Why Zepto's Aadit Palicha Turned Down Stanford to Deliver Groceries

If you remove all constraints and you just remove all the laws of physics and you just think from first principles, what's the most extreme positive customer experience you can give and you start from there and then you work backwards from how can I make that possible. So audit everyone in this room knows Zepto almost everyone uses the product many of them probably every day. I don't know how many people know the true story of how Zeppto got started. So I thought we'd start there. How did you end up starting this incredible company? Well, firstly, you know, Jared, thanks for having me and it's a privilege to be here at YC Startup School. I mean, you know, we started our journey, Keville and I watching Startup School videos in my bedroom and at home and so it's it's a wonderful privilege to be here. I think you know the the thought process of starting this company at the beginning was you know we didn't really start with the intention of building a company as you know right we were you know two kids how how old were you we started I think working on it when we were 17 right so uh we KB and I loved building when we were kids you know we've been friends for quite some time and you know we said wow around 15 16 we said you know people in Silicon Valley are doing this for a living building out cool products for a living. So why don't we explore doing that? And so you know when we were 17 we applied to study in California. We were lucky enough to get into a a good school there. And then co happened and once co happened we found ourselves back in Mumbai uh which is where we're originally from. And at that point you know we really had nothing to do because we were not in person in Silicon Valley which is where we wanted to be and so we said there's no point doing co college online learning. And so we took a year off from college and said, "Let's just work on a project." And that's how it started. And I think really the big problem at the time, first wave of the pandemic was people couldn't get groceries in their regular neighborhoods, right? Like the the small mom and pop stores had, you know, issues with labor. The big guys that were delivering next day, you know, were all backlogged. And so we just started a a WhatsApp group chat where we used to deliver for our neighbors in Mumbai. And that slowly kept evolving until we made it into an app. And the first rendition of the app was called Kiranakart. And then it kept going from there. And then that's when we were we stumbled onto YC. So, and I remember the first time I met you, Audit. This is a story that I don't think has been told very often. Um, you were just about to start as an undergraduate at Stanford. You'd gotten into Stanford, amazing school, and you were going to leave India, go to Palo Alto, and start at Stanford. This is just a couple months probably before you really um two three months before metriculation. Yeah. Like two or three months before you started the company that was then originally called Coranocart. And I remember you asked me basically should I go to Stanford or should I drop out and start this company maybe for anyone in this room who's maybe thinking through sort of a similar sort of question in their own minds. How did you think about it? How did you end up deciding to start Zeppto rather than going to Stanford? Yeah. That's true, Jared. I think and by the way, for everyone here, you know, Jared has been pretty instrumental in getting us off the ground. He was our group partner on YC and you know, he's been really the only thing I did was convinced him to not go to Stanford. Everything else he's done. No, no, he did a lot more than that. But, uh, but yeah, I mean, I think in the early days, you know, really a lot of people ask me this question that you dropped out of college, should I drop out of college and do the same thing? I mean, actually, I have a less heroic answer, which is that we we thought of it a little bit more tactfully, right? where we said okay obviously it's a risk because going to college is a good opportunity etc etc so I think we took a year and in that period of the year you know we did a lot of a lot of work to be able to get to some level of PMF I mean we got to maybe a very early sign of PMF I think 8 n months into the journey and that's when we started saying okay there's retention there's compounding in the business and only up until we got to the point where we started doing about 10,000 orders per day I remember that point but that's when we said okay this is a real business and this is a once in a lifetime opportunity and we should go for it, right? So we I think we took our time before we took the big leap, right? Um and so that's usually what I tell folks is that you know you should have some product market fit. You should have a real product that's working and you you shouldn't take the risk before you have the proof of concept of something that's actually working and to in our case we got to like a decent like 60 70 cr revenue run rate or GMV run scale before we said wow we should go for this business opportunity and also we were lucky enough to get you know investor interest at that point when we got to that scale so we said okay we've got a term sheet we've got a business that's working customers are liking it we should go for it one thing that's interesting to me about the Zeppto story that I think might lead to a good lesson for everyone in this room is when you started Coranocart originally grocery delivery