The hierarchy of engagement | Sarah Tavel (Benchmark, Greylock, Pinterest)

1. 0:00 – 3:33

   Sarah’s background

   1. STSarah Tavel0:00

      I think a lot of people think about markets almost like these, like, bodies of water, you know? And it's like, "Oh, it's this big body of water that we're going after." And I actually think that the most interesting markets, you have to think of them like currents, where you're just trying to... There's something happening in the market that's creating this current, where you can have, like, a plank of wood that you put on the, on the river and it's gonna pull you forward, versus a market that doesn't really have that momentum to it, you're gonna have to build, like, something really big and fancy to make any progress. And that's why we care less about market size, because really what you're looking for when we- you're looking at a market are: What are the dynamics of change? What's the current and momentum that's gonna pull the company and make the job easier for the founders to actually build something that endures?

   2. LRLenny Rachitsky0:53

      (instrumental music) Today my guest is Sarah Tavel. Sarah is a partner at Benchmark, one of the most pre-eminent venture capital funds in the world, where she focuses on investing in consumer and marketplace startups. Prior to Benchmark, Sarah was the first product manager at Pinterest. And though I normally have a policy against VCs on the podcast, as you'll see, Sarah thinks very much like a product and growth leader, and I always learn a ton talking to Sarah about startups and marketplaces. We also learned in our conversation she used to play rugby, and was apparently one of the best tacklers in her league. In our conversation, we unpack two of Sarah's killer frameworks for building a startup: one, the hierarchy of engagement, which is an incredibly useful lens for trying to figure out how to grow and scale your consumer startup; and then the hierarchy of marketplaces, which is an incredibly useful guide for helping you build your marketplace startup. If you're building a consumer startup or a marketplace, this episode is for you. We get really nerdy and really deep, just the way I like it. With that, I bring you Sarah Tavel after a short word from our sponsor. This entire episode is brought to you by Gelt, your gateway to smarter tax solutions. Navigating the world of taxes can be complicated and stressful. Gelt is here to change that. They blend advanced technology with expert CPA knowledge to handle your taxes effortlessly. Whether you're a business owner, investor, or executive, Gelt is here to optimize your tax strategy and enhance your financial success. People working at companies such as Notion, Google, and Lyft have already experienced the Gelt difference. They've seen significant savings and a transformation in their tax handling. And now, it's your turn. Why choose Gelt? Because they offer year-round, proactive tax planning and clear, fast communication. With their technology, you'll stay ahead, making the most of every tax benefit. Ready to elevate your tax strategy? Visit joingelt.com/lenny today. With Gelt, you'll save more and stress less. Again, that's joingelt.com/lenny. Join, G-E-L-T, .com/lenny. Redefine your approach to taxes with Gelt. (instrumental music) Sarah, thank you so much for being here, and welcome to the podcast.

   3. STSarah Tavel3:12

      Super excited to be here.

   4. LRLenny Rachitsky3:13

      So you may or may not know this, but I actually normally have a policy of no VCs on the podcast, but you're a very special VC because you're a former product manager. Many people don't know you were the first product manager at Pinterest, and, to me, you still think like a product manager, and so I'm very excited to break my rule and have you on the podcast.

   5. STSarah Tavel3:31

      Thank you for the, the exception.

2. 3:33 – 6:03

   Framework 1: The Hierarchy of Engagement

   1. STSarah Tavel3:33

   2. LRLenny Rachitsky3:33

      What I want to do with our time together is dig into two frameworks that you've developed and use with founders that you work with to help them build successful companies. One is focused around consumer businesses, and one, marketplace businesses. So, let's just start with the first framework. I think you call it the hierarchy of engagement. Just to start, could you maybe just share a broad overview of this framework and also just where it emerged from, where you kind of came up with this concept?

   3. STSarah Tavel4:01

      Sure.

   4. LRLenny Rachitsky4:01

      ...

   5. STSarah Tavel4:01

      ... originally. So, I think one thing you'll notice about me is that I have an allergic reaction to vanity metrics, what people talk about as vanity metrics. And when I made the transition from Pinterest, where I was leading product for the discovery team, so I was responsible for all the discovery surfaces on Pinterest, the home feed, the search, the recommendations, a couple other teams. And ultimately what those teams were about was about kind of engagement, increasing engagement of Pinterest. It was helping people when they were on Pinterest find something that they loved enough that they wanted to pin it to one of their boards. And when I started to meet with all these, you know, really talented consumer founders building consumer social products, this was during a time when everybody was getting excited about, uh, growth hacking. And what you would see is that you would see all these founders coming in, and they all had these up and to the right graphs, where whether it was, you know, it was sign-ups or downloads or MAUs. And it felt to me like it wasn't obvious that those metrics that they were all getting very attached to and focused on and showing in, in these presentations was the wrong thing to focus on. It didn't get to the heart of whether they were on the path to building an enduring cons- consumer social product. And they were missing, at the core, the, the criticality of engagement. And, and, and so I, I just started to feel this, and it became like a pebble in my shoe, and I started, you know, just to kind of go through that process that you've gone through many times before of writing and distilling it, and that's where I, I, I came up with this, this hierarchy.And

3. 6:03 – 10:33

   Level 1: Core action

   1. STSarah Tavel6:03

      what I, what you realize when you look at social products is that there almost always is this action, which I call the core action, of that product that forms the foundation of the product. When, when a user completes this action, it's clear that they both understand the, the utility of the product, they understand what that product is all about and it's an action that if they perform the action, they're very likely to come back. So for Facebook, you know, the obvious action is friending in the beginning days. For Pinterest, it's pinning. I don't know if you're an Evernote user, for Evernote it's, you know, writing a note. And when you perform that action, that means you're an engaged user, and of course there's a lot of other actions that you have to do, you can follow people on Pinterest, you comment on Facebook, but at the end of the day, if you're not doing that action, you're not really a user to the product. And that's why it's, you know, the MAU thing doesn't really mean anything, it's really looking at, I think of users completing the core action, you can look at this on, you know, as a cohort, weekly, daily, whatever the right, the right kind of cadence is. But that's, that's the foundation, that's level one.

   2. LRLenny Rachitsky7:16

      Basically there's some definition of an active user that a company, that a consumer business basically needs to define and essentially figure out what is that, what is the action they're taking that makes them an active user versus just opening up the app or, or whatever it is. Like for Snapchat it might be sending a snap, for WhatsApp-

   3. STSarah Tavel7:34

      That's right.

   4. LRLenny Rachitsky7:34

      ... it might be sending a message. Okay, cool.

   5. STSarah Tavel7:36

      Yes. Yes. And, and by the way, this action, it's very important to pick the right action, you wanna make sure it's an action that scales to enough users, you wanna, like, think about if I optimize for this action, like if I think about my product roadmap and optimize for it, what, what do we end up doing? And just to give you a kind of more concrete example, actually in the early days of Pinterest we weren't sure what it was. You know, we had all these things we were measuring. We were measuring follows, we were measuring, you know, uh, clicking through, liking something, pinning obviously, uh, time on site. And so you would, we would do these experiments and you would get all this different data where some things would go up, some things would go down. And what were we optimizing for? And so there's something that's really, really important about having that clarity of this is the action that is most important for our product, all things around NUX has to lead to this, our, you know, if, if a user isn't doing this, then there's something missing from their experience of the product. And so it's super, super important to get real clarity and, and, and, and know exactly the one that you're gonna be picking to go forward.

   6. LRLenny Rachitsky8:49

      One more ques- question along these lines, how do you think about this milestone versus the activation milestone for a new user? Do you find they're often the same, like getting a new user to, to that aha moment versus-

   7. STSarah Tavel9:00

      Yes.

   8. LRLenny Rachitsky9:00

      ... just ongoing?

   9. STSarah Tavel9:01

      Right, and part of it is, like, how do you measure the success of your NUX, of the activation? To me it is are you taking those new users and getting them into, you know, helping them understand the mental model of your product well enough that they start completing the core action?

   10. LRLenny Rachitsky9:18

       Awesome.

   11. STSarah Tavel9:19

       There was a NUX once that we played with at Pinterest where we didn't even teach people what pinning was, it was like their feed magically appeared and it was a real, you know, that was an example of like, okay, we have to help people understand that they are here to discover something and save it to their board.

   12. LRLenny Rachitsky9:36

       And just in case people don't know what NUX is, new user experience, just an acronym for new user experience.

   13. STSarah Tavel9:40

       Thank you, yes.

   14. LRLenny Rachitsky9:41

       Okay. And then maybe one more just foundational thing that I think we maybe skipped is-

   15. STSarah Tavel9:44

       Please.

   16. LRLenny Rachitsky9:44

       ... what's the best way to think about using this framework? Is this essentially where you should focus your energy, on which elements of your product you should be driving, which metrics to focus on, is that the simple way to think about it? Basically we've talked about layer one and there's three layers. What's the best way to think about as you're talking through this framework of how to apply it and use it?

   17. STSarah Tavel10:02

       In a way you have to think holistically when you're building these consumer products, when you're designing it, but I think it's very important as once you get the product out and you're starting to optimize, to start thinking through each of the levels and, and, and that just very focusing to help you, to help a founder understand where, what are the levers that I really have to be focusing on to improve.

