Delivering Everything Right Now

Published Apr 7, 2022, 4:03 AM

Rafael Ilishayev is the co-founder of the instant delivery company Gopuff. His problem: How do you deliver everything from bananas to hot coffee in around 30 minutes -- and still make a profit?


On today's show, Jacob Goldstein surprises Rafael with a live Gopuff order. And they discuss the problems the company is working on in real time as they wait to see if the order will arrive on time and in good shape.


If you’d like to keep up with the most recent news from this and other Pushkin podcasts be sure to subscribe to our email list.

Learn more about your ad-choices at https://www.iheartpodcastnetwork.com

Pushkin. If you live in a city, you've probably seen ads for one of these new instant delivery companies. You download the app, go on your phone and order the kind of stuff you might buy at a drug store or seven to eleven, and then in like twenty minutes or a half hour, your stuff gets delivered to your door. I have one of those apps open on my phone in front of me right now. It's called gopuff. The apps called gopuff. It's one of the oldest and biggest of the instant delivery companies. And also right now I'm talking on my computer to the co founder and co CEO of gopuff. Can you hear me loud? Claire? Great, thanks for doing this. I've been reading about you and playing with your app. Can you just say your name so I have it and so I don't say it wrong. Yeah, of course, Raphael illashayev. As it happens, I live in Brooklyn and New York, and go Puff delivers to me, and so I thought a fun thing to do, if it's okay with you, would be that I order something now and you sort of talk me through what's happening as a way of understanding both your company, how it works now, and also some of the things you're trying to do. You know, It's sort of the thing of the show is the problems people are trying to solve, like I love it. I love it, problem too and I love it. I'm Jacob Goldstein and this is What's Your Problem, the show where entrepreneurs and engineers talk about how they're going to change the world once they solve a few problems. Raphael and as co founder started go Puff back in twenty thirteen. They were college students delivering munchies to their friends. Today you can go on the gopuff app and order most of the things you'd find at the drug store, snacks, diapers, advil, and the company is trying to figure out how to add more stuff like produce and hot food. Gopuff operates in hundreds of cities and they've expanded from the US to the UK and France. And the number of problems the company has to solve boggles the mind. Competition, technology, logistics, hiring, politics, it is all here and today on the show, we're going to do it in real time as I place in order to figure out how the business works and the problems they're trying to solve this is fun. By the way, I've never I've never done this a live order. I'm enjoying too. I'm enjoying too. So okay, so I got the app There it is, and it says on here fifteen to forty minutes. So let me see. So the first thing is like, what do I order? So look, I'm going to order. What I'm going to order oreos. To me, the spirit of gopuff is oreos. I assume you have you do have oreos, So it's seven to fifty. So seven fifty for a bag of oreos is like higher than the grocery store, right, I would say it's pretty pretty in line to like middle tier retail, closer to what you would experience in a drug store. And then looks like my minimum order is twelve ninety five. So let's see. I mean, I feel guilty about the oreos we always need, say bananas, yeah, out of stock. It looks like you have bananas, but they're out of stock. You have let me see, should try they're too popular? Let me just see? Am I just well? You have fugie apples? Okay, eighty nine cents for fogie apples, So let me just get some fujie apples here, maybe about of water. You're a gatory appreciate cardon milk, maybe smokely um Sky's hustling, always always be selling, always being um Okay, so let's see. It looks like there's a two ninety five delivery fee and then there's a tip which it's auto populating at three ninety nine. So that's fine. I'm gonna go ahead and place the order just so we can, you know, pass go here, place order. Let's do okay, my order is being placed in here just for fun. Started the Stopwatch on the cassio twelve. Let's do it two twelve. I love it. So I just well, first of all, what's happening now? I just pushed a button and something somewhere in the world just happened in what warehouse that you want right? Tell me what's happening right now? So there's a micro fulfillment center somewhere in your neighborhood. I don't know where you live. So in a neighborhood in Brooklyn, they received an order and they got routed to a packer. Okay, so like did a bell ring? Did somebody like get a buzz on their Yeah, like, what really happened? There's an actually a buzz. Yes, there's vibrates vibrates on their handheld device. So I'll take you a kind of step backwards. When we were building out this thing, and we're building out all the technology within the four walls of the building, and then later on, you know, the last mile, we realized that there was nothing out there that can solve this for us. We didn't want to build all this software to power all these buildings. But the reality is, because nothing like this