was not like some new idea in India there were already what two or three billion dollar companies doing grocery delivery in India at huge scale what made you think so what I'm just going to do this anyway what made you feel like it was worth launching something that competed with them I feel like a lot of people in this room would look at a market like that and say like it's too late I missed it no I think it's fair so I think you know we started this company in 2020 or we started working on it in 2020 and at that point there were you know multiple different models that existed in this space of grocery delivery because grocery as an opportunity in India to digitize grocery to build like an organized player is a very large opportunity and people have been chasing it not just for 3 four years they've been chasing it for 20 years you know even in the offline world right and so there were some large players out there nobody really had the model that we were doing there were some players that were doing sort of you know doorstep delivery from you know other stores there were some players that were doing next day delivery to delivery but we didn't really think about it as oh we'll try a different model that was not really the thought process what gave us conviction that we should do it was just bottoms up like when we started that WhatsApp group in Anderi in Mumbai and most of the deliveries Keville and I used to do ourselves and in that process we would talk to a lot of customers at their doorstep organically we would give them a pack of groceries and say hey you have 2 minutes to chat and then we would talk to them and say okay why did you use us why don't you use ABC or why don't you just do it yourself and that's when we learned a lot that when you spoke to the end customer, they were not satisfied. There was no real compelling player or compelling model that really was hitting their needs including us at that time by the way. So that first rendition of Kirano card we had not nailed the product market fit and so customers had they said okay you don't have all the selection your pricing is too high the delivery times are not in our control at that point and so we would just speak to customers and say okay what do customers really care about at the end of the day they care about speed quality selection and price right and so how do we push the bar on all four of those things and that's when we started for the first time you know trying to do these mini warehouses or mini dark stores and that's when the customer experience significantly improved and that's when we saw product market fit. Yeah, let's let's talk about this because a lot of people probably don't know that Zeppto is really a pivot from what you started with which even had a different name. You even like changed the name of the company when you when when you changed the model. So all during YC you were working on this thing called Coronicart. Maybe maybe explain to people what the original idea was. For sure. So look, I think from our lens like I mentioned, we we never started with a thought process of building a company. We started from a thought process of just solving a problem for like the 30 aunties in our neighborhood, right? And that's why we started initially with the easiest solution was just a WhatsApp group chat where we used to deliver from existing stores, right? And so we would just go around the neighborhood, deliver to people's doors when they asked us for it and we would pick it up locally, right? But then eventually we made this meaningful pivot I think early 2021 where we said look if you're just doing doorstep delivery from like a mom and pop shop to the customer you've got no control on the customer experience no control on the delivery times you can't really and you can't solve the problems that people were telling us at their doorstep. So that's when you know KB and I basically you know picked up a a bunch of stock and we put it into his his uh apartment and then that was the first mini warehouse in a way. Oh, your co-founder's apartment was the first dark store. Yeah. Yeah, it was. It was. So, we started from there and but what we noticed is that okay, in this like small area where we could, you know, control the customer experience, the response was phenomenal, right? And that's when we took it to the next level where we said, okay, let's try to do like a mini cardboard version of a warehouse. And we did that in in Juu if you're familiar with that. And then finally, we actually launched a proper warehouse. And then we saw the the pivot point for us was that controlling the customer experience you know we thought okay you improve the delivery times incrementally you improve this the speed inc the quality incrementally the selection incrementally the pricing incrementally there'll be like a little bit of an improvement here or there but we saw 10x like the one neighborhood where we were doing a dark store versus all the other neighborhoods that we were doing just mom and pop shop delivery the one area where we were doing it from a dark store had three four times the volume of the rest of the city for us and so we said wow there's really like a lot of customer demand here and So that's when we made the pivot saying that we need to control the customer experience and so that's when we started going into these mini warehouses. Uh so that was a thought process and at that time the idea of doing these like mini warehouses and delivering to people's doorstep at least back then obviously now it's fairly common