 Good morning everyone. Good morning everybody and welcome to slash It's really hard to follow Stage about psychedelics. They're talking about data So I'm Jay does I am the president and founder and partner of sapphire ventures and We have been investing in data for a long time all the way from my sequel onwards And I wanted to you know have to the issues our CEO of touch spot joining me on this conversation Thank you Jay, and thank you everyone for joining us. My name is Sudeesh Nair. I'm the CEO of touch spot like Jay said Only in Finland. I think you start with magic mushrooms and mpna and then follow with the data analytics So I'm glad that we started with it. Now. Let's get into analytics So Sudeesh, you know, we have an investing in data for I don't know how long I've been invested in my sequel Altrax Dasper soft looker How come the data dilemma still hasn't been solved like why do we need touch spots though because we are not interested in solving We are interested in making money in data. I am partly joking, but partly serious in the sense that analytics Should go away. I mean I really mean that I am Speaking as a CEO of data analytics company Analytics has become too much about analytics and not enough about outcomes Like why are we doing analytics? I'll give you a quick example Like maybe 15 years ago or 10 years ago when you were taking a taxi You had no idea what time will you get there and how much it's going to cost, right? Are there anything more important any other KPIs more important for a taxi commuter than Estimated time of arrival and the money in that's going to cost you know Today we take it for granted that it happens. Why is that happening analytics? Yeah, right? It is real-time traffic. It is weather. It's the location. All of those things are coming together But what do you not think about? analytics In the world of business though, it's too far So there's too much money invested in doing analytics as opposed to focusing on outcomes And it's also all about dashboards, right? It is not about taking actions like what I see every time you talk to a Startup and he has to be data-driven, but then they don't know how do you use the data? And what does it mean to be data-driven, right? Look Jay, the dirty secret of data analytics is that Most people are afraid of numbers Yeah, because when you speak in numbers, there is no place to hide it is black and white Whereas words and opinions you can look smart by talking about adjectives and all of that stuff So what happens is a CFO who's going to a meeting don't want to be responsible for speaking numbers They would rather have the fpna team feed them a dashboard Because if something is wrong you can go and yell at your fpna team. That's how it works Yeah, so what happens is the status quo of analytics team getting paid to make analytics more complex works for them business leaders to make decisions based on gut instinct and Sort of anecdotes it works, you know, it works for them So somewhere in between the status quo is maintained I don't think that is going to be the future and in a generative AI and all the New-age workforce that's coming in will change it, but that's currently the status So what have you done differently, right? I know I've been on the journey with you like the second time You know with Nutanix and their thought spot. How do you use data to drive your business? Yeah, look, I think I I'm a believer that Experience matters. I believe that human experience our knowledge is extremely important It's like Jeff Bezos said if data and anecdotes are a conflicting I will bet on Experience and the anecdotes. However, it is important to make sure that you are able to test the thesis That you believe you know is right to against data Yeah, and what is happening is that the lack of expressibility standing in the way that's I speak business. I speak English But I don't speak let's say sequel or I don't speak data schema Yeah, so there is this this this chasm between people who speak data and people who speak business a thought spot What we believe is that if you can bridge it a forward-looking marketing specialist will be able to test the thesis of Money that I'm spending on a campaign. Is it really working or not and answer is never going to be from the first Question that you're asking it's almost always the 15th question and dashboards are like poor poor way of doing the next level of Questioning right that's why you know we fundamentally believes that dashboards should just go away and die No, I love it like because I think in the board meetings You almost use the hotspot as a way to describe the business to all the investors and you know as we are talking about what to do So maybe you can talk a little bit about hotspot like how are you changing analytics and why you know It's very different than what we have seen previously. So that the topic of today's session is Noise and signal right there is so much data Everyone is generating data. We are all sitting here and generating data and the data is captured and processed the question is how hard is it becoming to really make sense of it and Without talking about hotspot I can tell you the hotspot mission in the sense that there is a book that a person a professor in Toronto named Shakil Chaudhary. He wrote about a topic or deep diversity and It is important to bring the connection here because human brain Let's say you are crossing a busy street in a loud loud city like New York or London At any given point in time apparently 11 million bits of information is coming to your five sensory input that we have 11 million of those so how does the brain process 11 million the answer is it can't so what it does is it Quickly figures out how to chop and slice it into pieces and then focus on approximately 40 to 50 So the most important ones if there is a hot dog vendor smell coming to my nose But there's a yellow cab coming to hit me. Obviously. I'm focusing on the yellow cab the reason why this is important for Diversities that you know, that is why we are biased against people who don't look like us in the world of analytics the ability for people to communicate freely that is the ability to connect all the sensory input Process them in the brain and then act. That's what human machine is doing in the new era of generative AI We are doing this. We are improving those three things. So number one Computing power