Inside Whatnot's Bet on Live Shopping | The a16z Show

Key Takeaway

Treat every marketplace or customer experience as a place people choose to visit, not merely a transaction funnel. Today, identify one moment where customers must know exactly what they want before engaging with you, then add a discovery layer: a live demo, Q&A, guided recommendations, or behind-the-scenes story. Make the experience useful even for people who do not buy immediately. Entertainment, expertise, and trust can create the repeat engagement that transactions alone cannot.

Episode Overview

Whatnot’s founder explains how childhood Pokémon-card trading, experience at YouTube and Facebook Marketplace, and a stagnant collectibles market led to a live-shopping platform. The conversation examines why live commerce blends discovery and entertainment, how it helps small businesses build direct customer relationships, and why trust, safety, and human connection are central to scaling the platform.

Full Transcript

Transcript of Inside Whatnot's Bet on Live Shopping | The a16z Show from A16Z. Auto-generated from episode audio; may contain minor errors.

If you look at e-commerce today, you have to know exactly what you're looking for. Live commerce accounts for over a third of all trade in Asia. Is this true? 30–40% of all trade in China is live commerce . And how many in the USA today? In fact, single digits. The interaction experience is truly imperfect . Transactions were poorly integrated, as were shipping, payments, search, and customer experience. And we just thought we could create a better system. Let's start with the time when you were a child.

Your first online sale was a holographic Pokémon card. At that time, eBay was just starting to develop. I thought: wait , you can make money on Pokémon cards? Like , I love Pokémon cards and I can make money from it. I sold the first one for 10 bucks. This is probably the true story of the origin of Whatnot. Users spend 95 minutes a day on the platform? Yes, about 95 minutes. This is incredible. Users don't care about the market. What matters to them is whether they will enjoy the thing or whether it brings value.

And it's this kind of experience that makes it interesting and keeps you coming back . What are the most exciting things you are working on and want to develop over the next 12 months? Yes, I mean Okay, I want to start with when you were a child. So, I understand that your first online sale was a holographic Pokémon card, the buyer sent you a money order, you cashed it at the post office, and then shipped the card. So, what age was that? Was it some kind of early, formative age?

I must have been in third or fourth grade, I guess. So, I don't know how old you were ? Somewhere around 7, 8, 9? Something like that. Yes, of course. So there are all these stories, you know, about the founder who found their market so early. Tell us how this experience became a formative element of what you ultimately created Whatnot. Well, I think, like all kids back then, there was just a crazy hype around Pokémon. Pokémon cards became popular, you know, I convinced my dad to take me to the store.

Opening my Pokémon cards. I had my own album. All my Charizards were there . It was everything. And , you know, down the street from where I lived, there was a small lake and a place to relax. My friends and I used to go there and trade Pokemon cards. I tried to trick people to get better cards. And just at that time, eBay was just starting to develop. And their biggest competitor was Yahoo Auctions. And, well, I do n't know how I found out about it , but I did .

I'm like: wait, you can make money on Pokemon cards? I love Pokemon cards. Can I make money from this? So I asked my dad to help me create an account. And I, you know, put a bunch of my holographic cards online. The first one sold for 10 bucks. It was Chancey's holographic card. Chancey is a card of luck, yes. And what always seems funny is that at that time the internet was so embryonic that online payments didn't exist , and that's just crazy. Yes. Yes, it was quite awkward.

I mean, yeah, the early days, you know, there was no trust for that. That is, there was no infrastructure, but of course there was no trust either. Yes, but it's also pretty wild that there was enough trust for someone to send you a money order in the mail and expect you to send a card in return. So I remember getting the money transfer and being like, okay, now I can send the card. Dad, can you take me to the post office? I'll package it up, you know, ship it, and that's probably the real origin story of Whatnot.

Whatnot is quite interesting in that almost all of my life experience, all of the big chunks of my major works, have come full circle in Whatnot. Yes. Pokemon cards, I worked at YouTube for a long time, so I was doing video, then I worked at Facebook, doing marketplace stuff, and Whatnot— it's kind of a combination of all those things. Yes, the perfect combination of all these trends. Yes, that's great. So, there are different versions of how you and Logan decided to start the company. How did you guys get to this point?