   18. LRLenny Rachitsky10:26

       Awesome. I wanna come back to how to apply it in maybe some examples after we go through the three levels, so I'll turn it back to you.

   19. STSarah Tavel10:32

       Perfect.

4. 10:33 – 14:00

   Level 2: Retention

   1. STSarah Tavel10:33

      So then you go to... So let's say you're, you're building your product and you're starting to see that you have users completing the core action. Then the thing that you have, like the next challenge you have, and I've seen this many times before is that then you have to figure out how to get those users to stick around. You want to retain those users. Obviously even if, if they do the core action once or twice and then they run out of steam, you're gonna be in a really challenged position to build something that endures. And so the test for me of whether you're building a product that has the ingredients to create a retentive product on a micro-level, just at the user level, is that the product should get better the more you use it and you'll have more to lose by leaving it. And so I'll give you just a couple examples there of a couple of my favorite products. Uh, you know, obviously Pinterest. So one of the features that I worked on when I was at Pinterest and we shipped was this idea of a Picked for You feed. And so the idea was every time you pin something to a board, we would take that information that the user gave us and use it to create recommendations in their home feed. It was actually, it may have been the first algorithmic feed that was in a social product because suddenly your home feed wasn't just things, people that you followed and the truth is people weren't really following other people on Pinterest, so we needed a way to make the experience get better the more you used it, so we started to do these recommendations in your home feed.And so it was this experience that the more you pinned, the more personalized your home feed got for you. And then the more you pinned, you also had more to lose by leaving Pinterest because all of a sudden you had all your favorite books, articles you want to remember, the recipes that you were planning on cooking one day, you know, the, the holiday planning that you were doing. And so you wouldn't abandon Pinterest because Pinterest was this repository for these different expressions of your identity or these different bookmarks that you want to go back to. So that was kind of this idea and it, it's very important that the core action is the thing that you use as a product to make the experience better over time. Evernote, another example where, I don't know if you're an Evernote user. People, uh-

   2. LRLenny Rachitsky12:58

      No. I, I tried it once and it was way too complicated and I gave it up quickly.

   3. STSarah Tavel13:01

      Okay. People still laugh at me for, uh, being an Evernote user.

   4. LRLenny Rachitsky13:05

      Oh, wow. (laughs)

   5. STSarah Tavel13:05

      I, I am just ... I'm, I'm a lifetime user.

   6. LRLenny Rachitsky13:08

      Wow.

   7. STSarah Tavel13:09

      Right? And it's one of those things that the more ... You know, I take all my notes in Evernote, I dump documents in Evernote, and what it means is that the more I use it ... So like now I know that I can do a search in my Evernote and I'm gonna find the document I'm looking for and, and now I can never leave Evernote, right? Like it has everything. I have thousands and thousands of documents in Evernote. And so that's a product that's incredibly retentive.

   8. LRLenny Rachitsky13:37

      Unless someone builds a great exporter and then there goes that piece of friction.

   9. STSarah Tavel13:40

      I, I know. Well, that's, that's ... People keep telling me I should, I need to move on to Notion or something else, but not yet. I, uh, I will remain a dinosaur here.

   10. LRLenny Rachitsky13:48

       Wow. I love that that's an example though of just, like, someone could break that barrier and make it less retentive by making it easier to get off.

   11. STSarah Tavel13:55

       Yes.

   12. LRLenny Rachitsky13:55

       It's like an interesting growth strategy, basically.

   13. STSarah Tavel13:57

       Yes. Yeah. It's very true.

5. 14:00 – 19:32

   Level 3: Self-perpetuation

   1. STSarah Tavel14:00

      So let's say now you actually, you have a product, it's growing, more people are completing the core action. When they complete that core action, the product gets more retentive for them, it gets better. The more they use it, they have more to lose by leaving it. Then y- your, your tall task, and this is a very difficult, this is, like, the hardest thing to overcome, is how do you make the product self-perpetuating? And this is where I really, I love to think with a, like, of, you know, every time a user uses your product, let's say they're, they're clicking on the mouse or they're tapping on their phone, I love to think of it as, like, this kinetic energy that they're putting into your product. You're taking that energy and your job with a great product is to take that energy and, as much as possible, convert it back to the experience that they're having with your product. Now, the biggest thing that you can do is a network effect, right? If the more I pin something on Pinterest, the better the experience gets for every user on Pinterest, right? Like, every time I add a pin to a board, I'm creating a new edge in Pinterest's graph that Pinterest then uses to create recommendations and, and enrich their understanding of all those objects on Pinterest. So these ... The network effect is the strongest thing that you can do and obviously if you, if you have that, which all social products have to in some way, you, you have to spend time as much as possible just maximizing where that shows up, fine-tuning it, removing friction so that it becomes ... it's a flywheel that spins faster and faster. But there's other loops too that you have to identify and then maximize, and these are the kind of growth and re-engagement loops, right? And so these are kind of classic loops. You talk about these a lot. They exist in marketplaces and social products. How do you get it so that as your users use the product, they want to share it with other people, they create metadata that you can then use for SEO, you, you know, you have collaborative experiences that pull other people in? There's, there's all different things you can do here, and then there's also things that you can do to re-engage a user. As an example, in the early days of Pinterest, if you pinned something, you're pinning something that you found on Pinterest that somebody else pinned, and so we would send a push notification, "Hey, Lenny, Sarah just pinned your pin to her, her art board." Now that, if you weren't a dormant user at that point, it's been a couple weeks since you've used Pinterest, that notification might pull you back into Pinterest and be like, "Hey, I wonder what Sarah, uh, what other pins Sarah has on her art board." And so it's a great re-engagement loop where Pinterest doesn't have to do anything there intentional. The user is creating the action that drives the outcome that Pinterest wants in that, in that example. Now, as much as I love, uh, Evernote, uh, this is a place where Evernote obviously falls down, right? There's no, there's no loops that they can take advantage of, that when I use the product, I make it better for you. There's no loop where, when I use a product, I want to pull you into the product. They tried, at some point, to do collaborative journals. It doesn't work. And so that's a place where, you know, because of that Evernote w- had to spend money to acquire users, they tapped out. Uh, so they, they kind of missed that, that level three. There's other examples I could speak to of companies that weren't able to transcend to level three that you wouldn't otherwise think. You know, I think of cl- companies like Houseparty and Clubhouse. You know, what was interesting about both of those products is that you would think that the more users that came into Clubhouse or Houseparty, the faster the flywheel should spin. But the challenge was that they relied on push notifications, right? And so I don't know if you had this experience, but with h- let's take Houseparty as a, as an example.... you know, you started to follow a lot of people on Houseparty and then all of a sudden it was like your push notifications just got overwhelming, right? Because it was a real-time product that you had to use push notifications to know the moment to join the product. But when you had so many push notifications, because the more people you followed, the more notifications you had, you just get to this point where you start ignoring them. And so, it becomes this thing where even though the flywheel should be spinning faster, it starts breaking down, and that's the, the real tricky nuance in, in this level three.

   2. LRLenny Rachitsky18:52

      So I think people hearing this are gonna be like, okay, getting people to use your product more often, increase retention, make it viral basically. It's like, it's like stuff they already know. But I think what's powerful about this is this is just a lens, like a clarifying lens on what is most important. There's so many things you can be focusing on to increase your product's success and help it grow. And what I love here is just, it's just like here's the three most important things and then here's the s- the levels, like they build on each other. So if you want to increase your retention, get people to use your... Do the core action more often. If you wanna help it grow, focus on helping it spread, self-perpetuate.

   3. STSarah Tavel19:30

      Yeah. No. That's, that's absolutely

6. 19:32 – 23:54

   The importance of focus

   1. STSarah Tavel19:32

      right. And I, it's... And the importance of it is like I can't emphasize enough how important focus is when you're building these companies. It's, it's so hard and there really isn't a playbook, right? Like every company, it, the, the ones that succeed have to be fundamentally different from anything that was before it. But there are these kind of first principles of the ingredients that they all share and you can use those ingredients to really focus you when you're going through this, you know, the inevitable growth curve and, and n- wanting to know like, well, how do we maximize this moment? How do we really focus on the things that will set us most up for success? And that's where I think a framework like this can be really clarifying.

   2. LRLenny Rachitsky20:17

      There's l- a lot of depth behind each of these things that you, you're being modest about. One is with the top of this pyramid of making it self-perpetuating, the reason that's really important is for a consumer app to work well, most of the time it has to s- be able to spread really cheaply. The cost of a c- acquisition needs to be very low, unless you somehow figure out some paid ad strategy where you can, you can make money with paid ads, which is very rare. Um, so maybe speak to that, just like why that's so important to a consumer product.

   3. STSarah Tavel20:45

      Yeah. I mean, what are you trying to build with a consumer par- product? You're, you're not trying to build something niche, right? You wanna build something that can be a mass market, hundreds of millions of users, billions of users. And it's, you know, it would take considerable capital to do that with marketing. There's one example that we have so far which is TikTok, of spending, you know, more than a billion dollars. Once they figured out actually the first two levels in order to grow, so they didn't actually put money, like user acquisition spend behind TikTok until they had a product that was retentive, and then they knew that they really wanted to, to maximize that. But every other social product has grown organically, and, and they've been able to do so because they figured out how to maximize these loops. And I, you know... Even, even TikTok got to millions of users before they spent all their capital to, to acquire those, those remaining users.