is exponentially increasing and number two of the energy, you know There is so much more energy creation that is happening and the most important thing that change with generative AI is that we are Fundamentally changing the input output. Can I express through natural language? Yeah, not just sequel sequel is so structured and so limited Can I use natural language to express my desires to a system? The system comes to me and says so this here is what you need to know Yeah, but here is what you asked for yeah, and can I start a journey of exploration with no tax on your curiosity? That is what thought sport is trying to do for business use cases Yeah, and you know it's always been the case that with analytics You only looked at the data that was there and tell you what happened, right? It was never the case where it would be predicting Okay, what is this data actually telling you because then you would go and kind of go to a data scientist or somebody else to do That right? Yeah, but with now with your platform I think it's kind of this whole evolution where you can ask the question Okay, what did the data say? Why is my sales going down over here? But then you can also say okay, what is my sales going to be in next year look? I think I'm very passionate about the idea of fundamentally making analytics invisible and outcomes visible and I want you know that I would like all of you to go sell and buy thought spot wherever you are But that's not the primary point if you are interested two of my colleagues are there Lena and Penelope they can help you what I believe passionately is that we need to move away from the idea of Asking questions that are pointed Every day you don't want cul-de-sac. You don't want like Alice that doesn't go anywhere. You want to start an explorative journey Every time you're asking a what happened question. What you're really asking for is why it happened Like how many customers churned how many customers left me today? Why are you asking that because I want to know what is causing them to turn? How can I make it better? Right? So going from what happened to why it happened to what will happen? And oh crap. I don't know what will happen. I don't like what is going to happen Let me change what will happen for the better So this transition the journey from what happened to why it happened to what will happen to what if I could change? cannot be be couple from the business now because Data people have the inside they have the left brain Business has the right brain. They have the knowledge if these two brains are Hemispheres are not talking to each other You can't go from what happened to why it happened to what will to what if and that is why? Generative AI is a fundamental game-changer because they all can speak that's right the language of English or German or you know finish And you can also analyze all the data in the way we actually Talk about things right like you can have you know come up with a dashboard if you want go with a pie chart But it can also give you back the results Yeah, in actual text that people can understand and not have to decide for the data that's being shown up I mean when I say you know dashboards should die I really mean that in a positive in the sense that like we were talking in the backstage there Sometimes in the world of business it is so hard to ask the right question like the data has it the like How many so there is a customer? Schneider electric I really like this story. It's like over 100, you know plus countries extremely large workforce imagine if a Country like Indonesia is not performing well Whatever let's say employee retention as an example Hypothetically speaking. Let's say they're not doing well in the boardroom. You make decisions at an aggregate level Oh, Indonesia is not performing well Brazil is let's just focus on Brazil However, Indonesia is not monolith there could be 100 teams and maybe there is this one hiring manager and She's doing a bang-up job. She's hiring well. She's recruiting well motivating well and guess what she's punished because Indonesia is not doing well decision-making almost always happens in aggregates And that's what is wrong with the old bi and analytics because ideally what you want is when you ask Why is Indonesia not performing well the system should answer and should say hey, did you know though? There is this young lady who's doing an amazing job of recruiting motivating and delivering value and that noise And signal ratio that is automatically amplified is only possible if machine learning works and We haven't had that in our system because like I said it works for the machine learning teams to go Crunch the numbers without ever knowing why am I doing it? I know Python, but I don't speak retention. That's right So a platform approach is required here I don't know if ThoughtSpot is going to be that company But I can guarantee you that the way we stop using ultra vista maps To moving Google Maps just like the way we go from yellow caps to Uber. This is a one-way street We will not be doing analytics the way we used to well. I hope really hope that thought spot is the one That does it for us Maybe talk a little bit about Genitive AI, right? I think thought spot was started kind of before Genitive AI really took hold But it had a lot of you know vision was the same thing that hey We want to make analytics more search based like what people are used to How has Genitive AI kind of changed the product and how you're looking at you know as you go in the next four Five years as thought spot evolves If you have young children you sort of know that zero to two It's kind of a difficult time because they don't speak You know they are almost always like angry or cranky and there's one of three things going on like hungry sleepy or want a poop Right, and you're just guessing But something magical happens after two for most children once they start speaking Their ability to learn just goes through the roof, right? From the very early civilization People learn through Speaking even before words written people huddle around the fire Elders will tell the stories don't make houses here because there is going to be flood coming they will draw pictures, right? We are who we are because we learn Constantly and we learn constantly because of our ability to speak naturally We're sitting here around you can watch this video on YouTube later, but