place but back then didn't really exist. So we didn't come in from a lens saying okay you know XYZ player is doing it like this I'm going to try doing it like this or this guy's going to do it like this and I'll do it like this and we'll see. It was more like okay customer is not satisfied and I would just we would just obsess over what can we do physically to get that customer satisfaction up and then that led us to this mini warehouse model which by the way when we started a lot of people said oh no this is not the right way to do it this not the right way to do it but the customer was telling us that yes this makes sense but all the pundits were not in a way when people were telling you that it was the wrong move to build your own dark stores what were their reasons a lot of people were building I mean in our business people were building the business initially backwards from a supply chain or from financial goals right which obviously you know makes sense to do and that's what we do dayto-day but really the starting point of any business has to be the customer so what we said is so what people were saying is that okay this is the model that's viable and this is what I'm going to give to the customer this is the best that I can do and this is what the customer is going to get we started from the opposite we said what is the most extreme that the customer wants and that's why this concept of 10-minute delivery right which is a crazy idea at the time. I mean 10 minutes still blows my mind like somehow I order something in Zeppto and like 10 minutes later it shows up and you know by the way the the thought process I remember K and I discuss this we had that similar discussion that you know Brian Chesky Airbnb uh concept where he said five star sevenstar I remember that so basically it was about where can you just if you remove all constraints and you just remove all the laws of physics and you just think from first principles what's the most extreme positive customer experience you can give like what's the most unbelievable possible at scale that is a better way from starting saying that what is possible for me that's the best I'm going to give the customer but what we realized is that if you nail the customer experience you get so many other advantages in the business that are impossible to forecast for example right because we nailed the customer experience of 10-minute delivery doing the right selection doing the right pricing and doing this mini warehouse model we got far more throughput and far more volume in our warehouses than people expected And when you get to such a high volume in the same mini warehouse, all your costs become much lower by nature, right? Without going into too many specifics. And so that was something that people didn't imagine because they didn't imagine that much demand, right? And so when people are pulling the product from your hand instead of you trying to shove it down their throat, right? Then magical things happen even on the business side. So at the end of the day, I think it's impossible to build a financially viable, profitable business without customer delight. Customer delight is where financial value starts. Starts with the customer. One of the things that I remember that stood out about you audit during YC was that you were constantly trying things and constantly experimenting. Like this is the first time, this is the first time that I heard the story that the first dark store was literally your co-founder's apartment. And I feel like that anecdote really encapsulates the way that you iterated on Zeppto, which is instead of having some like grand strategy from the beginning, you just like started with this WhatsApp group, you started like delivering groceries when things didn't go exactly according to plan, you like iterated on the like exact design of like how how you did it and you just kept like tweaking the model and going out there. You did a lot of deliveries yourself. You were super on the ground understanding what was happening. I guess my question is how did you know to build a startup in that way? Was it just instinctive to you that that was the right thing to do? No, I may give you more credit than is due. I think Jared is also skipping over all the failures, right? I think at some point in the batch, I remember all our group companies, we were like the worst or something. We were like not able to get any of the tasks done. I remember this very vividly. But uh but yeah, I mean I think the the you know we we didn't really have like you mentioned a grand insight or strategy in a way you know like Justin Khan says right the naivity of knowing how to build a company was a big advantage like we were extremely young and naive and the advantage of being young and naive is that you don't know how difficult it actually is right so and you don't know all the complex ways that people have tried and failed before the only thing that was real in front of us was us and like the neighborhood neighbor about down the road that wanted her groceries, right? And so when when you just cut out all the noise and in a way being in co locked in a room, it cut out all the noise of startups and VCs and this is the way to do it and you know read this blog post. It was just like this person just doesn't like us and we would just sit and talk between ourselves of why doesn't she order from us, why does he order from us, why can't we get everybody to order from us, right? And so it was that like singular focus on this like 30 40 customers at the very beginning