Yes, this is Logan's version, and this is my version. Yes, okay. Let's hear it. And, you know, they're both actually correct. I think Logan is just more, uh, nostalgic, his story is probably even a little better. So, Logan's story began somewhere in January 2019, I think. We went to Tokyo for the holidays, for New Year's, sat at a bar, drank; We both had experience in our own businesses, so after a few drinks we started talking: "Oh, what could we do one day?" And they started talking about creating some kind of marketplace.

We bought about 30 or 40 domains, and then the trip was over. We came back to the real world, and, you know, for about 6, 7, 8, 9 months, nothing really happened. And this is all the truth. We discussed it, said we would do it, but then this happened ... It wasn't the launch moment, right? Yeah, there was such a huge gap in time, and, you know, my moment came somewhere around October 2020 or 2019, yeah. And I'm like, "Logan, remember when we were talking about those businesses and buying a hundred domains while drinking at a bar in Tokyo that we never used?" Maybe now is the time?

Logan was just finishing his work, and I was still on mine. But I thought I needed to get back to creating something, and we saw a lot of our friends start collecting again. It was Funko Pop, comics, sports cards, and I thought I knew a little bit about this industry, because I had always been interested in it, even after Pokemon. I was into sneakers for a while, then some Funko stuff, so I knew the market, and it hadn't changed much for a long time . And everyone, like us, now adults, has free money; they grew up with a mobile phone in their hands.

I think there's a future in collectibles and social commerce, and if someone's going to invent it, why not us? Let's see what we can do. Yes. In December 2019, we both quit our jobs, went full- time, and launched our marketplace of certified Funko Pop figures. So I have to ask, was Whatnot among the domains you purchased back then? Oh no, that would be... can you imagine? Pretty good. That would be incredible foresight. Yes, that would be amazing. Actually, the origin of the name Whatnot , which I'm always asked about, because Whatnot is a very, very apt and recognizable name.

She has been with us for a very long time. This was one of the domains we bought, I think we were drunk at a bar in Tokyo. So don't think our judgment was flawless, it was " always legit." And I'm not kidding , because that was the first domain for Whatnot. And the reason why this was the first domain for Whatnot was because we were obsessed with the idea of rapid progress. Yes. Let's just get it done. Yes. I said we don't even have the product .

We have nothing . I don't want to sit here and discuss branding. Yes, of course. Just do it. Let's just get started. Yes. So, it was "always legit". And then our first designer at Whatnot, a guy named John who I worked with at my previous startup, said right before the launch, "I can't be associated with a company called Always Legit." That's a terrible name. And so it was. I made it up, and it was a terrible name . But I said: it doesn't matter. We do n't have a product.

Who cares about the name if there's nothing? Yes. So he made us gather in a room and brainstorm for an hour. We have compiled a list of 10 names. We decided: okay, any of these names will suit us. Let's see if we can buy the domain. I chose three, John three, and Logan three or four. And Logan got Whatnot. Of all those domains, each one would have cost over a hundred thousand dollars, and we didn't attract any investment. Logan contacted the guy who owned Whatnot. And he said, "I'll sell it to you for one bitcoin." So I was one of the early investors in Bitcoin.

That's why I had Bitcoin. This is incredible. And Bitcoin was worth $10,000 then, not as much as it is now. And Logan brought the price down to half of Bitcoin. It can't be. Yes, I got it for five thousand. Yes, yes, for five thousand. And so I hope the guy held him. I hope so too. What, I mean, is a great price for that domain. And, uh, the escrow service we used was a little weird because they had n't worked with Bitcoin before. So, in short, we sent the bitcoins, he sent us the domain, but he sent the wrong one.