this is special you are here You can see me you can see my pauses the way we are communicating with each other These things are adding color and texture to the conversation now think about how we communicate with the computer We computer with computer you don't speak English you speak code Forget about code the keyboard is so stupid that it is ASDF I mean just think about how archaic it is, but that we ever assumed like that's how it should be Just like ten years ago We used to think that the only way to get a cab would be to have a totally random Transaction to happen I come out stand in a random corner raise my hand a random taxi driver gets up in the morning drives around And these rule random events happen and somehow a commercial transaction happens and we used to take it for granted Yeah, that's how taxis are today. We don't we don't get out of here without knowing the rating of the What kind of car? Where is it? What time will I get there? How much it's going to cost? We take it for granted. I believe Generative AI forget the current noise. Okay, the the people and the drama and all of that This is all going to settle there is hype and all of that but at the core of it Generative AI is going to open up the bandwidth Where we will think that just like pointing and clicking as SDF thing is going to be archaic We can communicate with an intelligent system. Yeah, our needs aspirations likes dislikes Discomfort and it can come back to you And if we can enable that there will be magical use cases happening and just like that kid Who's growing up faster and learning after two we might be able to build amazing intelligent systems Yeah, I think like the way we interact with applications are going to change completely It's all going to be voice driven or text driven the way we talk to each other right applications should be like that right because intelligent system Why don't you tell me a little bit about this? I don't want to flip it on you But when you look at the noise with this AI plus AGI and numbers versus words Where is the early ecosystem going and what are you making the best? Yeah, so the early ecosystem definitely is all about enabling all our apps and all our products that we are invested in To have an ability to you know have voice enable or you know the kind of conversational enable right But very quickly what conversation kind of goes down to is actually searching for all the product, right? So conversation and search is going to go hand in hand because as soon as you start like you see that with your customers Right, it becomes a knowledge base And it's this intelligent system as you said and people start asking all kinds of questions to it Right because it's so a different way of Interface than when you type because of the type it's very static you can it's one way two way with a conversation You think it's another human being on the other side, so that's dramatically going to change, but I agree with you I think something Revolutionary like a noob or a door-dash build on the generative AI platform is still not out there That is an important point, you know I Thought support I challenge our company We are all were search, but let's just come out of analytics and think about search. Why are we searching? We are searching because we are already lost when you get up in the morning and go to school when you go to work You don't search the direction right because you know it If there is a traffic accident it should come back and say hey don't take the normal route take a different route In the world of analytics while the rest of the world is catching up to search Thought spot I want to move even farther because we have been on search and AI for the last few years as The world is catching up. I want to take it the company thought spot to the next frontier, which is why search If I have the data if I know your intent if I know what your business value priorities are Why can't I come to you and tell you things before you ask me? So we have built this product capability called monitor and our thesis is I want analytics to go away So a lot of you are startup founders. You are building other applications All of your applications have data and data requires analytics and I am pretty sure some of you probably are building analytics My my request is that don't do it focus on one thing that you do well and let us do analytics for you because our Thesis is thought spot should be invisibly accessible through all of your applications in a way That it tells the customer the user that you are going off course And when you are lost you can search so taking proactive monitor of your business KPIs in ways that are bespoke Where it isn't add more noise to your life and then delivering what to why to what if to what will All embedded through other applications where you're not thinking about thought spot just like in uber I don't think about Google Maps even though Google Maps is powering a lot of those in the back end That is the future that I wish to see analytics deliver So maybe talk a little bit about generative AI and the ethics and you know all the people are worried about What do you think of the regulations you think regulations will really stop innovations? Can it actually even regulate what the bad behavior that people are worried about with generative AI? Look, I think this is very early and I am definitely not the foremost expert to speak But my opinion is that in coming here in Europe I don't want to poo poo the idea that government should take a Active part in it because unlike other technologies Fundamental building block that you're building will require some level of ethics and guidelines Particularly when it comes to the weights and biases behind the data that is being trained on For example, just like that two-year-old kid that I talked about if you have a two-year-old kill as a parent You're huge responsibility to shape this little thing into a responsible human being and trust me You can shape that baby into anything that you want You can shape that baby into a loving boy or a loving girl or you can somebody that hates and pisses off everybody because you have that power Why because they're learning from the model that your parents are building at home The same thing applies to the formative days of model right now What we are seeing is very early. There are lots of data There are a lot of models and it is not a