that was super super advantageous, right? because it was just a no noise in our life and so if the only goal that we had was how do we get the 30 people around us to order from us and that was how all the questions started so it was nothing else about VC wants this funding and this guy wants this and you know this blog post told us to do that all was scrapped that that's super powerful I I I don't know if folks in this room know but uh one of the like iconic pieces of YC advice is instead of trying to get thousands of users just try to make a hundred users really love your product. Um, and it sounds like that's exactly how you built Zeppto is you just focused on like those 100 users and that's how you ended up knowing that Corona wasn't the right model and you have to go to Zeppto is because like those 100 users just weren't happy enough with Coronart. Yeah. Exactly. Ex. And so that was the thinking and in a way obviously you know YC also kept us accountable but I think the big thing like you mentioned is we were able to just create this insulated environment where there was only us and the customer and then there was nowhere to hide. If a customer was not happy there, it was right in front of our face, right? And so that was a big advantage. Most people's experience with Zeppto as a consumer is probably like like the app experience is like like a a digital product. I bet a lot of people here don't realize that you don't really see Zeppto fundamentally as like a soft as like a a consumer like app company. You see it as a logistics and supply chain and like traditional retail company. Could you talk about the um incredible supply chain and infrastructure that makes it possible for you to deliver uh one bottle of Coke to somebody in 10 minutes? No, absolutely. So, I mean that we we we say this to each other all the time, right? If you remove all the software and the tech and the dark stores, I'll use a bit of Hindi, but if you remove all the software and the tech and the dark stores, fundamentally we're an ATA dal travel company. You know, we sell Subji and Dud, right? We basically sell day-to-day groceries to the customer. And if you still talk to the customer and we still talk to the customer every day, the customer is like you're my arta where I get my art from, where I get my rice from. They don't think of us as some grand, you know, software company and technology company, right? And so that that clarity is helpful. Uh and that's what we focused on. So just to go into Jared's point, so there's there's a lot of depth beyond this just the front end of the application. Like obviously people aren't aware of the dark stores but even within the dark stores there are hundreds of minute design decisions from the way that you design the selection the way that you orient the store the infrastructure that you build there the way that you build replenishment algorithms the way that you build you know the trucking on the back end how you manage the tasks you know today thanks to Y cominator you know we employ north of 200,000 people right and so how to manage those people efficiently at scale including delivery partners pickers packers truck drivers. So there's a lot of you know efficiency that goes in there including the backend supply chain and you know we've talked about it publicly as well where you know we've built out industrial grade automation on the backend supply chain. So if we are getting you know products from a uni lever or a proctor and gamble or a Coca-Cola how can we move that through our supply chain with the least amount of people and keep shipping it to the dark stores as fast as possible right so it's at the end of the day a logistics company there's a lot of hardware robotics actually there's a decent part of our tech team now works on hardware right and some of the like you know the videos that we saw today and so there's a good amount of work that we're doing on the back end and the thought process there is also not just for the function of again right it starts from the customer why am I doing all of this, I'm doing all of the depth in the supply chain only because every rupee of cost you save across the supply chain by becoming more efficient is a rupee you've saved the customer or a rupee that you can invest in better selection, better delivery times etc. And the last thing that I'll mention also for us the biggest culmination of this is our fruits and vegetable supply chain. uh without going too many specifics but we run one of India's largest or you know at this point the largest fruits and vegetable supply chain uh in the country and we source you know millions of units per week from farmers across the country everybody from Mahabaleshwar for strawberries or let's say outskirts of Karnataka if you outskirts of Bangalore and cocot or beyond can you give people a sense of the scale that you're operating Zeppto I mean the numbers are just jaw-dropping I just learned today that audit at 23 years old is the largest seller of fruits and vegetables in all of India on the platform. on the platform. Yeah. No, so I mean we have like a a decent amount like Jad mentioned on the Zeppto platform. There's millions of units per day of fruits and vegetables that we sell and of course all sorts of other products. But yeah, I mean we're at a point where today we've got a customer base, you know, in the crows, right, of monthly transacting users. There are millions of people that order every day and we do millions of deliveries per day. Um and we've scaled to I mean you know