He sent us "why not" instead of " whatnot". But we thought: you know, not bad. Also a pretty good option. Pretty good name. But we're like, let's try to get the " whatnot," because we've always had the idea of ​​selling a lot of different categories of merchandise and collectibles. So "whatnot" always felt a little better. And luckily, you know, I think it took us 36 hours to get in touch with the guy, but 36 hours later we were able to get a "whatnot," and for a while we almost got a "why not," which would have been nice too.

This is great. This is a stunning story. I haven't heard that part before. Um, okay, so by the time you started in the US, Facebook Marketplace was already up and running, right? In the sense that there was a lot of activity there, right? And you could see the network effect at that time. Is that so? And then it already started happening in China , right? The live shopping trend has become a separate category. This was at an early stage. And in the US and the Western world you had, for example, QVC.

So tell us about the state of the market when you started and what you saw that gave you confidence that this would be a very good idea. Well , that's probably the state of the market when we started. Keep in mind, this is around December 2019, right before COVID. First, I'll say up front that we probably didn't understand the market very well. We had no idea about live shopping in Asia. We've heard about QVC. Yes. And my impression of QVC was that no one wants to shop there.

It's like grandmas shopping on QVC. So no one my age wants to shop there. Um, listen, there were a lot of other big marketplaces around. Um, but we kind of, you know, for our initial market of collectibles, there were a lot of options. Um, you know, there were quite a few players who were into it , but it seemed like things were getting a little stagnant. It's as if nothing has changed in this area for a long time . So, this was approximately the state of the market.

And by the way, I'll just say that it's probably an advantage, not a disadvantage, that you guys didn't know much about this market. Because if you knew a lot about the market, you would probably try to build on what already existed, just by improving it a little. Yes. Instead of something new, completely new . Yes, that was a 1000% advantage. Yes. Listen, I think the main conclusion, and I can elaborate on this , is that users don't give a damn about the market. Yes. They don't care about the market at all.

They care about whether they will enjoy the thing. Does it bring value? There’s this counterintuitive point: if you’re trying to build a marketplace, it often distracts from whether you’re actually creating a great user experience. Again, no consumer says, "Oh, this is a live shopping marketplace and I can transact in this marketplace." They ask: “Well, no, what do you offer me? » What is the value? And so, um, there were a lot of players who tried to do that because they saw what was happening in Asia.

This is a huge market. Today it is a market worth over a trillion dollars. And we are like: we do n't know. We created the Funko Pop marketplace and communicated with our customers every day. Then we saw that our clients were using social media in interesting ways, trying to sell products through live broadcasts, but the experience was very imperfect. Yes. Transactions were poorly integrated, shipping, payment, search, customer experience—all of it didn't work. Um, and we just thought we could create a better service that would solve our customers' problems on the platform.

Um, Logan locked himself in his room for a few weeks, came out with the first version of the product, and, um, I went live. They sold a lot of goods, and the rest is , you know, history. This is incredible. Yes, cool. So, it's interesting to think about what exactly you're providing. I think it's a very apt observation that buyers don't care about the market. They want a product that appeals to them. What value do you offer them? Shopping often begins with intention. So how much is entertainment here, and how much is intent, and how do you assess the distribution of buyer behavior and what they get from you?

Yes, I think Whatnot is like going to a store. Yes. I mean, I don't know if I would go to a shopping mall or a series of retail stores . It's like I know I want something , but I don't know exactly what it is. I will go to this store. I'll go and look around, so I don't know what I've been buying there lately? I bought a few new pairs of shoes, so I went looking for new sneakers. I said to myself, "Okay, I need sneakers." I entered the website address and spent 30 minutes browsing the products.

I did n't know what exactly I wanted. I walked out of there with a purchase, and it's kind of a Whatnot user journey. Whereas, if you look at many of the e-commerce companies that have been created in the past, before us, you usually know exactly what you want. This is not very suitable for discovering something new. This is not very suitable for getting impressions, entertainment, or a good mood. So I think that's exactly the analogy with Whatnot. Maybe you know you want something, but you don't know what it is.