good thing that the largest models that the most high-performing models currently are Fully closed sourced when it comes to the weights and biases and the fine-tuning models behind it I don't think that's a fair thing to do I like the fact as much as I'm not a huge meta Facebook when it comes to the products I like what they've done with the llama where everything is available Yeah, will it be the Linux for operating system for this? I don't know but I do think that in this specific thing It is very critical because I mean look at this place, right? I mean I was asking the person was putting makeup on would you have the right shade for me because everyone is blonde here, right? And she said no I have it This is the story wherever you go. We are all somewhat like clickish So when the data that you are going to use impact the whole world It needs to be Intentionally built without biases and that's not something we are there yet. I actually think regulation is you know There has to be some kind of guideline regulation But I also think that that's an opportunity for entrepreneurs to figure out, but will it stifle innovation though? I I don't think so I don't think ever that regulations that I was stifled like you know Look GDPR was something that people are all talking about and saying hey It's going to cause lots of issues But in some ways that has actually not stifled in innovation and people have figured out ways around it to meaning provide Solution that enables you to do GDPR. So yeah, you know, it's always like regulations are never perfect But I think over time if you don't have any regulations You know as you said you're going to train these models to with all the data that you don't know where it's coming from And maybe you know, you don't like to have that kind of models, you know When you're using an application and this is an important point again within ThoughtSpot. We have taken a very Opinionated approach that we use large language models for intent capture only Because they're not good for numbers and analytics So what we have done is to take the intent and use our system to securely and Go on the data model that the customers are running in fact, so they have visibility into it But these things require time Look, there will be a lot of fair lot of noise lot of machine and I know that a lot of investors are jumping in and then learning the Hardware, but I don't know if the door dash and the ubers of this is even founded yet. Yeah, that's right Well, let's turn a little bit about founders you're need right because you build Nutanix 10 billion dollar company took them to billion dollar in sales you're doing that again with ThoughtSpot What are the learning there are lots of startup founders here? What what are the Key thing that you have learned through these actually you had a journey even before Nutanix right a startup journey What have made you so successful? I'll make it so instead of What made me successful is like making all the mistakes and then not making them again But if I were to look back and summarize them into three things number one is Focus It is so important that you do one thing really well and then get paid before you do the next thing Don't I mean founders specifically technical founders have this thing to just keep building and have no respect for sales or GTM Sales is one of the most noble profession there is you are actually solving for value for problems The customers are having which means founders should respect sales and do one thing and do it really well second thing Is don't listen to any of us who have done this before sitting on a stage like this and Pontificating because all of us have been lucky in somewhere other none of us are infallible You know all the way from Elon Musk to Sam alp man's of the world if you don't know them completely as a person This is all a show charade that that you're actually getting to see here So do not trust any of our advices as that something will work for you you will have to do your own thing So that's a number two. I'll become an icon of class in my own mind I don't trust any of these people if I don't know them that we both know each other for the last ten years I'll call him, but I will not listen to someone who's actually on Twitter and the last thing is That it is so important to remember how lucky we are and be grateful for everything that we have because I guarantee you Entrepreneurial journey. It is really shitty There will be days where you just want to say what the heck am I doing this for and why am I doing this to myself and If you want to have the ability to sort of keep moving forward you need to figure out how to be grateful and Trust me if you don't know what you should be grateful. It is because you're not looking this is special and That's my third thing be grateful I think also great founders are always self-aware because they you know no matter what they might Project outside they always know what their weaknesses are and they kind of go in and surround themselves with people That can fill their weaknesses right that that is like for me I think the best founders and the best CEOs are the best You know people who can hire and motivate and kind of get people to kind of do the best they can do I remember when I met Jay early on ten years ago. They were just founding. He's also a founder of the firm and It was my first entrepreneurial journey at Nutanix as well and it is the point that he just made is so important You know as founders sometimes it is important to remember that It is you know as more successful you become you'll be surrounded by people who wants to get something out of you and they Say yes to you and you can be in a bubble and it is nothing worse than being in a bubble So one of the things is make sure you have good family and good Friends who can always bust the bubble and say hey don't get too far out of your comfort zone What's the last thing you want to leave it with the start of founders here? Enjoy enjoy life This is the right there is no destination out there the journey itself is the destination enjoy this this is these are the good old days And I would say just go start companies out there That's the only way you're going to learn how to build and you know scale and create companies of consequence as we call them Talk to ThoughtSport get investment from Sapphire a great great firm. Thank you. Thank you everyone