The rituals of great teams | Shishir Mehrotra, Coda, YouTube, Microsoft

1. 0:00 – 4:13

   Teaser

   1. LRLenny Rachitsky0:00

      I generally value the reference check over interview signals. You know, if I had to stack rank in interviews what is the best signal, the, uh, reference check is the top of the list. Tho- those people, you know, they work with this person for, you know, sometimes for years. Like, their knowledge ... What you're gonna get out of 30 minutes of artificial scenarios is just, like, never gonna compare with what a good reference check will give you. (instrumental music) Shishir Mehrotra is the co-founder and CEO of Coda. Before starting Coda, Shishir led the YouTube product engineering and design teams at Google, where he spent over six years. Before that, he spent six years at Microsoft. He's also on the board of Spotify. As you'll hear in this episode, Shishir is an incredibly deep and very first principles thinker on all kinds of topics. And in this episode, we cover growth strategy, specifically a framework that he calls blue loops and black loops. We talk about the rituals of great teams, something that Shishir's been passionate about and has been collecting from all of the best leaders in tech for the past two years, and which he'll soon turn into a book. We talk about Eigenquestions, which is not a German game show. He shares how he evaluates product talent and s- gives some really great advice on doing reference checks. We go into so many other topics. This is the longest episode that I've recorded yet, and you'll see why. Shishir is so full of wisdom, and we could have kept going for at least another hour. And so with that, I bring you Shishir Mehrotra. Hey, Kasey Winters, what do you love about Coda?

   2. NANarrator1:29

      Coda's a company that's actually near and dear to my heart, because I got to work on their launch when I was at Greylock. But in terms of what I love about it, you know I love loops, and Coda has some of the coolest and most useful content loops I've seen. How the loop works is someone can create a Coda and share it publicly for the world. This can be how you create OKRs, run annual planning, build your own map, whatever. Every one of those Codas can then be easily copied and adapted to your organization without knowing who originally wrote it. So they're embedding the sharing of best practices of scaling companies into their core product and growth loops, which is something I'm personally passionate about.

   3. LRLenny Rachitsky2:05

      I actually use Coda myself every day. It's kind of the center of my writing and podcasting operation. I use it for first drafts, to organize my content calendar, to plan each podcast episode, and so many more things. Coda is giving listeners of this podcast $1,000 in free credit off their first statement. Just go to coda.io/lenny. That's coda.io/lenny. Hey, Ashley, head of marketing at Flatfile. How many B2B SaaS companies would you estimate need to import CSV files from their customers?

   4. NANarrator2:38

      At least 40%.

   5. LRLenny Rachitsky2:39

      And how many of them screw that up? And what happens when they do?

   6. NANarrator2:43

      Well, based on our data, about a third of people will consider switching to another company after just one bad experience during onboarding. So if your CSV importer doesn't work right, which is super common considering customer files are chock full of unexpected data and formatting, they'll leave.

   7. LRLenny Rachitsky3:02

      I am 0% surprised to hear that. I've consistently seen that improving onboarding is one of the highest leverage opportunities for both sign-up conversion and increasing long-term retention. Getting people to your a-ha moment more quickly and reliably is so incredibly important.

   8. NANarrator3:16

      Totally. It's incredible to see how our customers like Square, Spotify, and Zuora are able to grow their businesses on top of Flatfile. It's because flawless data onboarding acts like a catalyst to get them and their customers where they need to go faster.

   9. LRLenny Rachitsky3:33

      If you'd like to learn more or get started, check out Flatfile at flatfile.com/lenny. (instrumental music) Shishir, welcome to the podcast. Thank you for having me. I don't think I've actually shared this with you, but you're actually the very first CEO that I've had on this podcast. I actually have a rule of no founders or CEOs on the podcast, at least at this point. And you're the first person that I've let break this rule. And so how does that feel? (laughs) I've always been a rule breaker. (laughs) Perfect. Uh, so bio for listeners, just briefly. Uh, so you're the founder and CEO of

2. 4:13 – 7:53

   Shishir’s background at Google and current role at Coda

   1. LRLenny Rachitsky4:13

      Coda. You spent six years at Google, where you were a VP of product and engineering for YouTube, basically leading the YouTube product team. Spent six years at Microsoft. You're on Spotify's board of directors. You're also a prolific online writer. And that leads to my first question, which is, I hear that at Coda, there is a contest internally for people who ... Actually, so maybe a little context. You encourage people to write a lot of stuff externally within Coda. You want people to be writing. And you have this contest of who gets the most views and likes of people internally. And so is that true? And who's winning? B- by the way, yes, it, it, it's, it's sort of true. I mean, we, we did a similar thing at, uh, at YouTube. And we, and YouTube creators, I mean, obviously kicked our butts. But we, it was a good way to make sure we understood our tools and learned how it worked. And, you know, I think for a while, at YouTube, I had one of the top videos. It was a, a really cute video of my daughter taking everybody to orders when she was like three or four years old, and- That's not fair. It was, uh, yeah, super cute kid is a, is a, it was a easy trick for YouTube. (laughs) But I get to learn all the tools and so on. You know, here, the equivalent at, at Coda is you can publish Coda docs. They show up in our, in the gallery. You go to, you know, Coda has this gallery. You can see lots of them. And, um, you know, at this point, thousands of docs have been published from lots of different people. It gets millions of views and li- you know, like at YouTube, the most popular ones are not ours. But it is sometimes helpful for us to make sure that we understand how the whole system works in order to do it. So I think that the current winning doc at Coda is, um, is one in a pretty deep niche. It's from a guy named Kenny Wong on our data team. And it's an Orange Theory workout doc. So it turns out that Orange Theory has this, like, deep subculture.

   2. SMShishir Mehrotra6:00

      ... that they all hang out together on Reddit and so on. And this doc just took off as a

   3. LRLenny Rachitsky6:06

      Mm-hmm.

   4. SMShishir Mehrotra6:07

      ... sub culture. I don't really understand the doc 'cause I'm not in that sub culture, but kind of it's similar to YouTube in some ways. It's like that the-the-the niches are actually much bigger than-than people think. You know, of the ones that are, right, you know, more work-related, I think Lane's doc on two-way write-ups still outranks all of mine, but, uh, I have a couple good ones on there too. And I would say this competition exists in my family too, and I don't usually even win with my own family. My older daughter, Annika, who also had that great video on YouTube, has a... has two docs that do really well. One is Family Quarantine Olympics, which is a thing she put together at the beginning of COVID, like a fun game for families to play, and the other one is a score tracker for a game called Ticket to Ride, which I don't know if you ever played the game-

   5. LRLenny Rachitsky6:49

      Mm-mm. Heard of it.