You will come to Whatnot. You will look for her there. You will have fun in the process. Yes, that's great. So all these elements are built into the product. View, communicate, bid, buy, return. Is there anything else I missed? Ahem. Do you mean discovering something new in general? Yes, I think you covered a lot. I mean, one of the reasons Whatnot has worked so well so far is because it's a great value, because you can do so many different things there. Sometimes you're looking for a good deal.

And often you just want to get an experience and have fun. So, on any given day, most people on Whatnot don't buy anything. The vast majority of people don't buy anything on Whatnot. They just come in, look around, and interact with others. Entertainment, yes. So I can draw a parallel, even if it's not perfect, but I like the analogy of going to the mall. You go to the mall, you hang out with your friends, at least that's how it was when I was growing up: we hung out, went shopping, had fun.

You know, I often didn't buy anything. You just come there for the experience. It amazes me, of course, we are your long-time investors, so this is not news to us, but when I tell others how big this market is and how big Whatnot really is, it amazes. So, you mentioned Asia. Live- commerce, it seems, recent data suggests that it is over a third of all trade in Asia. Is this true? Yes, I heard that about 30-40% of all trade in China is live commerce. Yes, but how much is that in the US today?

Oh, I mean, single digits, actually. Single digits. Is there any reason in your opinion why this gap cannot narrow over time? No, I think it will definitely decrease. It's simply a richer way to sell and build a business. And it's simply more efficient. So if you're a fashion brand or a store, you can open a physical store on, I don't know, a famous street, like Fifth Avenue, it's going to cost millions of dollars in rent, and you're limited to people who come to you from 9 to 5, those who are in New York.

And it's a wonderful experience. You can experience it, connect with people, discover things on Whatnot. You can start with zero dollars and reach a completely global audience with no limits . Yes. So I think that kind of value proposition means that if you're a business that sells consumer goods, eventually you'll come to live commerce. I don't think you can ignore it. I think you can attract a large audience, and that's why in China we see numbers of 30 or 40 percent. Yes, if you think about the dynamics of the marketplace and the network effect, the supply side is very attractive because it actually costs nothing.

And with such strong supply dynamics and a large number of sellers, that drives demand, right? Yes, that's quite logical. I mean, the numbers are just staggering . You said that in Asia it's about a trillion dollars. This is a huge number. I checked this before the conversation, it seems e- commerce in the US is also around a trillion, if I'm not mistaken. One of the criticisms I often hear about Whatnot is that Etsy or eBay have limited sizes. I always find it funny when people mention this.

You know, about 10 years ago, eBay had about 15% of the entire market. By the way, their volume is still 80 billion. This was about 15% of the total market. Now it is less than 1% of the market. So the big change is that the market has actually just outgrown them. That's why I always say that limiting your vision of the market or your opportunities to just these indicators is too narrow an approach. eBay is proof to me that there is at least an $80 billion niche market that can be tapped online.

But the main question is: how much of the gap between 80 billion and 1.2 trillion can you capture because your experience is so much better? Do you agree or disagree with this? Yeah, look, I think ... If you break it down a little bit more, we're often compared to eBay because we started with collectibles, which is what eBay started with. But in reality, the format of live broadcasts is very, very different. Yes. This means that a wide range of different types of sellers can use this product; When eBay first started, most of them were small resellers.

If you look at Whatnot today, it's mostly small businesses, and now we're also bringing in brands. So, many of these brands and small businesses would not normally be present on eBay. Yes, they would never ...yes, it shouldn't... Yes. Maybe they have a brick-and-mortar store, or a Shopify store, or, you know, a lot of things like that. So, I think the seller base is inherently much more diverse. And also, the thing is that live shopping expands demand. Because you go in and discover new things that you did n't even know would interest you.

Yes. And if you look at e-commerce today, you have to know exactly what you're looking for. So, all it really did was move offline sales to online in a more efficient way, but you still have to know exactly what you want to buy. True. And as live commerce spreads, it will simply expand this market. So this could be not only 30 or 40% of the market, but the market itself will grow significantly. Yes, it's more like the emergence of department stores or shopping malls rather than e-commerce.