   4. LRLenny Rachitsky21:52

      I didn't know that number. So TikTok spent a billion dollars on paid ads to get-

   5. STSarah Tavel21:56

      Yes.

   6. LRLenny Rachitsky21:56

      ... to where they are today?

   7. STSarah Tavel21:57

      Yes.

   8. LRLenny Rachitsky21:57

      Holy moly.

   9. STSarah Tavel21:59

      Yeah.

   10. LRLenny Rachitsky21:59

       I knew it was... I thought it was... I didn't... I knew it was like many millions, but I didn't realize it was a billion dollars.

   11. STSarah Tavel22:04

       Yes.

   12. LRLenny Rachitsky22:04

       That just shows you how hard it is to get, to break in and s- and become a new social network, essentially, or a new social app.

   13. STSarah Tavel22:09

       It, it is, it is hard. And at the same time though, we do see exceptions to it where I, I still feel like consumers are hungry for new experiences. And so, we've seen a couple over the last, you know, few years that have gotten to tens of millions of users. Nothing that has really s- created the experience that's sticky enough to kind of s- uh, pull minutes away from really TikTok and, and Instagram, which just are these dominant forces right now.

   14. LRLenny Rachitsky22:42

       Your message also reminding me of a recent guest post on my newsletter about this app, Saturn. I don't know if you read it? Where-

   15. STSarah Tavel22:49

       Oh. I didn't. But I know of th- the app, and it's just awesome founder.

   16. LRLenny Rachitsky22:53

       Yeah. It was really interesting, and they had a really interesting insight about... First of all, very few apps ever make it to the top of the app store that break through what's already there. And the ones that do almost always spend a few days there and then they're gone 'cause it's like a one-shot thing.

   17. STSarah Tavel23:05

       Yeah.

   18. LRLenny Rachitsky23:06

       And what they basically talked about is many apps have this viral K-factor where they spread like crazy, everyone starts using them, everyone's in there. But then there's this... The opposite happens very quickly too, where if a few people miss using it a few days, you go in there and it's like, "Oh, nothing's happening here anymore," and it quickly crumbles. Clubhouse essentially went through this. BeReal started going through this where people stop posting and then like, "All right, forget it. Everyone's gone." So most apps have this destructive K-factor that kicks in the same way it kicked in to get it to spread.

   19. STSarah Tavel23:35

       Yes. As people say, if... When something grows really quickly, it also can collapse really quickly.

   20. LRLenny Rachitsky23:42

       Mm. Yeah. Tough. I guess along those lines, is there anything you've seen or learned? I guess the e- the idea of this framework is to avoid that essentially.

   21. STSarah Tavel23:50

       Mm-hmm.

   22. LRLenny Rachitsky23:50

       But I guess is there anything you... Comes to mind to avoid that?

7. 23:54 – 26:04

   The challenge of anonymity

   1. STSarah Tavel23:54

      Maybe the one other th- one of thing that I've seen that runs into a challenge, and I find it runs into a challenge particularly with, with, uh, level two, is anonymity. You know, anonymity is something where if without... You know, pseudo-anonymity and anonymity are very different concepts, right? With pseudo-anonymity, we have it on Twitter, you have it on Reddit, you have a persistent identity that you're creating-... and that identity then can have accruing benefits, mounting loss. My, you know, my, uh, pseudo-anonymous account on Twitter that, you know, just flames other VCs... Ah, I'm just kidding. I, I'm not funny, I'm not-

   2. LRLenny Rachitsky24:35

      (laughs) Just kidding. Is this

   3. NANarrator24:35

      Right, Phoebe?

   4. STSarah Tavel24:35

      ... funny enough to, to make (laughs) something-

   5. LRLenny Rachitsky24:38

      Ah.

   6. STSarah Tavel24:38

      ... like that. But, um, but if, you know, for the people who do have it, they get accruing benefits because they got people to follow them. And that's a mounting loss, too, where they've grown their, their identity on the, on the platform even though it's a pseudo-anonymous identity. The challenge when you have pure anonymity is that you don't have that accruing benefit or mounting loss that, that happens. I don't know if you ever played with Secret, but it's the sa-

   7. LRLenny Rachitsky25:07

      Mm-hmm.

   8. STSarah Tavel25:07

      ... similar to TikTok or something like that. It, these are the types of experiences where over time the, you know, the self prop- like, the network effect doesn't work because the anonymity creates the conditions for bad behavior. And so that starts to pull down the community. And then it as- it is also the experience where you use a product and you can delete it and then come back three months later, and because of the anonymity aspect of it, your experience isn't any different. And, and so I find anonymity, people keep trying it and super talented founders keep hitting their head against that wall because it's, it has the, like the very early fun growth but it just continues to be a product that isn't able to, to persist without some persistent identity.

8. 26:04 – 29:34

   Advice for founders who want to increase retention

   1. STSarah Tavel26:04

   2. LRLenny Rachitsky26:04

      I'm gonna pull on this thread of just ways to increase retention. Basically, this is one of the things, maybe the main thing you're telling people to help increase retention, make it so that it's really hard to leave. Before we get there, just to make it clear, is this framework mostly for social consumer products or do you generalize this to consumer products

   3. NANarrator26:23

      It-

   4. LRLenny Rachitsky26:23

      ... in general?

   5. STSarah Tavel26:24

      I, I think consu- I think it's, it consumer products in general-

   6. LRLenny Rachitsky26:28

      Okay.

   7. STSarah Tavel26:28

      ... but it's most-

   8. LRLenny Rachitsky26:30

      Mm-hmm.

   9. STSarah Tavel26:30

      It makes the most sense for a consumer social.

   10. LRLenny Rachitsky26:33

       Okay, got it.

   11. STSarah Tavel26:34

       Like it works for Evernote, right? Like, like and so Evernote isn't a consumer social product.

   12. LRLenny Rachitsky26:39

       Awesome, okay. So it's like most helpful for if it's a social, your friends are in it sort of product?

   13. STSarah Tavel26:45

       Right, you have the chance of level three with, with, uh, social.

   14. LRLenny Rachitsky26:48

       Right, awesome. Okay. So on this thread of retention, I've, it's a thing that I speak about often, everyone always talks about the most important thing you gotta get right is get retention to a good place 'cause if you can't retain people, nothing's gonna work anyway. So just to spend a little more time here, I guess is there anything else you recommend to founders to help them with increasing retention and getting to a healthy place with retention?

   15. STSarah Tavel27:13

       Number one is actually measuring it. Like I can't tell you how many times I suggested to a founder to f- to track cohorts, and that's a new thing. And so you know, just being really, really clear and intellectually honest on looking at cohorts, I think is, is number one.

   16. LRLenny Rachitsky27:34

       And maybe just describe that bre- briefly and then we'll link to an article that I have about how to do that.

   17. STSarah Tavel27:38

       Perfect. Um, these, this, these are the moments when I should probably turn the mic over to you, uh, to talk about.

   18. LRLenny Rachitsky27:44

       (laughs) No, it's, you're not, so.

   19. STSarah Tavel27:44

       But, um, (laughs) but you know, a cohort, you just, you, what I always like to look at is weekly cohorts for these products. And you look at them in two ways, which is one, for each vintage of cohorts you're looking at like a group of people who signed up in a given week and that's one, you know, one cohort. You look at it both on, you know, an active user basis, like are those people continuing to come back to the product? And I, and I particularly love to look at that on a weekly active user completing the core action basis of like our users completing the core action, what, how is that changing over time for each of the cohorts? And then also looking at activity level within those cohorts. And so what you love to see is the kind of classic smile graph where as a, as the network grows and users start to use a product more and more, you start to see that over time they actually become more retained in the product as opposed to kind of the classic, you know, leaky bucket where the cohort just keeps on dwindling down. And until you reach a point with your cohorts where there is a plateau, you have more work to do on figuring out the retention of your users. The second thing is just focus. Um, so I can't tell you how, how many times also you'll meet a founder and they will, you know, there was one example I saw, it was a, kind of a dating app, friend-making app, and he was growing it via TikTok which is not geographically constrained at all. And so you're growing, you know, he's growing the, the user numbers but without focusing on a specific geography,

9. 29:34 – 31:43

   What founders often get wrong

   1. STSarah Tavel29:34

      it's gonna be really difficult to make that type of product a high retention product.

   2. LRLenny Rachitsky29:39

      Coming back to the framework, which layer do you find most often is the one founders maybe are under-investing in/maybe the most important to focus on, if you had to pick one of these? Or is it super dependent on your situation?

   3. STSarah Tavel29:54

      What I often feel is that a lot of times product founders, consumer founders see where they want to get to. You know, they see, they compare themselves to the full expression of a product that you see with other products, you know Roblox, Instagram, TikTok, whatever it may be, and they want to get there but what that ends up meaning when the, when a new user signs up...... is that there's almost, there's too much in the beginning to, to take in and understand. And you don't then have a very focused product that gets a user to the thing that you most want them to do. And that's actually the importance, part of the importance of understanding, you know, level one, what your core action is, is that you wanna make sure that when a user comes to your product, signs up for the first time, doesn't have necessarily a lot of context on what the product is itself, that they see the thing that they're supposed to do and you get them to do it. And a lot of times, in the beginning, there are just so many other things that a found- that a product might have, that you're diffusing that attention that they would otherwise have had on the, on the thing that's most important. And so, I feel that, that's kind of where I always, the, the, that's the first mistake and it's kind of natural that it would be the first mistake, but maybe the largest mistake that I see people make.