   6. SMShishir Mehrotra6:50

      But it's like, it's the most, the most complicated scoring system on the planet 'cause like you play the whole game and you spend 10 minutes scoring it, and she built this whole scoring calculator and that turns out to be super popular too. And so anyways, but my-my docs do okay, but yeah, some of the kind of interesting variety of what people that are doing.

   7. LRLenny Rachitsky7:06

      I love that. Which of your docs has been the most successful? Hopefully we end up talking about it.

   8. SMShishir Mehrotra7:12

      I-I think it goes back and forth between two docs I've written, one called Eigenquestions, which I-I think you intend to talk about, so we'll get back to that.

   9. LRLenny Rachitsky7:20

      Yeah. Absolutely.

   10. SMShishir Mehrotra7:20

       And the other one is one I wrote a while back called Four Myths of Bundling, um, that's all about how subscriptions work and it's kind of how I ended up on the board of Spotify, was Daniel and I geeking out on bundling theory, which is like a super weird hobby, but, you know, I have normal hobbies too, but I like to, I like to explore bundling as a, as a kind of fun hobby and-and people enjoy that doc as well.

   11. LRLenny Rachitsky7:41

       Yeah. You've shared a lot of good thoughts on that and we're not going to cover that 'cause you've covered that in depth in many places. But just to clarify, did you write that doc and that led you to being board, on the board of Spotify, or is that after the fact?

   12. SMShishir Mehrotra7:52

       The conversation

3. 7:53 – 8:58

   How Shishir got on the Board of Spotify

   1. SMShishir Mehrotra7:53

      led to-

   2. LRLenny Rachitsky7:55

      Okay.

   3. SMShishir Mehrotra7:55

      They-they... So it was, it was a napkin sketch, um, at YouTube that turned into a really fun lunch I had with Daniel, Daniel Ek, the Spotify founder and CEO, and then he encouraged me to write it down. And for me, I mean, I don't... You write prolifically and I-I have a, I have a... Writing for me is actually surprisingly hard, okay? I feel like I have to think about it all the time.

   4. LRLenny Rachitsky8:20

      Same.

   5. SMShishir Mehrotra8:21

      Like (laughs) you make it seem really easy, but the, uh, um, you know, for me, like, he's like, "Could you write that down?" And it's like, "Great, now I'm gonna, like, take a year to write this thing down." Because you, like, you, like, think through each part of it and you kind of come up with the right framing. I have a little review process I use for my docs that, that allows other people to kind of help me make it better, which is, uh, which is always really helpful. But yeah, so that's how that started. So it got-got written down after. Yeah.

   6. LRLenny Rachitsky8:47

      So we were talking about writing and content and things like that, and that's a really good segue to the first thing I wanted to talk about, which is black loops and blue loops, which to folks that haven't heard this before might sound

4. 8:58 – 9:52

   Black loop and blue loops and how Coda uses this internal diagram

   1. LRLenny Rachitsky8:58

      like some kind of ultimate fighting nightmare scenario, but it's something that is really important to you and Coda. And so to set it up, can you just talk about what black loops and blue loops are?

   2. SMShishir Mehrotra9:08

      Yeah, and it's probably worth mentioning, I think a lot of businesses have a diagram that describes their ecosystem and-and how it works. Sometimes it happens a little later in a, in a company's journey. For us, we were probably three or four years in before, uh, Matt Hudsons, uh, um, runs our data finance teams here, he came up with this, this diagram and-and it really stuck for people. But I h- highly encourage drawing a diagram like this for your business. I'll flash it up on screen for a second and I'll describe it, but this is, this is what the diagram looks like. Um, black loop, blue loop, and it's basically the two different ways that our product spreads. The black loop is someone comes in, they make a doc, they share with a group of people, some sub-subset of the people turn around and make another doc, and the process repeats itself

5. 9:52 – 12:15

   The black loop is how a product is naturally shared

   1. SMShishir Mehrotra9:52

      over and over again. The blue loop is someone comes in, makes a doc, and instead of sharing it with, with the team or, or with the collaborators, they publish it to the world. And in that process expose it to, they can choose how it should appear. What publishing in Coda is a lot like building a website. So you pick a URL, you tell us whether or not Google should be able to find it, show up in the gallery and so on. And what-what ends up happening from that is they turn into broad promotion of Coda, but really it's about that person, what they're trying to get, trying to get done. And I'll stop sharing so I can talk a little bit about the dynamic. So the-the... I-I sometimes refer to them as the, the Microsoft loop and the YouTube loop because those are the, like, two inspirations for it. So the black loop feels a lot like how documents naturally spread, like the viral actions of a, of a document platform are share, create, share, create. It happens over and over and over again. You know, the best way you learn about Office or Google Docs or so on is somebody shares one with you and you're like, "Oh, that's pretty cool. I bet I could create one." And that loop can happen very, very quickly. And it really drives, for us, a lot of how we think about how we work mostly within companies and teams, but sometimes across them as well. And so for an example, it led to our pricing model. So our pricing model is, um, a little different than most, most companies in the space that we do a thing called maker billing. So basically all, all document products or all products with a document metaphor have, have three personas, people who can see things, people who can change things, and people who can create new things. Basically everybody charges for the top two, they charge for editors and makers. If you can make changes, then you have to pay. And that's like every document product you can think of, including ones, you know, if it's drawing or something, they all, they all kind of do the same thing. And we decided that we're only going to charge for people when they make a document. So the, the, uh, if you think about it, you get a Coda doc, only, you only need one, you know, if you're using any of our paid features, you only need one paid license, uh, for doing it. And the reason we do that in terms of that diagram is I wanted no friction on the share edge. I mean, the share edge for us is like that's the moment of, "Hey, look, I-I'm doing this thing. It's so cool." Like, and-and that's the moment where the line I gave to the team is, "I want no dollar signs in the share dialogue." Going to that, that's your... Every product has its moment of how you, how you

   2. NANarrator12:03

      Yeah.

   3. SMShishir Mehrotra12:03

      How you

   4. NANarrator12:03

      Yep.