Yes. True? Which actually expanded the characteristics of market demand. Yes, that's right. Yes, that's quite logical. Another thing that's impressive on the demand side — and I want to talk in detail about the supply side, because I think it's one of the most interesting aspects of the platform. This is because on the demand side, users spend 95 minutes per day on the platform. Is this true? 95 minutes. This is incredible. This is a shocking number that puts you on par with social and entertainment platforms. Yes, because it's fun .

You know, to put it simply, what is value for the buyer is a pleasant shopping experience. And online shopping doesn't give many bright emotions right now. And you could go to the mall before, but the dynamics there are changing, so it's not as interesting anymore. And, as I said, most people don't even buy anything. They just come to have fun and gain new experiences. Most people, over 80%, don't buy anything on any given day. They just watch and have fun. This is amazing. Yes, this is an incredible combination of entertainment and satisfaction of shopping needs, where buyers discover new things for themselves , and more than 80% of people do not even make transactions.

This is definitely a form of entertainment. Yes, I actually spend a lot of time there. Eh, that's entertaining. I make deals too, but, uh , yeah, I understand. That makes sense. Um, okay, so this is like the origin story, like the market structure at a high level. And I want to move on to categories and stuff. But I would like to talk about the supply side, like the sellers on the platform. Talk about the scale of sellers you have . Um, and you know, this is one of those underrated things.

By being a marketplace , you are helping to create new businesses or grow existing ones that otherwise wouldn't exist. So I'd like to explore this a little bit. Yeah, I mean, I think one of the most interesting things about Whatnot is that it fundamentally empowers small businesses. And if you look at what's happened in e-commerce over the last 20-30 years, it's become so effective ; and one way to make it that way is to hide who you're buying from and hide the opportunity for people to build brands and businesses around it, and Whatnot is the complete opposite.

If you are on Whatnot, you will have your own store, your own brand, and your customers will come back to you. And so, you know, I think that's one of the best things about the platform, because we generate billions and billions of dollars for small businesses. These are companies ranging from one or two people to several hundred employees in large projects. This is amazing. Yes. I think you've compared it to the owner of a local shop or a local pub before. And if you think about what Whatnot allows such a person to do compared to say Amazon, which is very utilitarian, yes .

Um, you know, very gamified and with competition between top-tier, third-party, and international sellers. It's just a completely different experience for a seller that will allow a creative and ambitious person to succeed, perhaps in a way they couldn't on those platforms. Is this fair? Yes, I think it just empowers small businesses, whereas the previous trend was to almost disenfranchise small businesses and hide them. Yes, yes, yes. To gather power, essentially. Yes, yes. Well said. Well said. Um, so you have a very interesting idea about all these different types of entrepreneurs, right?

So, um, I'd like to start with maybe just, you know, the initial categories , like collectibles, cards. What do you see, uh , about the behavior in that market? Um, so I think we're seeing a continuation of the trends that we saw when we first started this business, which was one of the reasons we got back into collecting. We've seen a lot of people coming back to this market. These were people like us who maybe had money and bought things that brought back childhood nostalgia and got re- enamored with the unique, you know, things that they loved.

Um, and this trend has essentially continued unabated for the past six and a half years. So, um, you know, I think collecting is pretty much becoming mainstream now, for lack of a better term. You have mainstream celebrities there, you have athletes there. So I think a lot of people thought and thought of collecting as a niche thing. But in reality, it's becoming a major part of pop culture. And I think over the last six years, we've just seen this evolution continue to grow, grow, grow, and grow.

And you know, actually, you know, a lot of the new art that people are collecting now is , like, trading cards that sell for millions of dollars. And you know, maybe your parents, if they were into high-end art, or you were lucky enough to have parents who could do that, they would buy, you know, famous artists, and now people are buying, like, unique LeBron James cards, things like that. Yes, yes, yes, that's super cool. Uh, yeah, and there 's obviously an element of entertainment there too.