publicly available information right we've scaled to you know tens of thousands of crores of topline or billions of dollars right and so from our lens uh obviously long way to go candidly I think that there is uh I mean we're in the grocery market and there's a huge amount of scope to run this up and there's a lot of work that we still want to do on squeezing out more and more efficiency from the supply chain squeezing out more efficiency on you know the consumer part of the business I think we discussed today as well. But we also have a meaningful advertising business on the app itself where the top brands in the country like Coca-Cola or Pepsi or Nestle bid on search keywords. So we're doing, you know, hundreds of millions of ARR in ad revenue today. So there's so many big opportunities in the business and right now we're growing quite fast and so we think that we still have a few more years of high growth and high amounts of operating leverage and efficiency that we can unlock and new big things like for example this ads business you know was very small 2 years ago or inconsequential and now it's become a very big part of our journey. So there's a long way to go but uh you know we we we're still in day one. Can you talk about the long-term vision of Zeppto? I think a lot of people maybe haven't really thought about what Zeppto could become in the future and I think it'd be really interesting to hear what you imagine it could be in 10 or 20 years. Yeah. So the way that we the way that we look at it is we want to build an urban grocery platform and we want to build urban grocery infrastructure for the country. If you look at it, every you know major country in the world has their own form of grocery infrastructure, right? Or you know logistics and supply chain and retail infrastructure for day-to-day goods, right? People in the US or people in Europe are buying much more sophisticatedly, right? Their costs are much lower. The sophistication they have on quality control is better. It's much more organized, right? We basically think of ourselves as an organizing force in the grocery supply chain in the country. Obviously we deliver a delightful experience to customers but the the big vision is I guess over the next four five years to be a large groceryer in the country and do that in the top 40 50 cities and if we can do that then I think we can build India's like homegrown version of Amazon or any other e-commerce platform right we want to build you know an e-commerce platform that's unique to India and if you look at the key nobody's really been able to do this hyper local e-commerce format had anywhere in the world, right? We think it works beautifully in India. We think it's the right model for India. Um, and so if we crack it, then we can build a very large outcome and hopefully also play a part in, you know, growing other businesses along with us. I think one of the other key things that I'll mention is that, you know, Zeppto and all these other quickcommerce platforms, not just Zeppto, have become a platform for a lot of startups to build their own consumer brands, right? And so, you know, eventually down the road, if we can create 10 more large outcomes in day-to-day consumer brands, maybe someone here will start a consumer brand. Uh, why not? You know, that happened in the US 40 years ago with a Walmart or a Costco and all these other players. We are still at the beginning of that journey. So, we want to be part of that journey. The last thing I'll mention, of course, is the people that we employ on the ground. I think if we keep scaling this business, we keep creating more employment on the ground and we think it's a wonderful thing. I I I'll thank you Jared here. I told Jared that you know everybody wants to meet Y Combinator because they think of YC as a very good positive force. A lot of people have told me that all these investors are creating lots of jobs in the country and we it's true you know. So so yeah we're we're here creating employment helping build a small part of India's infrastructure. So that's our thought process. Um so next I want to talk a bit about AI. So even though Zepto got started before the whole AI boom, you guys are now really deep into implementing AI at Zeppto and really on the cutting edge of like what companies here are are are doing. You you were telling me this morning about all the ways that you've transformed your internal operations with AI. Can you talk about like what's how how you guys are using AI, how it's changed things for you? Yeah, I mean I think so there are two parts of it. Obviously there's the customer side and there's a supply chain side. And so from our lens, there's a lot of work that we've done on this on the supply chain side. So for example, we we showed Jared today, one of our dark stores and we're basically walking through how it used to take a team of people, you know, days to manually forecast the entire supply chain. But now we've built out a machine learning algorithm that runs, you know, millions of units per day and doing forecasts without any humans in the loop. And what that's made is that that's made my supply chain more agile. I'm able to dispatch faster. I'm able to do more throughput in the same place because my forecast accuracy is so high and it's I'm able to forecast much faster. So it's purely as a function of a lot of work we've done on the machine learning side on the consumer side. One of the big benefits for us has been the ads business. So we've seen a big bump in ads over the past 2 years because a