   5. SMShishir Mehrotra12:03

      ... spur growth and, like, you know, imagine, going back to YouTube, imagine you had to pay for people you shared with, like, nobody would ever share anything. But that's how basically every productivity product works is the moment they charge you is when you share

6. 12:15 – 14:55

   The blue loop is the emotional loop on why products are shared

   1. SMShishir Mehrotra12:15

      with somebody.The blue loop I, I often call the YouTube loop because the emotions of publishing a doc are, are, are incredibly close to that of publishing a YouTube video. And people have all sorts of reasons why they do it. I mean, sometimes people do it, um... You know, there are people who do it for money, but a lot of people do it for exposure, for brand building. They just want to get an idea out in the world. They wanna get feedback. Some people do it for fun. Some people do it as a, as a charitable contribution. There's lots of reasons why people do it. But the net effect of what happens is... You know, for YouTube, the vast majority of how people found out about YouTube was through a video that was shared with them, and that's, that's sort of the, the, the, uh, the, uh, impact. But what it changes the dynamic that, uh, allows everyone who publishes a Coda doc, now has a very natural incentive to go share it with the, with the right population. You know, if you're, if you're into Orangetheory, you share it with the Orangetheory population. If you're into, into playing Ticket to Ride, you share it with the Ticket to Ride population. But the, um... You know, if you're into bundling, you try to find a, you know, small group of people that care about bundling (laughs) and tell them about that. But the, um... And, and what happens for us is that then becomes a loop that, um, means that most people's exposure, um, uh, and almost a third of our users come through this, this loop. They're not actually exposed to Coda. They're exposed to a great idea for, you know, how to run an off-site or how to, how to, you know, how to win Ticket to Ride or whatever it might be. And in that process, they learn about the product, and so then they come in, uh, through this vehicle. And, you know, one, one reason it's very important is 'cause what... For products like ours that are very horizontal, you get, you know, you get, um, you get different types of users. There's some users that I call the building block thinkers. They like to kind of build up from scratch, and the blank surface of Coda is really amazing for them. But for most people, that's intimidating, like, "I don't, I don't really know what to do." Most people in the world are, are, are problem-solvers, and so they start not by, "Do I need a new document?" They start with, "I've got a problem." You know, my... "We don't make decisions fast enough at our company," or, you know, "My family can't figure out what to do on the weekend," or whatever it might be. And then when they find a solution to that problem, they then pick the right tool. And so the blue loop allows us, uh, to go after that, and it kinda changes, changes how the motion of the ecosystem works as well. But that's what blue loop and black loop are.

   2. LRLenny Rachitsky14:31

      Awesome. (clears throat) I have a bunch of questions I wanna ask. The first is, how did you... How did... For a founder listening to this and they're like, "Oh, man. What are my, what are my loops? What's my flywheel? How do I think about my business?" Can you talk a bit about how you kind of came upon this way of thinking about the company? And then also, how do you structure your teams to kind of work in this way if this is the way you're thinking about growth?

   3. SMShishir Mehrotra14:50

      Maybe on the first part, uh... And I think you just hosted Casey Winters. He's like, is, like, pretty famous

7. 14:55 – 18:20

   Why you should think in loops instead of funnels

   1. SMShishir Mehrotra14:55

      for talking about loops not funnels. And I do think there's a, a very natural thing when you're building a product or building a business to think about your funnel. And you think about things as being linear, that somebody comes in. You know, they go up through your sign-up process, and then they, they see your onboarding, and then they get exposed to the first magic moment of your product, and the second magic moment, so on. But the truth is, it doesn't really work that way. I mean, it... A- almost all products have some form of loop. I mean, that, that person turns around, and maybe sharing is built right into your product or maybe it's not, or maybe there's a way that it happens through advocacy. And, and so by understanding the, the way products actually grow and spread happen through some type of loop, not funnel, is, I think, pretty fundamental. So the first piece of advice I'd give is, you probably do have a loop. I mean, I, the... Whatever the product is, there's probably something about it that causes that loop. And understanding how that, that works is really important. I mean, in terms of, like, wha- what it is, you know, in our case, I'd say if you take black loop and blue loop. You know, the black loop is... You know, every product in our category has that loop. I mean, it's not... It's actually... We didn't invent it. I mean, you, you, you, you build a document. You put a share button on it. You know, e- e- every product has that loop. Sometimes it's just recognizing what's there. I mean, it's not, it's not that, it's not that interesting. The blue loop, on the other hand, is not something that every product in our category has, and it's not really a thing that you expect to do with Google Docs or Office or so on. It's a... It's kind of our unique take on, "Hey, we're gonna build a publishing platform that isn't just for sharing ideas and building things with your team, therefore putting things out in the world." I mean, one of the, the best compliments we hear about the Coda gallery is... I had this user tell me this line I really love. He said that the Coda gallery feels halfway between Medium and an app store. And you can come and you can read about anything interesting in the world, and you can go shopping and say, "I need one of those, I need one of those, and I need one of those." And it's my view that this category, what we call the All-in-One Doc category... I think this is gonna be critical, 'cause I, I don't think that there's enough people out there that are looking for, like, a horizontal new blinking cursor. I mean, they exist, and you can get through your first million users that way. But I think to get to, like, the level of, of impact we wanna have, we've gotta find this problem-se- uh, seeker. So you probably have a loop, not a funnel. And it might be hiding in plain sight or it might require invention. That's, that's the Bounder dance and the, you know (laughs) the Find where it is. But, but, uh, you know, finding it, writing it down, I think, is really helpful. Um, it's a second part of your question? Oh, how do we organize the team?

   2. LRLenny Rachitsky17:18

      Let's... Yeah, but let me ask one quick question. So you said this, uh-

   3. SMShishir Mehrotra17:20

      Yeah.

   4. LRLenny Rachitsky17:20

      ... I think it was a data scientist that kind of first imagined-

   5. SMShishir Mehrotra17:23

      Yeah.

   6. LRLenny Rachitsky17:23

      ... and kind of diagrammed this out? 'Cause I'm, I'm curious as... You know, as a founder listening, they're like, "Oh, how do I, how do I find someone like this?" I imagine part of this was, oh, this person brought you this interesting way of thinking, and there's kind of this process of, "Oh, wow, this is cool. Let's think about this." What was that like, kind of just over high level, that process?

   7. SMShishir Mehrotra17:40

      He currently runs data and finance for us. He's actually one of the early founding, uh, members of the team, Matt Hudson. So he's kinda handed out every job here around the go-to-market team-

   8. LRLenny Rachitsky17:47

      Mm-hmm.

   9. SMShishir Mehrotra17:48

      ... and marketing duties. Very insightful. But honestly, the idea can come from anywhere. I mean, it's, uh... You know, there's a really famous loop diagram for, for Uber that I think, I think one of their board members drew it or Travis drew it. I don't (overlapping) Yeah, right.

   10. LRLenny Rachitsky18:01

       ... like an napkin.

