Yes, there is some lethargy around, you know, I see some level of kind of like, you know, spread. ⁓ but you know, even in Latin America, like I'm you we have offices in Latin America, and I they're they're all very interested. Like the yes, there is a gap, I would argue, but the gap is l vastly, it's it's reducing every day. So ⁓ like even South Africa, they're like super interested in. you know, these kind of kind of investments. So I I would say like, you know, yes, there are some of them are futuristic, but ⁓ but I would dis I would not kinda like, you know, I personally do not kind of believe that, you know, these things have to be slower or anything like that. It's like, you know, it's universal. speaker-0: I think the acceptance rate is still ⁓ way too low. I mean people are not even accept ac accepting synthetic data data or digital I mean I think the best part of for this conversation is that you specifically mentioned there are certain use cases on which people should think around it. Not necessarily synthetic data would work on every single case or digital twin would work in every single case. And that's exactly what we, you know, trying to build our company as well. That, you know, there's a certain use case on which you should work on. Entire methodology could be different. ⁓ or the way you're looking at consumer per s ⁓ you know no details are different. We we are just part of this group. I mean just like you know data collection can be just ten percent of the entire market research project but it has its own use case. So a lot to digest for a lot of people and eventually that the pace which is going right now is like hundred and twenty miles an hour, seems like that you know people are by the time they they are able to understand, you know, what digital to when is something is else is coming. speaker-1: But that's kinda like par for the course right now. Things are moving fast. It's like internet. When internet came out, like, okay, shit, nobody's gonna buy anything online. No. But eventually they did. But it did t it takes some time really, right? So the only I mean at the saving grace I would say is like, yes, things are moving fast, but it's okay even if you're slow, you know, if you came came to the party late. It's all right. You can you can get up to speed. Yeah. Really, right? It's like you can get up to speed about almost everything. You know, so there's lots of benefits also really, right? So it's like all right, you don't have to kinda like get depressed, ⁓ shit, shit's moved fast. Yeah, it's moving fast. I get it. Right? But again, I would also argue that like not you know, not everybody's figured everything out. Mm. Not everybody's yes, there's a lot of froth in the market and that's everybody like think about the internet, you know, back in two thousand, like ⁓ yeah, pets dot com and a web band would take over the world. Well neither did neither of them did. Yes. Right? So okay. So a lot of them came, a lot of them went. And so there's like a lot of you know speaker-0: There's always been a debate between everything. I mean, you know, when ⁓ face to face was there, online panel came and then there's so many things which are happening. But this is this is really, you know, an amazing conversation, an amazing story that I got to know from ⁓ Vivek all the way from the US he flown all the way from here for this podcast episode. And and glad that I am able to see through your eyes is as to where exactly, you know, the US is heading to versus the entire world. This is a refreshing perspective on where technology actually meets human understanding. speaker-1: Yeah exactly. speaker-0: I mean, I think the takeaway is very clear. Tools and models will evolve at lightning s lightning speed, but clarity of purpose and staying close to real customer problems, that's remained timeless and nothing's gonna change there. Thank you so much, Vivek, for this whole energy of yours. And I'm sure Jetaya did mention that, you know, you'll come with a lot of stuff that that I can look into it. And ⁓ you know, his engineering mindset, his vision towards the future of the research is like outstanding. To our listeners, if you enjoyed this conversation, make sure to subscribe and leave a review and share the episode with anyone building into tech insights or customer experience. Until next time, keep listening, keep questioning, and keep building what next. Thank you. Thank you, Vivi. speaker-1: Thank you, Voss. Good one.