existed, we had to build a whole bunch of proprietary software to then pick and pack and then ultimately been in batch so put it into a batch of orders and then eventually send it out to the end customer. And you have to do that constantly under thirty minutes, right, So I just, oh, yeah, under thirty minutes. And also I want to pause here just to make a distinction clear that might not be obvious to people listening, and that is this is not instacart. This is not You're not a company that has people picking things up from existing stores. Right. This is your MicroWarehouse and you own the oreos and the apples. I just bought right, you're paying for the warehouse, you're paying for the apples. This is the opposite of like an asset light business, right, you're buying all this stuff and reselling it. Just classic retail. Yeah. Yeah. The problem with a third party delivery model it's notoriously unprofitable and it's not consumer centric, right, So when you don't own the inventory, you can't really tell the customer, like the same way that you just saw the bananas were out of stock. If you're ordering from a third party app you know, twenty five percent at a time, you're ordering an item and then you get in a replacement item instead of the item that you've order, or to ask you for something else because there's no real inventory, so they don't know. I'm placing the order for a thing that's not there, and it can a company that doesn't know that it's not there. That's right, You know that because it's your banana. That's not that's it is it is our banana or as our milk. Wait, I'm gonna pause and say my order is on the way. We're running behind on the interview, it's going too fast. But tell me, what does this fulfillment center look like, yeah, it looks like somewhere in between of a retail location and you would expect an Amazon warehouse to look like. Think about you know, seven to twelve thousand square feet, so larger than what you would expect the drug store to look like, smaller than what you expect a grocery store to look like. But we can pick a lot more inventory inside of the building because there's no walk in traffic that's going through the aisles. The building is optimized for you know, rapid picking, because the key is that you want to pick it as fast as possible, throw it in the bin. Then you have an algorithm that then determines like, hey, do we send out this door order right away or do we wait another dirty seconds to forty five seconds to an order work that could come in that's from the same block or the same name. A right, so we could have been and batch it better. Okay, so the driver is on the way or the delivery guy is on the way, let's talk a little bit about that. I mean, first of all, is he an employee of yours or not? So all the folks that are working inside of the four walls in the buildings are employees anyone that's delivering an order, A driver partner there ten ninety nine d a contractor. Okay, So I mean, what are they making? Fifteen bucks an hour? Twenty bucks an hour? So the national average is eighteen to twenty four bucks an hour. How much of what you eat do you order from? I order Go Puff between two and four times a day. So wow, I was a big user. If you got something on the way Now, I ordered in the morning, I got coffee. You try and break it, You try and do something really hard when you order all the time, all the time. I'll order from like a whole bunch of different categories. I'll order from items that I know that are not located in newar Introl inside of the warehouse. And I always order from a pseudo account, so no one notes me. So I'm constantly looking at things and then like, you know, my operators, I don't know if they love or hate me. Right, always always funkle, and you know how we're going to move forward. So if your order comes and it's not just right, you call the you call the warehouse and be like, okay, right, it's not hot. First, I'm trying to understand the root cause, right, it's not like like what you don't want to do is like this is just a blame game of like, hey, this person messed up, and like what caused this defect? And then you drill in like one or two layers deeper with like was it a process issue? Was it a tech issue? Like was it a staffing issue? Like like there's a whole host of issues that could have caused that problem. Ninety five percent at a time, when there's a problem, it's within something new that we're trying to do. So it's it's activated by me because I'm ordering a lot of the new stuff. I want to be testing it, and like they'll they'll be issue because it's like it's it's experimenting. So like you don't want to like fault people too hard for trying and like you want to fail fast and learn quickly, you know what I mean. So you don't want to hit people too hard for trying stuff and not working out the first time around. That's okay, you know, I mean, we'll learn and we'll move past them. We'll you know, we'll get better. Dex Can you talk about out a problem you're trying to solve now, one of the hardest problems for us to solve is grocery. Grocery meaning basically produce fruits and vegetables. Fruits and vegetables. Yeah, so okay, first of all, tell me why it's a problem. The products don't have an expiration date, so like a banana might start looking bad, you don't want to start delivering that product. So