   11. SMShishir Mehrotra18:02

       Yeah, the napkin sketch. You know, who knows how true that is? (laughs) The, uh, probably lots of dispute o- on that.

   12. LRLenny Rachitsky18:07

       Mm-hmm.

   13. SMShishir Mehrotra18:08

       Um, uh, I think Airbnb had a similar diagram. I'm trying to remember.

   14. LRLenny Rachitsky18:12

       You know, we, we tried, you know.

   15. SMShishir Mehrotra18:13

       Yeah.

   16. LRLenny Rachitsky18:13

       We had some-

   17. SMShishir Mehrotra18:13

       (laughs)

   18. LRLenny Rachitsky18:13

       ... we had some sketches.

   19. SMShishir Mehrotra18:15

       Right, right. So the, um... You know, it can kinda come from anywhere. I do find that...The

8. 18:20 – 24:37

   Mining for your business’s loops by looking at what you tell job candidates

   1. SMShishir Mehrotra18:20

      most natural place to, to see it is just when you're talking about your business to someone you're pitching, a customer or a candidate. So I actually think candidate, they, you know, I think we're gonna talk a little bit about energy, but I think talking to candidates is one of the best ways to hone what your business is about. Because those, those people are, in some ways, even more critical than investors. I mean, they're investing their, they're investing their time, not just their money. And, and so your ability to get across to them why this thing is going to be interesting and how it'll grow, and they're kind of the most discerning investors out there, in, in a lot of ways. And they, and they're actually not that easily confused by metrics and so on that can be temporary and, and, you know, they're, they're, they, you know, they had, they sort of put themselves in that picture of like, "Can I see that happening?" And so for us, like, the black loop part is pretty obvious. For the blue loop part, like you had to, like, squint a little bit to think, you know, "Will people really do that? Will people come and, you know, publish these documents," like, you know, some hybrid between websites and blog posts and templates. Like what, what, you know, what are they gonna, what are they gonna do and why? And, and so it required a little bit of creativity, which forced me to get better and better at pitching why that's gonna happen and, you know, what, what that role is gonna feel like. And, you know, this analogy of halfway between medium and an app store is like, that will, like, help people criti- uh, crystallize what that, what that promise has to feel like. So I think that the, the initial idea can come from anywhere, but if you want to mine for your own loops, go look at what you told the last few candidates you talked to.

   2. LRLenny Rachitsky19:48

      I like that, and it just takes some, takes some attempts at drawing some kind of diagrams. That's kind of how I thought about that-

   3. SMShishir Mehrotra19:53

      Right.

   4. LRLenny Rachitsky19:53

      ... when I was thinking about it. Do you find that the quality of user is different amongst the loops? I imagine one is, like, 80% of the growth, but maybe the other is a different type of user, maybe higher quality. I think about it a little bit with Airbnb referrals drove higher quality hosts, even though it was still a small portion of all hosts, and so it was a really lucrative and interesting channel. Do you find anything like that?

   5. SMShishir Mehrotra20:14

      So, I mean, quality and activation are a little bit different. I mean, they, they, they... Blue loop definitely the... So it's actually kind of three entry points on that diagram. Those people come through the blue loop as people get shared through the black loop, and then people come through the top of the funnel. Like they, they have to be, like, your, your, your seed population, right? They, they, they have to... Somebody starts with blinking cursor. Nobody shared anything with them, either a blue loop or templated document or, or black loop a- and, you know, the way the team is running. You know, if you look at activation retention, so on, certainly the, the, the highest is the, the black share loop. I mean that, that... If you kind of... Somebody shares a document and say, "Hey, this is how we're running this app meeting," you're kind of, you're just gonna use it. So the, uh, the job of retention there is, um, is a little different. And actually one, one thing that is... Actually, let me come back to that observation. S- Second, second best is through the blue loop, and then the third, you know, the worst, the hardest is activating through the very top. And, you know, from, from there, you know, s- roughly one in five people make it to what we think of as our activation moment, which is hard, right? It's like you're gonna hand somebody a new product that they didn't start with a problem on, and nobody handed them a document to say, "Just work in." Like, that's, that's hard. Now you gotta... Now all our next flows are really important and so on. But if you think about these three different dynamics from how you, how you- and you asked about how you structure your team, they're like incredibly different mindsets, right? Because the, you know, coming to the top of that diagram, we get to own the conversation. Like we, we tell you, we, you know, we, we have our opportunity to tell you what the product's about, what you should do with it. "Here's the, here's the minimal set of things you need to know in order to be productive," and gradually reveal the other things that you might need to know. Getting the order of those things wrong, real trouble and so on. But it's actually a minority of how people get exposed to Coda. Because in the black loop, you know, the person who owns that conversation is the person sharing the document with you. Like, if that person does it and, like, mispositions it or doesn't, like, just makes a crappy document or so on. Like, you know, we can't, we have to, like, help them onboard their users, which become our users.

   6. LRLenny Rachitsky22:11

      Mm.

   7. SMShishir Mehrotra22:11

      And the same thing in the blue loop. You know, that conversation is now owned by the publisher who's, you know, they've, they've, they're really not that interested in teaching anybody about Coda. They're mostly interested in, "Here's this, like, really cool way to do Orange Theory," or, "Here's this, here's this interesting way, you know, to run a meeting." And, and so one of the interesting things about building platforms, which I think is a little bit different than, than, uh, products that, that get to be sort of direct. I mean, I just, uh... For better or worse, most of the products I've gotten to work on are platform products. And, and I find that there's two very different kinds of people that like that challenge. So some people, and I think Steve Jobs was the, the, the quintessential example of he didn't really like being a platform. Like, you know, the, the iPhone shipped without an app store. You know, they locked down the screws on the back of the, on, on, on all the devices so that nobody could open them. And, you know, the- his viewpoint was, "I'm gonna control every element of what my users see." And on the other hand, you know, platform thinkers, you, you sort of assume that my connection to my eventual user is through someone else, like YouTube. You know, regularly I come to U- day, I come to work at YouTube and somebody would say, "Well, here's what happened last night," you know? And, and, uh, sometimes it's heartwarming. Like, "Oh my gosh, this person, you know, this kid bit this other kid's finger and it took off like crazy." And, you know, this, this, uh, Korean pop star just, like, broke through the billion view mark before everybody else did. Uh, and sometimes it was not heartwarming (laughs) like, and you don't get to control that 'cause that's, you know, part of, part of being a platform. And so it does change, you know, how you think about the, the way you run the team, because if you have a loop where your community ecosystem users so on control that narrative, then you have to, you have to in- incorporate that. I mean, another close analogy I think is like, you know, Airbnb and the famous story of them, you know, taking pictures of people's apartments. It's like, you know, they had to, they had to reach out and try to control that. And, and eventually you can't do that and you have to, you had to sit back and let people, you know, market their, market their hotels. And, you know, thankfully the ecosystem got good at it, but it kind of has a similar dynamic, I think.