   4. LRLenny Rachitsky31:31

      Essentially not getting the activation moment right, not focusing enough on getting people to that a-ha moment, as people describe. And then again, that repeating again and again as a core action.

   5. STSarah Tavel31:42

      Yes.

10. 31:43 – 37:37

    Examples of core actions

    1. STSarah Tavel31:43

    2. LRLenny Rachitsky31:43

       To close the loop on this sort of discussion, could you share a few, again, a few examples of core actions/activation moments, just to make this fresh again in people's minds? And then also just whatever you can share, but helping people decide what that is. I know there's like a whole, it's all, it's a difficult thing to do well, but any advice for how to find that activation?

    3. STSarah Tavel32:03

       I mean, I'll tell you the exercise that we did at Pinterest, because, uh, you know, as I mentioned before, it wasn't obvious to us in the beginning of Pinterest, like, were we a social network? You know, like, should we optimize for the follow graph? But you have to remember, Pinterest, this was 2011, you had Twitter growing like crazy, Instagram was growing. Those were both asymmetrical follow graphs that they were creating, right? And so, it wasn't clear at the time whether Pinterest was actually just creating a new follow graph, but that was about, you know, things, in- instead of what with Twitter was or what Instagram was. And so, it wasn't obvious. And what happened actually is that there was two efforts that I, I think of as like kind of bottoms up and top down, that helped really clarify that actually pinning something was the core action. You know, it was gonna be North Star. We called it weekly active pinners. And, and the two things were, you know, there's always this like, you know, bottoms up, um, analysis that you do, which is, we looked at every action that you could do on Pinterest. So we had liking, following, clicking through, time on site, pinning, repinning, um, and we looked at, first of all, what percentage of users complete those actions? And if you do that action in a week, what's your propensity to come back the following week? And we basically ranked that. And what we saw was that if you pin something, repin something, which is finding something on Pinterest, I'm using those synonymously, uh, or click through, you had an incredibly high probability that you would come back. So, if someone's pinning something, they're coming back to Pinterest the next week with a, with a super high, you know, more than 90% probability at the time. And then there's like the top down way of thinking about it, which is, well, what is Pinterest for? And if a user does come to Pinterest and they never add something to a board, do they really understand what Pinterest is? And yes, clicking through makes you wanna save it. If the click through is valuable, you wanna save it, so clicking through is obviously really important. But at the end of the day, if they don't like that pin enough that they wanna save it themselves to their board, then we haven't done our job. And so we did this kind of top down, bottoms up analysis to make it very clear that when we launch an experiment or launch a new feature, if the pro- probability of the percentage of people that pin something doesn't go up, or the number of pins for that cohort doesn't go up, that experiment is not a, a successful experiment.

    4. LRLenny Rachitsky34:58

       Awesome. There's a post I'll link to in the show notes where there's actually a shir- a guide to doing this regression analysis on how to figure out your activation milestone-

    5. STSarah Tavel35:07

       Amazing.

    6. LRLenny Rachitsky35:07

       ... basically what is most causal of retention as you're describing. And I love this example from Pinterest. Any other just examples? Again, you shared a few at the beginning of just examples of great activation milestones/core user actions.

    7. STSarah Tavel35:20

       I'll tell you one other one that I thought was super interesting, which is that when I initially published this post, I had assumed, wrongly, that YouTube's, uh, core action was watching a video.

    8. LRLenny Rachitsky35:34

       Yeah.

    9. STSarah Tavel35:35

       And then Shishir Mehta, who is the kind of CPO of YouTube, he reached out to me and he said, "You know, that's what it was in the early days, but we started to realize that it wasn't actually our core action. And we did a lot of analysis on YouTube, and what we realized was that subscribing was the core action on YouTube." And it makes so much sense when he said it, right? You're a creator. You're uploading content onto YouTube. There are a lot of other places at that time that you could have uploaded that content to. But you care about YouTube because that's where you're growing your audience, right? And so, the more people that subscribe to your content on YouTube, the more you have the accruing benefits and mounting loss of using YouTube. And, and so that was super important.... on the creator side, which, you know, one of the funnier things that we'll bridge to our next conversation is that all social products are really marketplaces, right? YouTube is a marketplace. You have the creators uploading content and the viewers who are watching it. Supply and demand. And so then the subscribe button is also very interesting because then from the demand side, the viewer, why would I come back to YouTube consistently versus, again, all the other places I could spend time on if it weren't for me having found creators for whom I re- like, their content really resonates with me, so much so that I subscribe to their content. And so I loved that example because it also, it showed a little bit the evolution that can happen with these companies, but then also just the beauty of like you know you've got something really right with the core action when it's helping both sides of your, of your, of your network.

    10. LRLenny Rachitsky37:26

        Awesome example. And I think it's also a good reminder. People change these things. You come up with your best bet, okay, we tried this for six months, maybe let should try something else, we've learned something new.

    11. STSarah Tavel37:36

        Exactly.

11. 37:37 – 38:12

    Finding your North Star Metric

    1. STSarah Tavel37:37

    2. LRLenny Rachitsky37:37

       I'm also gonna link to this post I wrote around finding your North Star metric, which this is like such a multimedia conversation, we have just links for all these topics to go deeper. Essentially, I figured out the North Star metric for, I don't know, 30 companies, and I feel like the North Star metric of a company ends up often being this core user action. So, like, at Pinterest is sounds like it was WAPs, or weekly active pinners.

    3. STSarah Tavel38:00

       Yeah, pinners. Yeah.

    4. LRLenny Rachitsky38:01

       And YouTube is subscribers.

    5. STSarah Tavel38:02

       In the beginning we called it, uh, weekly active repinners and WAR-

    6. LRLenny Rachitsky38:06

       (laughs)

    7. STSarah Tavel38:06

       ... 'cause we felt like we were at war, so it was good.

    8. LRLenny Rachitsky38:08

       And then it became peacetime maybe.

    9. STSarah Tavel38:10

       Yes, exactly. (laughs)

    10. LRLenny Rachitsky38:11

        Awesome.

12. 38:12 – 38:54

    Who should use the Hierarchy of Engagement framework

    1. LRLenny Rachitsky38:12

       So maybe just to close the thread on this framework, can you just briefly summarize who this is most helpful for and just how they could apply it if they're maybe not growing as fast as they want or they're just getting started as a founder building a consumer product?

    2. STSarah Tavel38:29

       Yeah, I just, I think about consumer founders, product founders, product leaders in these companies that are thinking through the roadmap, what are some most important things that they need to prioritize, the big rocks for the immediate short term. And it's, the framework then provides a lens on what to prioritize.

    3. LRLenny Rachitsky38:51

       And we'll link to the whole framework for people that wanna go deeper.

13. 38:54 – 46:09

    The Hierarchy of Marketplaces framework

    1. LRLenny Rachitsky38:54

    2. STSarah Tavel38:54

       Awesome.

    3. LRLenny Rachitsky38:54

       Okay. So let's move on to your second framework that you call the hierarchy of marketplaces. And first of all, I wanna give a shout-out to Mike Williams, who's the founder of Everything Marketplaces-

    4. STSarah Tavel39:03

       Yeah.

    5. LRLenny Rachitsky39:04

       ... who gave me a bunch of good question i- suggestions to ask as we go through this.

    6. STSarah Tavel39:06

       Wonderful. Now I'm nervous.

    7. LRLenny Rachitsky39:08

       (laughs) He's, he's Mr. Marketplace.

    8. STSarah Tavel39:11

       Yes.

    9. LRLenny Rachitsky39:11

       Okay. So let's start with the same question, just what's kinda the broad way to think about what this framework is and who it's for, and then where did it come from?

    10. STSarah Tavel39:18

        It's a, a framework or hierarchy that is for marketplace founders. It could actually be B2B or B2C. Or, uh, marketplace product leaders that are just continuing, like either getting started or scaling their marketplace and helping kinda prioritize and focus on the things that most matter. And it came out of a similar reaction, a very similar parallel to what I was experiencing meeting all these consumer social pro- founders. Like, they would talk about MAUs, and I just felt like it didn't get to the heart of whether they were building enduring value. And I actually started to realize that GMV was very similar. You know, GMV, it's, it's a metric that it seems like, you know, of, of course you can't build a marketplace without growing GMV, in the same way that you can't build a social product without growing MAUs. But if you focus, if the race that you think you're running is to grow MAUs or to grow GMV, you're not actually gonna run the right race. It can take you, and actually with marketplaces very, very interestingly, it can take you in the wrong direction. So what would happen is I would meet all these founders, and everybody got fixated on this milestone of a million dollars of GMV, right? And, uh, I think a lot of people started to feel like you hit a million dollars of annualized GMV, it means you're ready for your series A. But it's kind of self-evident that not all GMV is created equal, right? Like, so imagine a food delivery company 10 years ago. You could have a million dollars of GMV by getting, you know, focusing on LA, or a million dollars of GMV by focusing on Boston. Or, you could have a million dollars of GMV by having $500,000 here, $200,000 here, two- you know, like spreading it out. And so the founders in... And I would meet so many founders who would say, "Oh, well, I thought we had to prove out that the value proposition resonated in these other cities. And so that's why we're diffusing our focus." But what they, what I would always feel is like they were just making their job so much harder by doing that. The other way you could think of it is like if you're trying to get to a million dollars of GMV as quickly as possible, you're actually motivated to go after a really big market, right? It's a lot easier to get to a million dollars of GMV when you're skimming the cream on this big ocean of a market than to be very, very focused on a smaller market, and getting to a million dollars of GMV in a constrained market is a lot harder, takes longer, but it's actually the path to building something that endures. And so I was having the same conversation again and again, and I just felt like I needed to crystallize it or synthesize it in a way that would be easier to communicate, and that's what led to this hierarchy.