I am amazed by the scale of some of the companies on Whatnot. So, you're obviously empowering millions of small businesses, but, you know, if you do it right, a business owner can actually build something with over $100 million in revenue and very profitable, all thanks to Whatnot. Yeah, you know, the biggest businesses on Whatnot are probably over $100 million in revenue and, you know, with very strong EBITDA margin numbers of 20, 30, 40- plus percent. You know, what we're trying to do is, um, you know, so if you look at Whatnot last year, we did, you know, over 8 billion in sales, and this year we've already done over 8 billion in sales.

And we don't, we only take a small fraction of that. Yes. You know, we, we would like to...we want to...we're not raising commissions or anything like that. It is very important that the business receives the majority of the profit, and our business grows with them. So we just spend all our time helping build great companies and providing the best experience for the people who work with them . Yes. Yes, that's cool. This is one of the great and perhaps lesser known, lesser covered stories of small business empowerment in the United States, and now increasingly in Europe, that we can talk about, but it's incredibly exciting.

Um, I want to change the subject, this was obviously a bit on the proposal side. Um, you know, tell us about some of the categories you're working on. Obviously, we started, you know, with collectibles and cards, and now you're successful, you've had great success in fashion, which is of course one of the biggest categories in the retail, and there are many others. So tell us about the evolution of category expansion over time. Yeah, I mean, if you look at Whatnot today, we sell products in over 100 categories.

Um, and we're always investing in new areas to help people grow their businesses. So some of the interesting areas today are fresh food and drinks, where you can buy fresh fish straight from the dock in San Diego or, um... By the way, Ben and I started buying it for dinner. Yes. Yes. Yes. Yes. I like it. This is great. Um, another one of my favorites is a barbecue restaurant where you can get amazing barbecue, and I have it in my freezer right now. Um, you know, I just think with fresh food...it's still early stages for us, but I'm excited to buy it because I think the food supply chain in the US is so opaque, and it's really nice to know where your food comes from and to know its history.

Um, you know, to change the subject, for some reason, all my friends are playing golf now. I'm at the age where it comes. I'm getting old. Yes. Yeah, um, my hip hurts, so I can't do any other real sports. Um, and golf at Whatnot is currently experiencing a huge boom; you know, it's growing hundreds of percent year over year, and we have all these golf stores selling, so that's, you know, another interesting category that I've invested in and that our team is investing in, and you can find a lot more .

Yeah, it's like anything that requires careful selection, right? Should be suitable for Whatnot, where explanations are needed, a charismatic personality presenting it, or some kind of story behind it, it might be suitable. I think anything that someone would build a retail store around. Yes. would be good for Whatnot. Anything, anything that would be in a mall, yes. Um, literally anything, because if you were going to visit a store, now you can do it from the couch via your phone. Yes, that's cool. What about clothes and accessories in particular?

Um, well, women's fashion, women's clothing is our biggest category right now. Oho. Um, and, you know, there's a huge assortment of all kinds of things. Um, I would say something like TJ Maxx dominates, meaning it's just last season's items at really good prices, so you can get great discounts on branded items. Same with sneakers on Whatnot. Um , I almost forgot the name, like, I don't know what it was. When I was a kid, there were these discount shoe stores where you could get Nikes for 40 bucks or something.

There's a lot of it out there , and these are just great deals. I bought, I bought from many of them. Yeah, it's like, you know, nothing beats a good discount while shopping. DSW. Yes, this is exactly the store I was thinking of. Yes, it's like TJ Maxx for shoes. Yes, and a lot of this is appearing in the field of fashion and clothing. And live broadcasts are great because you're obviously moving huge volumes, and when you can sell so many products, you're willing to discount, which gives customers a very good price.

Yes, 100%100%. Um, the other part, you know, of the expansion—is that you've obviously had tremendous success in the US. Um, tell us about the expansion into Europe: what worked, what didn't, what conclusions did you draw? Yes, well, I think we are represented in about 10 countries today . So, the US, Canada, North America, five countries in Europe, and now Japan and Australia. So we cover a significant portion of the globe, so to speak. I think the most interesting thing that might seem counterintuitive to people is that we see how everything works the same in all of these countries.