   8. LRLenny Rachitsky24:12

      Absolutely. With Airbnb, pricing is even more of a challenge where a lot of hosts don't really know what to price. They think their place is worth a lot more and we can't tell them the price. There's laws around that. And so it's like-

   9. SMShishir Mehrotra24:23

      Mm-hmm.

   10. LRLenny Rachitsky24:23

       ... "Hey, maybe you wanna price it at this rate."You know what's good for you. So it's a lot of, uh, encouraging. So yeah.

   11. SMShishir Mehrotra24:29

       Yeah.

   12. LRLenny Rachitsky24:29

       I've seen that in action. So you talked about teams and how you think about structuring them a bit. And that's a good segue to our second topic, which is around a book

9. 24:37 – 26:30

   Shishir’s upcoming book The Rituals of Great Teams

   1. LRLenny Rachitsky24:37

      that I hear you're writing called The Rituals of Great Teams. And I think what you're doing there is exploring rituals that have emerged at some of the more successful companies. And so just a question there. When ... How'd you get interested in rituals, so much so that you decided to write a book, which is such a trudge and-

   2. SMShishir Mehrotra24:53

      (laughs) .

   3. LRLenny Rachitsky24:53

      ... endless amount of work? And just like how... Yeah, how was it and how's it going? And then I'll ask you a few more questions there.

   4. SMShishir Mehrotra24:59

      You know, the, the ... Writing a book. So writing this book, it's called Rituals of Great Teams, and the, uh ... When I signed up to do it, I thought it was gonna take six months. I'm now almost two years in. Uh, I-

   5. LRLenny Rachitsky25:09

      Classic.

   6. SMShishir Mehrotra25:09

      I owe my man- (laughs) I owe my manuscript in four months, and I, I'm probably half done, so there's a lot of work, uh, uh, ahead of us in, in, um, ? because it's one of the most fun projects I've ever done, so ... So the history behind this was a- as in many cases, is a lot of, you know, sort of odd luck and, and happenstance. I got hosted ... Right at the start of the pandemic, I was interviewed for a different podcast called Masters of Scale by Reid Hoffman. And the way Reid records, which is not... I'm not sure I would recommend this, but he does it a little bit differently than, than, than, than you do. You sit down with, you know, no idea what you're gonna talk about and you talk for like three hours, and then he has a group of editors, the same group that actually edits Ted, and they come in and they pick 20 minutes of it, and they turn into an episode. And you have no idea what it's gonna be. So you just talk and talk and talk, and he gets 20 minutes out. And they're pretty, pretty good at getting to a nugget. So they picked, out of this whole discussion, this part that I thought was like really small, and it was, um ... Reid had asked me for one of my favorite quotes, and I talked about this quote from a guy named Bing Gordon. If people don't know Bing, Bing was one of the founders of Electronic Arts. He's a famous investor, Um, Amazon, Zynga so on. Mo- lots of great companies. And I happened to sit on a board with Bing, and he used this line ... I think Bing's one of the best non-linear thinkers in the valley, uh, always learn something with, uh, with Bing. And

10. 26:30 – 27:10

    The 3 Golden Rituals Of Teams

    1. SMShishir Mehrotra26:30

       he, he used this line that really stuck with me. He said, uh, "Great companies has a very small list of golden rituals, and there are three rules of golden rituals. Uh, number one, they're named. Number two, every employee knows them by their first Friday, and number three, they're templated." And, and he has great examples. You know, Amazon has six pagers and Google has OKRs and Salesforce has V2MOM, and there's all these different rituals that, that, that people do. And I ended up sharing on the podcast a little bit about what Coda's golden ritual is. If you were to ask, uh, a set of Coda employees on the first Friday what Coda's golden ritual is, they would almost certainly tell you about this thing we do called Dory

11. 27:10 – 31:29

    Coda’s Golden Ritual: Dory and Pulse

    1. SMShishir Mehrotra27:10

       Pulse. It's, it's sort of the key of how we run meetings and do write-ups and so on. It's a pretty simple idea is that, in our write-ups, instead of, um, a- a- and in our meetings, instead of, you know, going around the room and hearing what everybody thinks, we, we do this thing called Pulse. Everybody writes down what they think and we hide everybody else's until you're done writing. So you kinda ... You force yourself to, to, to be eloquent about your opinion, go on the record about it, and unbiased. And then the second thing we do is called Dory, which is instead of s- randomly asking questions, we ask people to put the questions on the table, and then we take a round of upvoting and downvoting them to actually figure out what we're gonna talk about. Dory's named after the fish who asks all the questions. It's some- uh, it's a tool we use a lot at Google that we kind of turned into this, this mini-tool. If you were to ask a set of Coda employees on the first Friday about Coda's golden rituals with Bing's three tasks, they'll almost certainly talk about Dory and Pulse. And it's not because they're meeting wonks. It's because it's indicative of the culture of the team. And so the, the, uh ... You know, I'll regularly hear employees say things like, "You know, I just joined Coda. It's been a week. It's amazing. The culture is so open that I got to ask a question in a meeting and it outvoted the CEOs." Um, or they'll say, "I got asked for my contribution to this, this really hard decision we're making, and it was thoughtfully presented. I had space to be able to do it well without bias, and it was actually read and considered as part of the, as part of the decision-making process." And so this, this, the, the, the podcast on, um ... You know, Reid's podcast did pretty well, and I got all these questions about rituals, and so I decided to do a, uh, a dinner, which turned into a dinner series, and we basically ... Every third Wednesday, we would host a group of people to share their rituals with each other. And I learned a bunch of stuff in this- in, in this process. And I've, I've now interviewed over a thousand people for this book, and, and, you know, there's lo- lots of really interesting rituals that come out of it, but first thing I learned is people love sharing their rituals. I mean, it's not ... I, I get... I, uh, you know, I've interviewed people from, uh, many companies that everybody's heard of, you know, Nike, Disney, New York Times, you know, so on, all the way down to many of the, you know, many startups that maybe people haven't heard of or, or companies and industries that people don't, uh, talk a lot about. Um, book authors, uh, pundits, and lots of different people that have come through this process. Everybody loves sharing their rituals. I mean, everybody has a little secret to how they run their business, but for some reason, the how we work part, th- you know, everyone's very willing to share. People also love hearing about them, and I was kind of ... I was like, "Is this gonna be interesting for a dinner?" Like, you know, geeking out about how a team works? And it turns out, not only do people like hearing about it, it's the, the littlest details that matter. It's like, "Oh, yeah, we kinda do that too, but we have this issue. How'd you get past this?" And you start discovering that actually those little details are what make, what make or break a thing that, you know, you can't quite do it the exact same way. And then the other thing I realized, which is probably the most important point, is that rituals are ... I like to say that they are a mirror of culture. The, uh ... One of the attendees of the dinner is, uh, Dharmesh Shah, founder of HubSpot and very thoughtful, uh, person. He talked a lot about this thing that, that his ritual he presented as something called flash tags, which is a really, really, uh, cool example. But, uh, Dharmesh talked about how we ... When we're building companies, we actually build two products. We build one for our customers, and we build another one for our employees. And that's actually ... That how we work part of it, that's ... The term he uses for that is culture. That's the product we build for our employees. I think it's a very, like, interesting way to define what, what culture is.Interestingly, when you ask people about their culture, "Hey, what's the culture of Google or Airbnb," or so on, the way they'll answer the question is through rituals. And say, you know, "Here's, here's what we do. And the way you know this is what we do is through this ritual that, that, that, uh, that's in place." So I, I thought that was pretty interesting. Um, started off just building a little listicle of like, here's all the great rituals. And then I started realizing that actually the comparison between them is kind of interesting. And so I started sort of filling in the gaps between them of like, when would you do X versus Y? And what did I learn through that process? Publisher asked if I would turn it into a book, and I, I, um, uh, agreed without really contemplating how hard it would be. And it's just bec- it's become what most of my evenings end up being on this. I have a wonderful co-writer, Erin Dame, who's, uh, who's incredibly gracious with her time and, uh, um, and sort of helping me sort through the best ideas and the worst ideas. But it's a really fun project.