    11. LRLenny Rachitsky42:30

        Awesome. Okay, cool. So you've touched on a number of the important elements of this...... hierarchy. So, let's dive in.

    12. STSarah Tavel42:35

        Okay. There's probably a couple core insights, but the first one that I think about with this hierarchy is that when you're building a marketplace, you actually have to start from the goal and work backwards from there. Now, what's the goal? Like marketplaces, I mean, you have so much incredible content on this. They are, they're, they're just incredibly difficult to build. You're basically building two companies at the same time, right? You have the product and go-to-market org for the demand side, the product and go-to-market org for the supply side, and somehow you have to make these two line up so that they, that they serve each other's needs. It's, it's an incredibly difficult, um, dance and, and it, it really, it, it, it, it's just, yeah, it's one of the hardest types of companies in the software world that I think you can build. And you do it because you think about the f- the full expression of these models, and you think about companies like Airbnb or Amazon or eBay or Google where once you have scale, you ha- and have, like, very dominant market share, you have a product that just is the best place, without question of fulfilling this, this need. And they end up being very profitable, cash-flowing businesses and, and you see that, and that's the business that you wanna build, right? That's why you're going through all the hardship of building a marketplace, getting the marketplace off the ground to get to that place. And, and there's kind of very clear analysis. There's, I published i- in my, in the hierarchy, I have a, this incredible graph that I remember seeing when I was an, uh, board observer for this, uh, online classified company called OLX, led by Fabrice Grinda, and it is, it kind of shows that the more dominant a marketplace is relative to the number two in its space, the more profitable that business is. And so, it's just like a very clear, like you don't wanna be... You don't, obviously, when you're building a marketplace, you don't wanna be in the middle of the pack, but even being number one, but just barely number one, doesn't give you any of the benefits of a marketplace model. You're fighting for each incremental point of market share. It's not clear to the supply side or the, or the demand side that you're, like, the main place to be. And so, it only becomes really clear once you have that dominant market position where you're just so much bigger than the number two. And, and the only way to get to that place, to get to that moment of real dominance where you're, you have this winner take most dynamic is that you have to tip your market. And, and that's the only scalable way to do that, and that's what level two is. And so, how do you get... How do you maximize your chance of tipping a market? You have to focus on a really constrained opportunity, and that is layer, level one. So, that's kind of the, the, like, the broad framing, which is, like, a really important thing of how, when you're building a marketplace, you actually are building towards that point of dominance. And so, it's part of the reason why being really, really, really focused in level one is so important.

    13. LRLenny Rachitsky45:57

        Awesome. So, you've walked through this backwards. Just to summarize in case people didn't totally catch it, basically, layer one is focus, layer two, tip the market, layer three, dominate the market.

    14. STSarah Tavel46:08

        Exactly.

14. 46:09 – 50:19

    Level 1: Focus on a constrained opportunity

    1. STSarah Tavel46:09

    2. LRLenny Rachitsky46:09

       Awesome. So let's maybe start with layer one of just focus.

    3. STSarah Tavel46:13

       Let's do it.

    4. LRLenny Rachitsky46:13

       Which, which I think is the most counterintuitive step-

    5. STSarah Tavel46:16

       Yes.

    6. LRLenny Rachitsky46:16

       ... for people building a marketplace where, as you said, normally it's go big as fast as possible, get as much as possible, just make everyone happy as quickly as you can. And your advice is essentially do the opposite. Make it very... Uh, you have this, like, great acronym of THIMBLE that I imagine you'll talk about it-

    7. STSarah Tavel46:32

       Yes.

    8. LRLenny Rachitsky46:32

       ... that I love, I always think about. So, yeah.

    9. STSarah Tavel46:34

       Yeah. And, you know, it's, and, and part of what's so hard about this idea of focus is like, you know, I'm lucky to meet with, uh, just incredibly ambitious founders. And w- the hardest thing, I think, about building these marketplaces is that the am- the ambition can often feel like the sun. You know, like the heat of the sun that's trying to warm everything up and you're going after this big, big market, you know, warming the ocean. And I think that really what the, the best ambitious founders do is they focus that ambition like a laser beam on a small market, the THIMBLE. And what you do is, like, you're getting that THIMBLE, you're getting the product-market fit really, really right with this s- this, like, small, constrained market. And i- if you heat that up really, really hot, then it expands from there, you know? And part of this is that we have to accept a couple things, a couple points of scarcity. One is that you can't raise hundreds of millions of dollars from the start, right? We're, we, especially today, live in a world where there is a constraint around capital. And, and so you have to be able to take that capital in the beginning and make the most out of it, and the only way to do that is to be very, very focused. And the second thing is that the constraint is the founder's attention. And again, these things are really, really difficult to, to get off the ground, and the only way that you can do it is by having myopic focus on a mar- on a, on a, on a segment. And the example that, you know, I, I always think about, and it's just was a very cool thing that we all got to see which was when the food delivery wars happened, right? Where you had companies like DoorDash and Postmates and Uber Eats, uh, and then the incumbents of Grubhub take very different strategies.... and what Postmates ... I, you know, I, I make comparisons to Postmates and, and DoorDash, both successful companies. But, you know, Postmates, from the very beginning, had very big ambition and went after all the big cities. They went after, you know, not just restaurants, but restaurants and Apple and, you know, retail, bicycle. I remember there were so many things that they went after all at the same time versus ... And, and when you're playing that game, you're always being compared to whatever other substitutes there are on the market, right? And you're ... It, it ... That just ... When you have constraint of capital and attention, you're, you're spreading yourself thin across a lot of different vectors of, of kind of preference for the customer and the s- and the seller. DoorDash on the oth- on the other hand famously and very controversially went after the suburbs, right, in the beginning. And the beautiful thing about the suburbs is that there was very little competition because no one thought that you could do it economically. And they probably weren't wrong in the beginning. Like, DoorDash lost a lot of money in the beginning fulfilling delivery f- in, uh, in, in the suburbs just because the delivery times, the driving was so long. But by going after a market that other people weren't focused on, it, it let them get to a place where they were able to really make the customers on both sides of their marketplace happy enough that they retained. And this word happy I use, I should, I should expand on it a little bit, because the

15. 50:19 – 54:47

    Sarah’s “happy GMV” and “minimum viable happiness” concepts

    1. STSarah Tavel50:19

       m- you know, when you ... The realization I had as I reflected on this kind of feeling that GMV is really actually a vanity metric, it doesn't get at the core of wha- whether you're building enduring value, it's, like, very clear and there's plenty of examples I could give where there are companies that had incredible scale and still were disrupted by a startup. And it's because, like, actually, customers don't care how big you are, real, right? Like, they don't care how many transactions you've accumulated. What they care about is when they have a transaction with you, how happy do you make them? How much better is the experience that you provide than any other substitute that they could use? And if you do a good job, then they're gonna keep coming back to you. And so I call that happy GMV, and that's actually the thing that I think you as a founder or a product leader have to focus on, which is what do I think is gonna be the experience of a buyer or seller that leads them to retaining and tracking that as, like, the happy path, and therefore the happy GMV.

    2. LRLenny Rachitsky51:29

       Amazing. I love that concept, happy GMV. You also kind of extend this idea and you s- I think you called it the minimum viable happiness is what you wanna create when you're building this marketplace, this thimble of water that you're trying to boil into something incredible, that what's the minimum version of that, that it creates really happy customers?

    3. STSarah Tavel51:51

       That's right, yes. And it's, it's ... And, uh, kind of the way to think of it is like obviously when you're a marketplace founder, like, you know that in order to build something that endures, you're gonna have to grow, right? But you ... In the beginning, you're, you're doing all these things that don't scale, you're working on your product experience, you're taking friction out of the, out of the, out of the transaction, and you want to get to a point where there's a certain percentage of people that retain after having done a transaction. And so you're kind of doing the work, doing the work, doing the work to get to that minimum threshold, and that's when you know you get to f- kind of go to the next step of starting to figure out the levers to really scale what you're doing.

    4. LRLenny Rachitsky52:36

       Awesome. And again, it's difficult. It's easy to say, it's difficult to do as a founder-

    5. STSarah Tavel52:40

       So hard, yeah.

    6. LRLenny Rachitsky52:41

       ... to tell them, like, you just need to get a small version of this marketplace working-

    7. STSarah Tavel52:45

       Yes.

    8. LRLenny Rachitsky52:45

       ... versus you go big really quickly. And so most of the time, I think it's actually, like, location. Basically start in a city if it's l- if it's location-oriented, or if it's all online, pick, like, one niche to focus on. Like Etsy was, like, you know, craft fair goods, right?