All those small businesses that we helped create in the US in collectibles, fashion, and electronics, we were able to create there as well, and many of the dynamics are incredibly similar. Therefore, we are very confident that the live format is an extremely sustainable and valuable tool for shopping. One of my favorite stories is a recent one about a seafood distributor who contacted you on LinkedIn. First of all, what an apt request that was. And then it went live the next week and sold, I think, three or four thousand dollars worth of fish during the pilot phase.

What caused this and how did it even happen? Again, the main conclusion is that live streaming is just a really good medium for discovering new things, right? His story is that he has been in the fishing business for a long time. He knows exactly what he's selling. It will bring you the freshest fish from San Diego this time of year and offers what is seasonal. He is incredibly knowledgeable in his field. You watch a show, you see a species of fish you've never heard of, and he tells you it's sustainably caught , that it was just brought in from the pier, and here's the fisherman who caught it.

It's just a wonderful experience. You can take it, we can translate it to Whatnot as a big seller, and then it's pretty easy to generate thousands of dollars in sales per hour with that kind of expertise. And so, imagine if he tried to sell over the internet in general. It may gain access to the local market, but it is unlikely to be able to truly reach the national level. Yes, except for something like this. And I think that buying food online, like fresh fish, is actually very difficult.

Well, it's one of those things where you think: fresh fish on the internet? How do I know if it is of good quality? Will I get sick from it? And video really increases the level of trust significantly. Yes. Yes. Yes. Yes. And the real personality of the person offering it. Yes. You can chat with him . You can ask him questions. You see exactly how he packages it. And if it's a static site, you just think: "Well, I hope it's good." Yes. Yes. Yes. Yes. This is true.

These are very correct words. So , if you think about the platform and the scale that you're working on, it's just a colossal level. And one of the characteristics I use to describe you and Logan is tirelessness. You are incredibly persistent in completing tasks. What problems of this magnitude do you constantly have to face? I think it's important, because now you're big and have a network effect, but I think it's this underrated ability to constantly monitor what could go wrong that helped you win in the beginning.

So I would be interested to learn about some of these global challenges. Yes, of course, it's not easy. There are so many aspects of trust. For example, is the platform reliable, does it work stably, and is it fast? Therefore, we measure absolutely everything. Yes. I think the main principle of trust is to measure everything. And we invest a lot in it. For example, about 40% of our employees are involved in security and trust at Whatnot . Oho. Yes, this is a huge investment on our part, because the task is complex, and no one wants to enter card data where there is no trust.

And buying fresh fish from a place you don't trust is especially risky. Rightly. So, for me, a lot of measurement, user feedback, and a lot of investment in people—that's what drives the process. And problems come in many different forms, so it's important to make sure you're building the right tools to solve them. So, as an example, we have a system in the background that we call our rules engine. It processes billions of data points in less than a second to detect or stop any unwanted behavior on the platform.

So if you insult someone during a live broadcast , LLM will detect it. This feeds into our rules engine, which checks the seller's history and, based on other signals, automatically blocks them, suspends their account, or issues a warning. If you don't ship an order on time, the rules engine will do the same. If you have a high return rate, the rules engine will do the same. And look, these are constantly changing problems and issues. Yes, weightings and things like that will always change. So you have to constantly keep your finger on the pulse.

Yes. I mean, if you think about how many concurrent users you have and how many transactions you process every day. Yes. I always draw an analogy with security and trust: tens of millions of people use Whatnot every day. Yes. Yes. So the trust and safety team is essentially the police and the judicial system of Whatnot. You have to write down the rules and have people who will enforce them. In a city of 10-20 million people, you're looking at a city the size of Tokyo, or perhaps a better analogy would be several New Yorks.

Yes, two New Yorks. Two New Yorks. Not everything works perfectly in New York, so some troubles happen. People are arrested, but thanks to our systems, we have significantly fewer of them. Yes. Problem. And if Whatnot is a business for a small entrepreneur, then everything happens on video. So, if you're a small business owner, you want to build a great business. Any successful business knows: it needs regular customers. You must invest in customer experience. That's why our sellers support it. And you're on live TV. So you won't be able to hide much during a live broadcast .