12. 31:29 – 40:38

    Shishir’s most impactful rituals: Arianna Huffington’s Reset, Gusto’s incredible hiring call, and Coinbase’s RAPIDs.

    1. SMShishir Mehrotra31:29

    2. LRLenny Rachitsky31:29

       What are some of the more, uh, wacky or, and/or impactful rituals that you've come across that you can share?

    3. SMShishir Mehrotra31:35

       Oh, boy. I, I did, I had sorted down the list. I'll, I'll tell you some that are, um, interesting and recognizable. One of the most fun ones, uh, is from Arianna Huffington, and she shared a, a ritual called Reset. And as a bit of background on Arianna, I mean, she's well-known for Huffington Post, she now runs Thrive. The, um... She had an accident a few years back and ended up going through this period of doing a lot of research on how the brain works, and ended up coming to this set of conclusions about how you can affect your own brain chemistry. And one of the things she does as a sort of personal ritual is, I think, called a Reset. And it's basically you make a, the, the, the ritual is you make a one-minute video that is, um, personal. And it's, uh, they have a, they have a little template for doing it, but it's a breathing exercise. So it's like you, you're supposed to, like, play it while you do this breathing exercise, but it's personal. So it's like, her video, if you go search YouTube for Arianna's Reset, you'll find it. It has, uh, you know, it has pictures of her kids, it has, uh, it has quotes she loves, it has, you know, videos of her hometown in Greece, and, and so on. But the way she brought it to the team was they start meetings by randomly picking someone, they call it Spin the Wheel. They randomly pick someone and they play their Reset. And the idea is you get this, like, like, two-for-one where you, uh, everybody gets, you know, a little bit, you know, the, the, the brain chemistry rewiring of 60-minute, 60-second bre- breathing exercise. Everybody gets back into that sort of zen state a little bit, and you learn a little bit about each other. And the... She was saying that, you know, the pictures people pick and so on are interesting, but actually the music people pick is one of the most interesting. Like, a lot of people pick calm music, some people will pick something they rock out to. Some people... But, you know, everybody does their Reset a little bit differently. So it's, that, that was a really fun one. Um, another really fun one that I was surprised by, Gusto does this thing in their hiring call. So you get an offer from Gusto, and apparently when you get on the offer call of, you know, "Congratulations, you got our offer," instead of just meeting the recruiter, which is what most companies do, they have the entire group of people that interviewed you join the call, and they all say something about why you're amazing and you should join Gusto. And like I said, it's such an interesting ritual in so many ways, right? I mean, like, it's a, for, for the candidate obviously, like, what an amazing experience. It's like, you know, it... They, uh, to use an Airbnb term, that's like a level 11 experience of, uh, (laughs) of what that, what that feels like. But also for the company, it's like, hey, we're like, we're gonna get, you know... I, one of the questions they asked, "What if I voted no? What if I'd said this person's a no-hire?" And they said, "Doesn't matter, you're on the call. You're gonna work with this person. You're gonna help them feel welcome, and you're gonna help them understand, you know, where, where they stand." I just, I, I think it's a, obviously takes a bunch of time, but it's, you know, it also is a signal to the company of how important hiring is. And it's something that obviously we all, we all try to prioritize. Those are maybe a couple other different kind of, uh, maybe different ones that people might not have heard of before.

    4. LRLenny Rachitsky34:31

       Those are amazing examples. Have you integrated any of these rituals from other companies doing this research into, into Coda?

    5. SMShishir Mehrotra34:37

       Oh, all the time. (laughs)

    6. LRLenny Rachitsky34:39

       (laughs)