    9. STSarah Tavel53:02

       Exactly. And, and sometimes as we see, w- as we saw with Postmates, sometimes it's both, which is, you know, geography and category. It's, again, you just, you have, you only have, there's s- you have to assume scarcity of your capital, of your attention, of your customers' attention. And so getting complete clarity of what is the experience that you wanna really nail, what's that happy path you're optimizing for, constraining that in the beginning gives you the option, the optionality to grow beyond that. But if you don't constrain, you just won't be able to get there fast enough.

    10. LRLenny Rachitsky53:42

        This entire episode is brought to you by Gelt, your gateway to smarter tax solutions. Navigating the world of taxes can be complicated and stressful. Gelt is here to change that. They blend advanced technology with expert CPA knowledge to handle your taxes effortlessly. Whether you're a business owner, investor or executive, Gelt is here to optimize your tax strategy and enhance your financial success. People working at companies such as Notion, Google and Lyft have already experienced the Gelt difference. They've seen significant savings and a transformation in their tax handling, and now it's your turn. Why choose Gelt? Because they offer year-round proactive tax planning and clear, fast communication. With their technology, you'll stay ahead, making the most of every tax benefit. Ready to elevate your tax strategy? Visit join gelt.com/lenny today. With Gelt, you'll save more and stress less. Again, that's join gelt.com/lenny. Join, G-E-L-T, .com/lenny.Redefine your approach to taxes with Gelt.

16. 54:47 – 56:36

    Thumbtack: a counterexample to this approach

    1. LRLenny Rachitsky54:47

       I wanna talk about what it looks like when you, you've got it working and it makes sense to start expanding. But before I do, there's one counter-example I found in my research which is really interesting, which is Thumbtack, where they, from the beginning, went national and across many categories. And the founders admit maybe that wasn't the right idea. Took 'em, I think, a long time to get to something that was really working. But on the other hand, their thinking was to create enough of a flywheel for people to keep coming back. How often do you need a plumber? How often do you need a DJ? So there's-

    2. STSarah Tavel55:21

       I was just gonna say, that's, that is the trickiness with Thumbtack's model, and it persists as the trickiness with their model, which is that you don't have a high repeat use case there. And so, you have to, you know, one of the really important things with building a marketplace is cornering the buy side, like having a relationship with the buy side where they don't even think about where they're gonna go, they come to you. And if it's a plumber, you use that so infrequently that every time you are gonna d- like look for a plumber, you're gonna start at Google, right? You're not gonna assume, "I'm gonna go to Thumbtack to find that." Because you've already forgotten by that point. And so, you need to have enough touchpoints with the consumer so that they'll have a relationship with you and start to think of you as a place to go to, and that's probably, that was the trickiness with Thumbtack. And by the way, when we talk through level two, that also is the trickiness that Thumbtack, you know, it, the level two will articulate some more of that trickiness.

    3. LRLenny Rachitsky56:26

       Awesome. By the way, I do use Thumbtack. I've, it's implanted in my head of that's where to go to find plumbers and stuff.

    4. STSarah Tavel56:32

       That's great. I mean, those found- I l- I adore Marco. He is an incredible,

17. 56:36 – 58:06

    Signs you’re ready to move to level 2

    1. STSarah Tavel56:36

       incredible founder.

    2. LRLenny Rachitsky56:38

       As a segue to layer two, what are signs that you've done a good job at focusing and things are working? L- you have this term that I love of creating a white hot center. So maybe speak to that of just like what tells you it's like, "Cool, I think we've got this very focused thimble." How do we, how do we know it's time to go?

    3. STSarah Tavel56:57

       Yeah, I think it's, it's very much that you just start, you see people retaining. You see people... You, you just, you just feel that where, like, people are texting you or emailing or whatever it is that they've had a great experience and you see them coming back. And that's, and you see an, a greater... Uh, you know, you're not gonna make everybody happy. That's, that's... Like, let's all accept that. But there's a core of users, a persona that you are able to make really happy. That's when you know that you're on the right track.

    4. LRLenny Rachitsky57:28

       Awesome. It's essentially the very fuzzy product market fit question.

    5. STSarah Tavel57:33

       Yes. Yeah.

    6. LRLenny Rachitsky57:33

       And I'll link to a few posts I've written that help people get there.

    7. STSarah Tavel57:36

       Yes.

    8. LRLenny Rachitsky57:36

       I guess, is there anything you wanna add there of just like...

    9. STSarah Tavel57:37

       I, I mean, I love... You know, the only thing I'll add, those two things r- related I'll add, I don't like NPS for this. Uh, and I have a whole blog post on this. I like Sean Ellis's question of, like, how disappointed would you be if this product disappeared? And I think it's if you have at least 40% of people respond that they'd be disa- like very disappointed, then you're on the right track. That's kind of the, the feeling that you, that you, that should tell you you're on the right track.

18. 58:06 – 1:04:15

    Level 2: Tipping the marketplace

    1. STSarah Tavel58:06

    2. LRLenny Rachitsky58:06

       Awesome. Okay. Let's talk about layer two, around tipping the marketplace.

    3. STSarah Tavel58:10

       So, layer two, I think will actually illuminate, again, the importance of that constrained focus for layer one. You're doing all these things that don't scale that you've ridden so much about, uh, Lenny, about kickstarting the marketplace, all those things that you, you can't just keep on doing forever. Now, layer two, level two is about figuring out how to do things that do scale. And really, you're, you're doing it in service of this kind of mythical moment of a tipping point, and the tipping point is this idea that you often reach a, some kind of saturation point in a market. Uh, you've reached an experience, you kind of, you've kept on working on the product experience, you've taken a lot of friction out. Where, and you keep on growing, growing, growing, and you're doing that in service of increasing happiness, and, and this is a really important thing. Like, the growth that you're doing should be, you know, you should have a lot of clarity on your flywheel at this point, and know that you're growing in service of accelerating your flywheel. And, and I think of the, people talk about the flywheel as liquidity. I could think of it as happiness, which is that the more the flywheel spins, the more you're able to make both sides of your marketplace happy, right? And what you're doing is you're growing into a greater percentage of your market.

    4. LRLenny Rachitsky59:38

       Mm-hmm.

    5. STSarah Tavel59:39

       And y- as you're doing all these things, again, that don't scale, then you reach this moment where things start to get a lot easier. You've kind of pushed the boulder up the hill and it starts to slide down, and what you notice first, I think, is you start to see it on a micro level, where you might have suppliers that start to tip to you, or buyers that tip. You might see your cohorts get a little bit better. You start to see some organic growth. And, and let, let me ex- kind of give a story or an example of, uh, a company that I got to see where the marketplace started to tip. So we're investors in a company called Reki, and Reki is a marketplace for restaurants and their suppliers. And it's, it's... Imagine in the beginning, this kind of app that you could use. It almost looked like WhatsApp, that a chef could use instead of doing a voicemail at like midnight when they're, when they're leaving the, the restaurant, or texting a bl- you know, typing something into WhatsApp and texting it to their supplier, an email, whatever it is, they could use this Reki app and it would make it a lot easier for them to do that.And, and then of course the marketplace element is that then they'll be able to discover other suppliers on Reki if they need them. Now in the beginning, Reki had to have their own sales force and they were going to London, knocking on the doors of these restaurants, finding and cornering probably the chef in the rest- in the, in the kitchen to show them the app, to get them to change to this new app called Reki. And it's like a very high cost of sales effort to have, as you might imagine. But then what started to happen is that every time they got a new chef onto Reki, that order would then go to the supplier that that restaurant normally ordered from, and so instead of getting the voicemail, the supplier would start to see a nice kind of order form that was from the restaurant and kind of provided by Reki, powered by Reki. And what would happen is that as Reki grew more and more in London, these suppliers started to get more and more of their restaurants ordering via Reki, and then some of the earliest suppliers started to be like, "Well, this is so much easier for me. Like, I would much rather get this order form from Reki than have to listen to an hour of voicemails from chefs with different accents and different ways of ordering their food in the morning when I come in." And so then the sellers would, the suppliers would reach out to Reki and be like, "Hey, here's a CSV of all of our customers. Can you reach out to them and onboard them to Reki? 'Cause I'd much rather have my orders come via Reki than the way I'm getting them." And so you go from this moment of like the hard, high cost of sales, like pounding the pavement, to getting a list on a silver platter of restaurants to onboard, and that's that i- that feeling of tipping, right? Where all of a sudden something goes from being really freaking hard to so much easier.

    6. LRLenny Rachitsky1:02:52

       It feels a lot like how people describe finding product market fit. Things start to feel like you're, you're not pushing anymore and it's just coming to you.

    7. STSarah Tavel1:03:00

       Yes.

    8. LRLenny Rachitsky1:03:00

       In your experience, is this tipping often just happens as you become more successful in this thimble of a marketplace? Or is it things you have to do actively to tip, to get to this place? 'Cause it sounds like in Reki it was like this feature they built of just advi- so recommending, "Hey, you should check out Reki and maybe other, other suppliers can, would help you have a easier time if they're joining." So is it usually like it just happens with scale or is there usually something you have to change to create this tipping point?