lot of products that we're building for our brands are AIEL. So let's say on the search platform for example there are lots of things where we're giving people the opportunity to generate keywords or we give them sort of we've built out a model that helps you accurately predict what keyword is going to give you the best return on ad spend. And so there's a lot of work that we've done there but we've seen huge results because once brands see that their marketing dollars are getting put to use in the most optimal way and they can see that objectively in their advertising efficiency spends like return on ad spend as we call it in the industry then it's been a huge bump in advertising revenue. So building out the search stack of advertising with Genaii tools has been massive. And the last thing that I'll tell you is internally internally we've been able to achieve a lot more with less headcounts like in our corporate office hopefully some of you you know have interacted with our team but you know in our corporate office we've been able to do a lot more with the same headcount. You know the engineering team that we've had I mean obviously everyone says this but I see this on the ground and it's still mind-blowing to me. The amount that we're able to build with such a smaller team is crazy. Especially like all the SAS bills that we used to have. I was saying Jared this, you know, I can say it publicly. We've cut almost all our software spends to zero. We've cut, you know, a lot of our managed services spends. We've been able to reduce a lot of inefficiencies and manual tasks in the system, right? That were happening with people. We've been able to automate it now. And so my overall organization, if you walk into the Zeppto office, there's so many more tools that are automated and there's no outsource tools at all. So we're saving like literally hundreds I mean huge amounts of costs on that on that side. So it's great for us. What is the software team like at Zeppto? And are you hiring? Yeah, we've got about 500 or so engineers and another 150 people on data science, analytics, uh product management, designers, etc. So, you know, we're very much hiring and, you know, actively in the market always for the best talent. So, if anyone here is is keen, let us know. You know, we're still very much at the beginning of the journey. We consider ourselves an early stage startup. So, we're not a big old slow company. So, if you're excited to come into a startup environment, please do join. Okay. And we're we're running out of time, but my my last question for you, Audit, is in order to run this company, you had to master an incredible number of skills. And you had to do it at a very young age. I mean, you started this thing at 17, and now you're running like a massive cold storage supply chain. You've got trucks and warehouses and physical sorting machines and 250,000 employees you have to manage. Um, how did you keep upgrading yourself as a person and a founder? keep learning these skills so that you could like do this incredible work at such a young age. Again, you know, very, you know, there's a that's a very generous way of looking at it. I tell the team this all the time that it's good that we're a Y Com because I've got exposure to what's happening in Silicon Valley and after everything we've done, we feel like peanuts compared to what people in the US have done sometimes, right? You know, like Uber or Door Dash or Airbnb. So, a long way to go still. But I think the the big thing for us has been that I mean both Kavali and I have just been incredibly lucky with the people that we've been able to surround ourselves with since the earliest stage and you know without exaggeration since the group partners that we had with you and Tim and and Seri at Y cominator um but every stage of the journey we've just been lucky to have incredible people that have been surround that have surrounded us that we've learned from and most of them most of that learning has come from our management team. So I think what the the luckiest that we've been is that we've been able to find a team whether it's our CFO, our chief operating officer you met today, our CTO, you know, our head of growth and marketing, our chief business officer, our, you know, leader in the product function. All of these people who've got decades of experience but are still hungry and believed in the vision very early on despite being much more senior in tenure than us. Having them join early in the journey and learning from them was the biggest one. So I think we woke up every morning and we said, you know, if I'm working shoulderto-shoulder with this exceptional leader in finance and this exceptional leader in operations, then I need to pull pull up my boots, right? And I have to learn from them. So you just have to be you just have to surround yourself with people that are smarter than you and just keep learning from them and keep asking and even today like with our finance team for example, I ask the most basic questions. How does accounting policy work or the ops team I ask them how does this hardware software work? You know, we also have looked very stupid a lot internally, but then that's been good because we shamelessly ask questions. We shamelessly learn and then you end up aggregating information from a lot of people and then you grow. So surround yourself with people that are smarter than you and learn from them shamelessly as as simple as that. Adit I and everyone in this room is in awe of what you've accomplished. Adapell to everyone.