    7. SMShishir Mehrotra34:39

       I mean, and they, I mean, and it's the cheapest form of research. I mean, I get to, like, borrow all these great ideas from all these companies. I mean, we, we just added one to our decision-making process. This is like, this, this is a good example of a little detail that really matters, is Coinbase has a ritual that's formed around, it's a decision-making ritual they call, uh... It's a, amazing how many companies have a decision-making ritual, with a name, with a verb, right? So like, at, uh, um, uh, at Square, Gokul called them spades. This is like a... It's like he, he kind of verbified it. There's a template. There's a, you know... But you use, you use Bing's 3/10s, right? It's gotta be named, every employee knows about it by the first Friday, and it's templated. So Coinbase does a thing they call Rapids. And Rapid is, you know, uh, uh, framing around what the, um, uh, re- the, the roles in it are, the, the, you know, responsible approver participating informed end decider. But their, their technique of doing it was really interesting. They, they had this, like, subtle nudge thing that we weren't doing that I've now incorporated at, at Coda. So we, at Coda, we have Dorian Pulse, like a very common ritual we do. It's spread through a lot of different companies that use Coda. You don't really have to use Coda to do it, but, you know, I think Coda's pretty good at it. But one of the things we were facing was that you would do this pulse, and so you'd have a meeting, and it could... If you did it wrong, it could feel like voting, and it could feel like consensus building. And so we would get this, you know, people would talk about it. And every ritual has its, like, pro and con, but people would look at it and say, "We have an open culture, you're allowed to share whatever you want." But on the other side, for the person that's trying to make a decision, it can feel like, "Oh, my God, I now have 30 pieces of feedback. Am I supposed to wait for, like, all 30 of them to be yes? Am I supposed to wait for it to be a majority? And, like, how do I know? What am I supposed to, uh, supposed to do here?" And so Coinbase had this really simple idea that we sort of smooshed together, which was at the top of their Rapids, they named who all the people were...And then next to each one, they put a little box that said, "What is the decision from that person?" And it just organized ... It's very similar to Pulse, but they kind of organized them and said, like, "Y- everybody that's in the inform bucket can, like, they can comment, that's totally fine, but we really care about the approvers, the responsible," and then of course, "the decider." It's the- the- the one that- that really matters. And so we just added a column to our Pulse, which is, "What is the role?" And we grouped the table by that. And the other thing that Coinbase does, which is, like, sounds really subtle and small, but it really d- de- detail that really matters, is the person running the meeting pre-fills that with what they want from that person. "You are an approver." Uh, you know, maybe I have three approvers, 'cause I have a budget approver, and I have a marketing approver, and I have a, you know, I have a, uh, sales approver, whatever it might be. "And I need you to give me this answer. I don't- I don't need you to comment on everything I- I'm doing, but I need you to tell me, do we have the budget or not?" Or, "I need you to tell me, you know, am I, am I authorized to make this change in this part of the product that we- we generally don't change? Or can I change the onboarding flow?" Or whatev- whatever it might be. And we took a process that I think was, like, doing a pretty good job of getting, getting rid of groupthink, which is really the heart of what we were doing with Pulse, but had this danger of being overly leaning towards, uh, you know, consen- consensus building to a fault. I think consensus building is a good thing, but consensus building to a fault is not. And we sort of stole this one from Coinbase and we squished it in and it, like, got better. And, uh, I, I think that's a good, good example very recently.

    8. LRLenny Rachitsky38:04

       I feel like you have a clear bestseller on your hands here, and I can't wait to read this (laughs) . I almost feel like you have a unfair advantage right now, having all these insights before you share them, being able to execute so much more efficiently.

    9. SMShishir Mehrotra38:17

       Well, yeah, I have to, it's interesting. I don't think it's, um ... I'm obviously not trying to keep any of it a secret (laughs) , so the, the whole point of publishing it is because I think other people, uh, will, will enjoy it, um, and, and can get, can get benefit out of it. But, you know, if, if people are interested, uh, I, one of the other choices I made in writing this book is, I decided to do it somewhat in the open. So there's a, what I call the Rituals of Great Teams brain trust. And so if you go, uh, if you just search for me and Rituals of Great Teams, you'll find it, and I'm sure we can add the link to the show notes. But, uh, you can sign up and, and basically as I finish a chapter, I put, put it out to the group. And there's now a few hundred people that are helping me co-edit this thing. Some of them, just because they, like some of them come in and give, give me help on storytelling, grammar, it's fun, but a lot of them are, are contributing. They show up and they say, "Hey, you missed this one." Like, "I, we actually do that, but we do this other thing a little bit differently, and you should really talk about that." And, and 'cause I kind of view it as a, um, you know, it started as a dinner series, right? It started with a, you know, e- everybody's gonna give to each other. And so I kind of wanted to bring that into the writing process. It's also a cheap way to get some pretty good editors, um, and it's pretty helpful.

    10. LRLenny Rachitsky39:23

        I love that. That is really smart. This episode is brought to you by Eppo. Eppo is a next generation AB testing platform built by Airbnb alums for modern growth teams. Companies like Netlify, Contentful and Cameo rely on Eppo to power their experiments. Wherever you work, running experiments is increasingly essential, but there are no commercial tools that integrate with a modern growth team stack. This leads to wasted time building internal tools, or trying to run your experiments through a clunky marketing tool. When I was at Airbnb, one of the things that I loved about our experimentation platform was being able to easily slice results by device, by country, and by user stage. Eppo does all that and more, delivering results quickly, avoiding annoying prolonged analytic cycles, and helping you easily get to the root cause of any issue you discover. Eppo lets you go beyond basic click-through metrics, and instead use your North Star metrics like activation, retention, subscriptions, and payments. And Eppo supports tests on the front end, the back end, email marketing, and even machine learning clients. Check out Eppo at GetEppo.com, GetE-P-P-O.com, and 10X your experiment velocity. How have you found these rituals

13. 40:38 – 42:50

    How do you find your own team’s rituals

    1. LRLenny Rachitsky40:38

       form for someone that's listening and are like, "Hmm, we need some of these, we need some rituals, we need to move more effectively." How do these come up? Are they just organically kind of organized? Do they come in from other companies and they evolve? Is it founder driven? What have you found so far? I know you're still working on the book, but curious.

    2. SMShishir Mehrotra40:53

       There's a whole section of the book on, I get this question a lot of, like, "Well, you know, uh, how do I find the rituals we have? How do I adjust the rituals to match? Are they supposed to change?" You know, there's lots of, lots of things you see, because some rituals are like great when you're a hundred people and they're terrible when you're a thousand people. And, and, and you should actively change those things. There's some, some are great in what sometimes people call peacetime versus wartime. It's like this, you should, you should do this during this time, but you should like actively not do it when you (laughs) , when you're, when you're in this other time. And actually that in itself is a ritual. I mean, every company has some form of a war room ritual that is, uh, I guess Facebook, I was learning... We did a dinner last night, so I learned a little bit about Facebook apparently calls them lockdowns, which is a term I hadn't heard before, but apparently like, it's well understood. When they say lockdown, everybody knows exactly what it means. It's like, "We no longer do the goal setting process. We, we owe you job, this type of work." And like, everybody just knows this is what, this is what it means. But I would say that, you know, when we talk to people about rituals, it's a s- set of rituals that happen organically, and those are the easiest. It, it, like Doreen pulse for us was one of the product managers running a meeting, just thought it was, we're a, we're a, uh, distributed first culture, but like, hadn't really adapted to it properly. And he, this, this product manager was just annoyed at waiting around on Zoom for everybody to go around and say their piece. It just took forever. And by the end, he was like, everybody's like, "Why is, can we just pause? Everybody just write down what you're thinking." And he just happened to do it in this thoughtful way and it just took off. And so s- some rituals grow organically and you just, you just got to, like, wait for them to, to, to do it. But there are cases where companies actively form a ritual to, to drive a certain behavior. And, you know, the best advice I've given on this topic is, um, to read one of my other favorite books. This is, um...It's a book called Switch. It's by Chip and Dan Heath, um, all- all their books are amazing. Um, these brothers also wrote Decisive, and Made to Stick, and Moments, and- and so on. But this one is my... If I could recommend five books,