it's a whole host of new technologies that we have to build in to go inventory, to really make sure that, you know, we could deliver it properly and profitably. And in two, the retraining of all of our associates. You have to train people what does a good banana look like versus what a bad banana looks like? You know, you can't put it into the machine, like you can't automate the meaning of a bad banana. Yeah, and in some sense, yes, So okay, I'm just gonna pause here and say we're at seven minutes. We're do a time check. Seven minutes looks like I can actually see It's like when you're, you know, calling an uber or something. I can see where the delivery guy is and the dude has looks like moved away from my house a little bit. So I guess I was not first on his list. The order might have been batched, so like the keys that still keep it in the in the delivery window that they use it. Like again, what a lot of people would do, or a lot of people that are entering the space for the first time would do is just like optimize, like bring it as fast as humanly possible, And that might not. I don't know what's happening specifically with your order, because obviously I'm not I'm not I'm not looking at it. But like the reality is is like we're optimizing to build the best kind of consumer experience and build a long staining, generational, profitable business. So, Okay, you were early to this business, and tons of other people with huge amounts of money are coming to copy you and do what you do, like lots of you know, startup VC funded businesses. But I also just read yesterday that like Walmart is trying to get into instant delivery, and both of those seem like potential problems for you in terms of like competition. What I'll tell you is, like I mean this in the most amount of respect, right, the vast majority players just won't exist six to twelve months from now, because the twelve months that is a fuego take and you're not saying in five years, you're saying in one year. Yeah, not all of them, but most of them. And here's why. Right, this business, if you don't focus on building out the tech early, and you don't focus on fundamental economics early, right before you scale the business, it's a fricking disaster. Right. You could be losing twenty thirty forty dollars every order that you deliver. So the easiest thing to do is to open up the building, right, the hardest, like, it's much harder to operate these buildings and operate improfitability and to deliver against the consumer promise every single time. And that took again, that took us a lot of years and a lot of iteration, a lot of work to do. So I just I just know what it takes to build a business like this, and if you don't invest in the fundamentals of the business early, the results are disastrous. My order is still on the way, apparently according to the app. The clock is still ticking, and we'll be back or a break to talk about more problems and see how long it takes my order to arrive, and now back to my interview with Raphael. I will say, for those of you keeping score at home tracking my order of oreos and apples, we're going to skip ahead a few minutes here and pick up the interview around twenty eight minutes after I placed the order. At this point in the conversation, we're discussing Gopuff's acquisition of this liquor store chain called Bevmo. Gopuff made the acquisition a couple of years ago. It costs them three hundred and fifty million dollars and go Puff has been trying to solve this problem. How do you keep this liquor store chain running as retail stores where customers can go and chop well at the same time using the stores as microwarehouses for its delivery business. The reason we bought it was because of the liquor licenses, so we could deliver alcohol in California in an easy way. And we did the same thing in Kentucky with a chain called Liquor Barn, and we where did these retail locations And our biggest problem, and the problem that we still have today, is how do you keep the integrity of the retail location while making it into what we call an omni store. So the store like a micro of a filment center within a very popular chain. Right, that's producing hundreds of millions of revenue. What's an example of the thing that went wrong in that in that setting? So I'll give a like an example. Let's say like something exists on the store shelf. Right, we never had walking customers before, and we have an active inventory account that's integrated between bevmo and Gopa. Let's let's say a customer there's a one bottle of something left and he took it, and it's physically in the cart, in someone's cart, So the inventory system is saying there is one of these bottles that exists in the shelf. So I go on the app and the app is like, yeah, you can buy that bottle of whiskey, and I pushed the order and then somebody who works then then would happen. They would go to the shelf, they would scan the shelf, they would look at the shelf and there's no bottle there. So that like, that's what we call an inventory defect. But the defect shouldn't exist because the bottle physically, like the way that we solve that problem is like anything under a certain amount, we're just gonna market out of stock because we're not so sure one or two bottles left. You just don't let anyone we're gonna market out of stock. And like that took