    9. STSarah Tavel1:03:32

       I don't think Ronan predicted that that was what was gonna happen, right? And your, your constant, that kind of customer obsession, having the relationships with the suppliers, having the relationship with the buyers so that you, you have your ear to the ground and when they start to, you know, lean in a little bit, you're there to receive them, right? And so I think this is very much like part of the what, what great marketplace founders are able to do is they're able to start to feel that there's something tipping in their direction and then create momentum behind it. And that's, that's a lot of what this level two is all about, which is that

19. 1:04:15 – 1:10:53

    Tipping loops

    1. STSarah Tavel1:04:15

       I think every marketplace has to figure out what I call tipping loops, and there's two types of tipping loops that work together symbiotically to help a marketplace scale. The first are the growth loops. And so we just talked about an example with Reki. There are, you know, many examples of this. Hipcamp, another marketplace I'm lucky to, to work with. If you and I went camping on Hipcamp and I booked, you know, a tree house on Hipcamp and you were gonna join me on it, I could then as part of my checkout flow essentially s- invite you to my reservation so that you could see the maps, the more information about the, the tree house, uh, any of the kind of sites around it. And so be- I'm the person that Hipcamp acquired but then I pull you into my experience, and that's a great buyer to buyer growth loop. There's kind of classic seller to seller loops or buyer to seller, you know, they, there are all these directions and then the great marketplaces see those loops and they figure out ways to accelerate them. So classic seller to seller referrals, right? You know, Uber drivers would get, you know, would tell their friend, "Oh, you should start, uh, driving for Uber. It's great." But then Uber incentivized it with a referral kind of bonus and you started to see Uber drivers like literally writing blog posts about how much they loved driving for Uber because some of them were making more money, uh, getting the referral bonus than actually driving. And so you're trying to find these growth loops and you maximize them.

    2. LRLenny Rachitsky1:05:56

       Awesome. I'll link to a post that has a lot of these, uh, insights from this work that I've done on marketplaces, but a few others that are just interesting. Basically you're trying to figure out how do you create a engine that continues to drive supply and also drives demand? So a fun example is when Etsy was getting started, as you know, they went to like craft fairs and pitched sellers, "Hey join, sign up for Etsy. You could sell your stuff online." And then all their sellers started telling all their buyers, "Hey, buy my stuff on Etsy."

    3. STSarah Tavel1:06:24

       Etsy literally gave them business cards. So Etsy gave their sellers business cards that, that made them feel like, "Oh, I'm a real cr-" Like made them feel professional but th- and the link on it was to their Etsy store.

    4. LRLenny Rachitsky1:06:39

       I love that. One of the most magical growth engines of marketplaces is when your supply brings on the demand. So that's a really good example. Another one is DoorDash where basically a restaurant sign up for DoorDash-... and then they tell other customers, "You want to order your- our food? Go to DoorDash and order it." And so all the restaurants were telling everyone about DoorDash, and then they become DoorDash users and they order from other restaurants.

    5. STSarah Tavel1:07:01

       Yes. And it just shows, it shows how accretive DoorDash was to both sides of the market in that example.

    6. LRLenny Rachitsky1:07:08

       Mm. And then Faire is an awesome example too. It sounds similar to Reki, as you described, where Faire is basically a B2B, uh, artisanal items marketplace where we- boutique shops buy like nice candles and blankets for their store to sell. And essentially they did what you described, where they signed on a store and they basically told them, "All your vendors can join for free and not have to pay any fees if they sign up for Faire. And, and it makes it easy for you to buy their stuff." And then the other, on the other side, a candle maker recommend- tells all the places they sell to, "Hey, you should use Faire. It's so easy to buy our stuff and we can communicate through there, and they can sign up without any fees." So basically everyone just invites all their existing partners on Faire, and then everyone's on Faire. So yeah, there's a lot of ways to do it.

    7. STSarah Tavel1:07:53

       I love it. And so the, the second type of loop, and again, this loop works symbiotically with the growth loop, is what I call happiness loops. And I think of them almost as like the kidneys for your marketplace as you grow. The, the happiness loop, the idea is like you have a lot of new sellers coming in, and of course you have new buyers, but you want to make sure that you're matching your buyers with the sellers that are gonna give them the best experience, right? And so, you know, normally like in a consumer social product, churn is, is, is like, is heartbreaking. Like, once a user churns, they're almost never gonna come back, and you're, with a consumer social product, you're trying to get to massive scale. But when you're building a marketplace, you actually... There's a healthy amount of churn that you want on the supplier side, because there are just going to be suppliers who aren't gonna create a great experience for the buyer, and you can't do anything about that, you know? And so we've all had the Uber driver whom you give one star to. Like, that person you don't want on your marketplace getting matched with, with, with buyers. And so you're trying to then make sure that as you grow, you have a natural mechanism in your marketplace to reward the suppliers that you want to reward and to churn out the ones that you don't. And of course, your job is to do your best to set all the sellers up for success, but it's an inevitability. And so there's two kind of great examples I think of as, in these happiness loops, um, which is around search ranking and then reputation. You know, search ranking is, is just a very obvious one, which is like you're trying to understand what, you know, creates a happy experience for the buy side, and then reward the sellers that provide that experience. So, I think a kind of an interesting example that ended up actually changing but still is illustrative is that Uber Eats in the very beginning, uh, of their journey thought that their advantage in the market was gonna be about having really fast delivery, right? Because they already had this network of drivers, and so they thought that that's where they were going to really lean in and have an edge over any competition. And so in the search ranking, they rewarded restaurants that prepared the food quickly, and so answered the Uber Eats request and then actually got the food out as fast as possible so that Uber Eats became synonymous with like the f- the quickness by which you got the delivery. And so, that restaurant that takes 40 minutes to prepare the food, they would get ranked low in the experience for Uber Eats, even if the food was really, really great.

    8. LRLenny Rachitsky1:10:45

       Okay. So we've been talking about tipping of the marketplace. Is there anything else there that you want to share before we move on to the, the third layer?

20. 1:10:53 – 1:15:55

    Not all markets are susceptible to tipping

    1. LRLenny Rachitsky1:10:53

    2. STSarah Tavel1:10:53

       Yeah. There's kind of two other things that, that come to mind. One, just as a reminder, um, and then, uh, one as a caution. So the reminder is, in order to tip a market, you have to reach a saturation point in that market, right? The, you can't tip a market before you've penetrated it to some saturation level. And it goes back to level one of why you want to focus on a constrained market. Let's take, you know, growing a food delivery product in Des Moines, Iowa versus LA or New York, right? When you're growing in a small city and you're focused on restaurants, you're gonna be able to get to this tipping point a lot faster and more efficiently than if you're going after a very big market. And so, there's a real advantage to being able to prove out the playbook as quickly as possible and as efficiently as possible before you move on to bigger challenges. And, and so I love... I, I think it kind of underscores again why it's so valuable in the beginning to focus on something that is, that is constrained. And again, you know, Postmates focusing on San Francisco, the city, DoorDash focusing initially on the suburbs, much smaller opportunity, but one that they could really knock the cover off the ball on. They had no competition. They had, you know, uh, a, a customer that was desperate for attention, and so they can make them happy, and that just sets you up for, for real success to go from strength to a- the adjacent market. The s- the second thing that I'd say is, uh, and this is so important, is that not all markets are susceptible to tipping, right? Like, there are... We've all experienced this, I'm sure, uh, Lenny, with, with your, with the companies that you work with, companies I work with, where-There are conditions of a market that make it vulnerable to tipping. I have six, um, in my post, just to name a, a couple, uh, classic ones. You know, the f- the first obvious one is just concentration on the supply side. In order to be able to tip a market, you need suppliers to want to lean in on your marketplace, right? My partner, Bill Gurley, talks about how great marketplaces create the new incumbents. So you basically have suppliers who aren't the incumbents. The incumbents aren't gonna lean in on their, on your marketplace. They're already fat and happy. But you wanna have these hungry suppliers who lean in, right? But if there's already concentration in your marketplace, in the ecosystem of the market that you're playing in, then you're not gonna have sellers lean in and you're not gonna be able to create a, a flywheel that spins where the more sellers you bring on, the better the pricing or experiences on the demand side because the sellers are happy and there's nobody that's hungry or fighting. And, and so, that's just, like, a critical, critical point.

    3. LRLenny Rachitsky1:14:14

       Bill Gurley and I actually had a little Twitter debate once about supply versus demand and which one's more important, and he had this... Basically, his pitch was like, all that really matters in the end to build a successful marketplace is can you find the demand, can you bring the demand to your marketplace. Makes all the sense in the world. Like, everything in the end is can you find customers. In the work I've done looking into marketplaces, what I find is the hardest thing to do is to bring people to your marketplace, is to find the supply and bring it, it to people, basically aggregate it. And your point here is exactly right, is that basically, like, the value of your marketplace is to find this non-concentrated demand that's all over the place and bring it together and make it easy to transact with. So I think that is really important, just like to, to double down on this point that what makes a marketplace valuable is supply that is hard to find that you make easy to find and transact with, and I feel like that's almost the, the core of what makes a marketplace work.

    4. STSarah Tavel1:15:12

       I think you're both right, uh, because-

    5. LRLenny Rachitsky1:15:14

       Yeah, I think we're both right. We're both right.

    6. STSarah Tavel1:15:15

       ... because in the beginning-

    7. LRLenny Rachitsky1:15:16

       We're talking about different things.