You know, all sorts of things happen. Uh, and then you kind of say: okay, we've allocated 40% of the company to, actually, law enforcement or the trust and safety team. And so, uh, you know, it's still a challenge, because if you're running the New York City police law enforcement system, it's a tough job and you always have to keep your finger on the pulse and look for new areas for investment and improvement. Yes. Yes, definitely. This is quite logical. Uh , so this is actually a good transition into the topic of artificial intelligence, because you obviously use advanced AI for a lot of your trust and security work.

Maybe we'll take a step back on AI, uh, you know, maybe on the client product side, not the internal systems. Uh , you know, you have a very human platform where buyers and sellers interact like people. What are the lessons learned on how to most effectively use AI while maintaining that human connection? Yeah, I mean, I think the good thing is that the platform is self-regulating to some extent, meaning, uh, if someone goes too far, I don't know, creates an AI avatar, people won't come back to it.

Right. This is not a good business decision. Because the point of Whatnot is that you visit the store, get to know the owner, and then get to know the people who frequent there . Yes. And it's this experience that makes it interesting and keeps you coming back again . So anyone who does good work is not trying to, uh, destroy the human experience. Everyone who is doing well is going to reinforce it , is going to try to improve it. So, where AI fits in here is in the question: how to help a person remain human, but run their business more efficiently?

Yes. So whether it's about displaying products, we can do it much better. Uh, we, you know, when you set up live streaming for detection systems, you often need to add a lot of metadata. Uh, we can define all this metadata using large language models. Uh, we're investing in this from an analytics perspective. So we will tell you what your business is successful at, what it is not, and where it can be improved. So, you know, honestly, pretty succinctly: we usually look at where we implement AI, where it's effective for us , and then we try to provide that to vendors.

Yes, that's quite logical. And if I understand you correctly, it doesn't involve giving sellers tools with avatars or anything like that in order to No, I mean, look, I'm trying to scale or anything like that. Look, ultimately, we're going to spend energy on something that's good for the buyer and good for the seller. But today we don't see the need for avatars, because that's not what people come to Whatnot for. Yes, they come for human interaction. Yes, I completely agree. This makes a lot of sense.

But we managed to discuss a lot. We talked about the platform today, the history of its creation, the supply side, the demand side. Whatnot is definitely one of the most interesting and unique companies we have invested in. And, you know, I'm very lucky to be a part of your journey. In conclusion, I would like to hear what happens next. What are the most exciting things you plan to implement or in what direction do you want to develop over the next 12 months? Yes, I mean, in some ways it's pretty simple: there's an incredible amount of work that needs to be done to help more companies thrive on Whatnot.

So over the next 12, 24, 36 months, it's a lot about enabling sellers to run a better business on Whatnot, giving them the tools to do that, opening up new opportunities, and improving the experience across more categories on Whatnot. One of the areas that I really want to implement is cars on Whatnot, because you should be able to buy a car on Whatnot, and how cool would it be to see the full process, to get a video about it. Oh, that's a great idea. Yes, that would be just incredible.

This might not be good for my bank account . But of course I have to try it. Yes, you have to use your own product. So, I think there's a huge investment needed here to continue to help people build better businesses. Oh, and you know, to do that, you have to provide a really great experience for the buyer. Yes. Because to build a good business, you need people to come back. So we will continue to invest in customer support, trust and security, reducing shipping prices, and improving the selection and quality of products on the platform.

That's all, actually. Maybe we'll launch in other countries, do something interesting like cars, and eventually we'd like to enter the spirits, beer, and wine market. There are many things that inspire us, but ultimately it all comes down to helping more people build businesses and making sure that users get a great experience from it. This is a great way to end. Your fierce and relentless focus on the seller experience clearly attracts buyers, and ensuring affordability and quality for buyers really works. So, thank you for your time, Graham.

It was very nice to chat.

Go deeper

• All A16Z episode guides