us, like we didn't we thought there was something wrong with the system. And then the engineers and product folks and everything we laugh about this all the time. The best person to fix a packing problem is the person that's packing the order, right, So it took us speaking to the people in the ground to like be like, hey, like what's wrong with it with this, Like like wait, there's something wrong with our inventory management software, and like somebody worked in the store. It's like no, man, somebody some customer just had the bottle in their cart. What that's literally exactly what happened. Instead of like going to the MFC's and really figuring out what the problem is, yeah, the stores, Yeah, yeah, we try to solve it an HQ and that just didn't work. So like you gotta have a bottoms up approaching. Sorry, I apologize, but I have to interrupt. Were at thirty one minutes. I'm gonna stop to stop watch. Just got a text your non content Go Puff order has been delivered to the address you provide it. You want to just hold on one sack. I'm gonna go downstairs and get it. Great, you're right back, and now back to the show. I just went to get my order from Go Puff. I think it worked. Let's see. Okay, let's see. I love this. Yeah, right, it's action. It's like this is this okay? So let's see at the apples. You'd be glad to know the apples look good. Great, we'll do the apple test here, you're ready, good crunch or good apple? And we got the orioles. I love it. Yeah, it worked. Were you worried at all? I think there's like like like a sub level of worry, especially about the apples because it's a new business. Great, let's do a lightning round. We close this show with like a bunch of fast questions, just fast. What's one piece of advice you'd give to somebody trying to solve a hard problem? Listen more than you talk. What's the hardest thing about growing is physical real world business? Insanely quickly. I mean I'm thinking I'm talking from my perspective more than anything else. We're working sixteen seventeen hour days, seven days a week. Yeah, and after months and months of doing it, it just waited, Like you're like, the business hasn't gone into the place where we got we wanted to get to, wen't raise any money, and like you start to deteriorate a little bit, you know what I mean, Like remaining mentally strong as you're building these operationally complex businesses, especially if you're acting as a founder operator, is probably one of the hardest challenges. What's a thing in the world outside of your business that you think should be faster. Maybe transportation probably should be faster, right Like we had we had concords twenty years ago, and I think we took a step backwards on transportation. If everything goes well, what problem will you be trying to solve in say five years. So I think, you know, gobus business can become a lot more broad than just what we're doing right now. What I mean by that like if you look at Amazon's business right like FBA fulfilled by Amazon is larger now than the retail business than Amazon owned products. And by owning all the infrastructure, we're starting to think about, like, you know, what are ways that we can utilize all the infrastructure and all the technology that we built to empower a whole host of e commerce websites to power their delivery. How will you know when it's time for you personally to do something else. You know, it's a great question. I used to think about that kind of all the time. I think, you know, you get one chan unless you're like Elon Musk, right, you get like one chance in life to build something really really generational. And when I said generation, I mean multi hundred billions in market cap and like a lot of magic needs to happen for you to get like timing, product, market fit, resources, damn, Like a whole bunch of things need to like come together. I'm not embarrassed to say I don't think I'll be able to do this again at this Yeah, Like I think this is our chance to do something really really a generational once once in a lifetime. And it's like it's so much bigger than money. It's so much bigger than you know, like how I'm gonna feel on how I'm gonna behave if I sold this thing. So I love doing this, not sounding a cliche, but like I really enjoyed doing this more than anything. I don't really have any other hobbies. It's a good thing, working sixteen hours a day. Right, It's like family and go Puff and most like go Puff. It sounds like, yeah, I mean this is my life, so you know, it's uh. I can't even imagine right now life would out it as crazy as that sounds. Thank you very much. Great to talk guys. Thank you. Rafael ilishaiyev Is, co founder and co CEO of Gopuff. Today's show was produced by Edith Russolo, engineered by Amanda ka Wong, and edited by Robert Smith. The music by Luis Gara. What's Your Problem is a co production of iHeartMedia and Pushkin Industries. If you like the show, please leave a review or rate us on whatever app you're using to listen. If you do not like the show, I've taken enough of your time already, no need to leave a raining. I'm Jacob Goldstein and I'll be back next week with another episode of What's Your Problem.

What's Your Problem?

Every week on What’s Your Problem, entrepreneurs and engineers talk about the future they’re trying  
Social links
Follow podcast
Recent clips
Browse 144 clip(s)