    8. STSarah Tavel1:15:17

       ... in the beginning, the only way you get to, yeah, the only way you get started with a marketplace, obviously, is you're bringing the supply online. And that's what opens up the opportunity. And the only way then for you to have a chance at tipping the market is if there's enough fragmentation on the supply side to create a differentiated experience for the demand side to care about your marketplace. But the only way to build longterm enduring value with a marketplace is to corner the demand side. And so it's kind of a little bit that evolution of a marketplace that, that ends up changing who, uh, who you focus

21. 1:15:55 – 1:20:29

    The challenge of homogeneity in B2B marketplaces

    1. STSarah Tavel1:15:55

       on. And I don't know if you've experienced this. In my experience with the marketplaces that I work with, it's a constant, like, "I have to focus more on the demand side right now. Okay, now we gotta focus more on the supply side." And it's just back to how difficult these businesses are to build, especially when you have limited resources, that it's like the forever battle, uh, of building a marketplace.

    2. LRLenny Rachitsky1:16:15

       Yeah, especially if it's a geographically oriented marketplace like a DoorDash. Sometimes you have more customers than restaurants, sometimes you have-

    3. STSarah Tavel1:16:22

       Right.

    4. LRLenny Rachitsky1:16:22

       ... too many restaurants, not enough customers.

    5. STSarah Tavel1:16:24

       Yes.

    6. LRLenny Rachitsky1:16:24

       So it's very city dependent. Airbnb, we had basically many attempts at a dashboard of which market is supply constrained, which is demand constrained. Along the same lines, I think it's also something I think about with B2B marketplaces, the reason they're challenging, in my experience to work, there's so few of them, is the supply is rarely fragmented. Usually, there are very few suppliers, so it's pretty easy. I'm just gonna pick one of these ten and I don't need you. Is that, is that true? Is that how you think about the challenges of B2B marketplaces?

    7. STSarah Tavel1:16:53

       There's definitely that, but I think actually the bigger challenge with B2B marketplaces is on the homogeneity of the supply. You know, when you think about a lot of labor marketplaces, and like, let's take a classic example like Mechanical Turk, right? What you need in a marketplace for that flywheel... You know, I, I have a graph in my, in my, uh, in my slides where you kind of show that the more you penetrate the market, the happier you're able to make your customer. That's, you, you should have an exponential experience where the flywheel doesn't slow down. Now you take Airbnb as an example, you can't... There's, there isn't a limit to how valuable it is to have more supply on Airbnb because everybody has their own special, you know, preferences and there probably is a theoretical limit, but it like... You just feel like there is always gonna be the bene- There's such heterogeneity in the supply for Airbnb that that's what has let Airbnb really close and escape competition. With B2B marketplaces, what sometimes happens, like a labor marketplace, is that you have this homogeneity in the supply where it's not clear that adding the next incremental unit of supply actually changes the experience for the demand side. And so at Mechanical Turk, whether Mechanical Turk has 100,000 turkers or five million turkers, it probably doesn't really change the experience for somebody who is using Mechanical Turk. And that's why in the early days of Mechanical Turk you saw a lot of other copycats or clones of Mechanical Turk doing something similar, because it didn't take that much supply for them to be able to create an experience for a segment of the market that was as good as the already scaled Mechanical Turk.

    8. LRLenny Rachitsky1:18:55

       Awesome. I'm glad we got into that. We got real- We're getting real nerdy about marketplaces, exactly what I was hoping for.Uh, you said that you had something else you wanted to share along the, these lines.

    9. STSarah Tavel1:19:03

       There's, and there's, you know, so many other examples of reasons why a marketplace will tip competition. Like, you know, again, you go after New York's... You're a door, you're a food delivery company and you go after the suburbs where there's no competition, that makes your job really easy to tip the market. You go to New York City where you have an entrenched competitor, uh, Grubhub plus Seamless Web, a different culture around delivery where the restaurants mostly have their own delivery fleets, and it's just gonna be so much harder for you to tip that market, right? And so, being, I think you have to be really context-aware when you're building a marketplace of what your competition's gonna be, so that you see if there is a path for you to, to actually tip the market.

    10. LRLenny Rachitsky1:19:52

        And even if you've tipped a market, it may get un-tipped, like Uber comes in and starts to eat your lunch. No pun intended, right? Like, it doesn't last forever. You have to keep fighting for it.

    11. STSarah Tavel1:20:00

        Well, and that's, it's such an important point, which is that you can never rest on your laurels when you're building a marketplace. Like, the history of marketplaces is one in which you have a marketplace that feels dominant and then gets disrupted by, by a competitor. HomeAway, VRBO being disrupted by the likes of you, uh, at Airbnb.

    12. LRLenny Rachitsky1:20:23

        I wish I could take credit for all that.

    13. STSarah Tavel1:20:24

        (laughs)

    14. LRLenny Rachitsky1:20:25

        Yeah. So one way I think about, as you're talking about this idea of

22. 1:20:29 – 1:21:43

    Signs you’re tipping successfully

    1. LRLenny Rachitsky1:20:29

       tipping, is a simple way to think about if tipping is happening, if you're tipping successfully is percentage of your growth that's starting to come from organic word-of-mouth, people just becoming, like, "This is the default way I'm gonna order food, travel, book a ride."

    2. STSarah Tavel1:20:43

       I think you see two things. You see, one, the cohorts getting better. Um, so people like, the people start using you more, right? Like, as your, as the supply side gets more diverse, you have more and more use cases that you start to, to be able to serve. And so that, you know, you get the demand side leaning in more, the cohorts get better. And then, to your point, you have the organic growth that happens where you don't have to fight to get each new participant on both sides of your market. There's something that starts to happen where the value proposition is just so strong, relative to any other substitute, that the market starts coming to you.

    3. LRLenny Rachitsky1:21:27

       Awesome. So essentially, cohort retention starts going up 'cause it's more and more valuable, and more people are coming to you just word of mouth or through some loops that you've built.

    4. STSarah Tavel1:21:36

       Exactly.

    5. LRLenny Rachitsky1:21:37

       Awesome. Anything else on this layer/level of the hierarchy before-

    6. STSarah Tavel1:21:42

       No.

    7. LRLenny Rachitsky1:21:42

       Okay, cool.

23. 1:21:43 – 1:28:29

    Level 3: Dominating the market

    1. LRLenny Rachitsky1:21:43

       Let's talk about dominating the marketplace and what that looks like and how you, how you do that. How do you dominate a marketplace and when? Sarah Tavel.

    2. STSarah Tavel1:21:51

       So yes, level three, and this is the one I feel like every, you know, every founder who starts a marketplace like they're chomping at the bit, right? To be able to really focus now on growth. 'Cause you know, all along I feel like it's been, we've been holding you back of just like focus, focus, focus. And, and what you're doing through level one and level two as you, as you're honing what you're, you're building, is that you're basically building a playbook. You don't yet know whether this playbook is repeatable, right? Like it's, it's not always the case that the, the dynamics of one market, uh, is the same elsewhere. I'll take Reki as an example. In the London food scene, there's like significant fragmentation, uh, of suppliers. In the Berlin food scene, that's not the case. And so, it's not always the case that whatever you did in one market works in another market, but usually there's something that rhymes, there's a hint of what, of what is gonna work. So what happens now is you, you have a market that's tipping and the question then is, are you ready to take your eye off the ball and start to diffuse your focus into other markets? You know, so there's, there's kind of three vectors then that any marketplace can grow at this point. The first vector is within the existing market where you already are tipping that market. You as a founder then have to make a very, kind of, very personal decision that's, that what you're looking at is the context of your competitive situation. So, if you're tipping a market and you have no other competitor going after that market, you're the first one to see this opportunity, then I think you have two things that you wanna do at the same time to grow. The first thing is that you wanna take the existing market that you're in and you just keep doing what you're doing and, and you may even find ways to stretch beyond that initial, uh, market that you're going after within, within the same market. And so what I mean by that is, you know, Uber went from black cars to Uber X to Uber Pool. You know, they kept on finding ways in the geographies where they were already dominant to keep on getting stronger by expanding the use cases that they serve. So you have continuing to grow and penetrate the market, find ways to solve, to kind of answer more use cases in that market that you can do especially if you don't have a lot of competition. And then the third vector of growth is you wanna then try to get as many plates spinning in as many markets that you can handle as quickly as possible. And this is like the blitz scale, right? This is the land grab. And again, your equity value that you're gonna create is gonna come from dominating the market. And so, you don't want to plant a thousand flags in a thousand markets at the same time, spread yourself too thin, and not decidedly-... win any one market. You're trying to put, you know, you always wanna put more wood behind fewer arrows. The more scale that you have, the more cities or categories where you have, you know, the, the, the flywheel's starting to spin and you're starting to see the tipping point happen, the stronger your company will be, right? Each market that you get, you have scale in, where you're s- you'll have contribution profit. You take that contribution profit, you invest it in new markets. The more markets and contribution profit you have, the more venture capital you'll be able to raise, which you then reinvest in growing. And so, this is really the place where you are as aggressively as possible while at the same time not losing sight of the fact that your goal is, in each market individually, to dominate that market. You're trying to grow into as much possible market opportunity as you possibly can.

Episode duration: 1:50:54