14. 42:50 – 45:01

    How to change things when change is hard

    1. SMShishir Mehrotra42:50

       this would take two slots on the list. And the subtitle of the book is How to Change Things When Change is Hard. And the basic idea of the book is, they use this analogy of an elephant... Uh, a rider on an elephant on a path. And when you're trying to change things, you have three options for what you can do. And they actually kind of map to Bing's analogy. So, you can direct the rider, so you can tell people what to do. You can motivate the elephant, so you can kind of give this thing a kick in the butt, and it's gonna move. You don't know exactly where it's gonna move, but it's gonna move. And you can shape the path. Shape the path is, "I'm gonna set this way up so that you can only do these things." And so, you know, the way I think about it is, you know, direct the rider, tell people what to do. That's why, you know, if you look at Bing's test, that's- that's why you teach employees this in- before their first ride. Uh, you- you tell people, like, "This is what we... This is what we do." And so some rituals, like, how do you get this ritual to work? You put it in your new- new hire onboarding and y- y- you make it work that way. Motivate the elephant, a lot of that's about branding, right? So what do you do with rituals? Like, I... There's an amazing number of rituals where I'll tell people, "That seemed like a great ritual, I would highly encourage you to give it a name." Gi- give it something that lets people anchor ideas to it. I mean, names are a very powerful thing, right? If I, if I said, "Oh, yeah, at Coda we do voting and sentiment writing," um, or something, I don't even know what- what you would say, it would sound boring. It wouldn't sound like something you could brag about, something you could form identity around. And so- so it's very important to give it a name. And then, and then finally, like, why do you shape the path, or you, like, set things up? You- you templatize, make it easy, m- make it as easy as possible to follow this ritual, and make it just a part of what we do enough that, you know, if you're at Gusto and you're like, "I don't really understand how this hiring call thing works," like, guess what? You're gonna be invited to one soon. (laughs) You're gonna see it, and then- and then you're gonna have to do it. Or- Or if you're at Square, like, you're gonna see this bay template and you're gonna- you're gonna learn how to do it. So I think Switch is, um... I- I- I recommend this book for lots of purposes, but as you're thinking about rituals for your- for your team, it's a- it's a particularly relevant frame.

    2. LRLenny Rachitsky44:45

       Awesome. By the way, that's a little plug for the YouTube version of this podcast, which is now a thing we do. So if you're like, "Hey, I don't see what you're talking about," uh, just search for Lenny's podcast YouTube, and I think you'll find it, and we'll probably link to the-

    3. SMShishir Mehrotra44:59

       Yeah.

    4. LRLenny Rachitsky44:59

       ... to that in the show notes. I'm also reminded

15. 45:01 – 46:45

    Airbnb’s unique rituals

    1. LRLenny Rachitsky45:01

       of Airbnb's rituals, which you probably already know about, but they're kind of all hilarious and weird. One is Formal Friday, where people dress up in suits and gowns on Fridays. Another is a human tunnel for all new employees, where every new employee has to run through a human tunnel and jump into, like, a beanbag or something.

    2. SMShishir Mehrotra45:17

       Yeah.

    3. LRLenny Rachitsky45:17

       And then- and then there's a new hire tea time, (laughs) where new hires drink some tea with some veterans and chat about where they're from and things like that. There's a bunch more, but there's the ones that are my favorite.

    4. SMShishir Mehrotra45:27

       Those are great. And I... uh, and there's a... I also included some of the, uh, the- the sort of level 11 thinking, uh, as well.

    5. LRLenny Rachitsky45:34

       Uh-huh.

    6. SMShishir Mehrotra45:34

       I think that's also, uh, a good Chesky favorite. Brian also has another one that's in the book that's about how to rank your to-do list by finding leverage, that's a, that's a really fun one as well. They're like, "Don't rank your to-do list..." But a lot of people do, like, importance versus, um, urgency or- or- or so on. And I- I guess he sorts his by, which one of these is most likely to create leverage of, like, getting rid of the rest of my list, which I thought was a- a very, like... You know, I started doing that in my to-do list and it's very interesting and impactful.

    7. LRLenny Rachitsky46:01

       That's actually an incredible segue to our next topic, which are Eigenquestions.

    8. SMShishir Mehrotra46:06

       Ah, yes.

    9. LRLenny Rachitsky46:07

       And so Eigenquestions, one of your most classic posts, you mentioned this at the top, may be your most liked post other than maybe bundling. It sounds like some kind of German game show, Eigenquestions.

    10. SMShishir Mehrotra46:18

        (laughs)

    11. LRLenny Rachitsky46:18

        Can you tell us what Eigenquestions are?

    12. SMShishir Mehrotra46:19

        Mm-hmm.

    13. LRLenny Rachitsky46:19

        And then I'm gonna ask you a few more questions around that?

    14. SMShishir Mehrotra46:22

        I hadn't thought about the German, uh, game question (overlapping speech)

    15. LRLenny Rachitsky46:27

        We should- yeah, let's start with that.

    16. SMShishir Mehrotra46:27

        Yeah. (laughs) We're gonna... I don't know, this would be a very interesting game. But we could- we could- we could tr- try to create one. Okay. So I'll describe what Eigenquestions are, but maybe I'll start by telling you a little bit about where the concept came from. And, um, this is actually a YouTubism that... And maybe just to- to place ourselves in history.

16. 46:45 – 53:00

    A back story on YouTube and valuing consistency over comprehensiveness

    1. SMShishir Mehrotra46:45

       So in 2008, I joined YouTube. And, um, many people don't remember this, but YouTube at the time was seen as a mistake. It was seen as Google's first- first bad acquisition. Everything else had worked. But this thing kind of seemed like a disaster. I mean, it was- it was grainy videos, we were losing lots of money. We had billion- uh, billions of dollars in lawsuits. It was like, none of it really seemed that obvious. There was lots of discussion on how to reorient it, fix it, and so on. And so everybody outside, that's what they... That's what they saw. If you stepped inside a YouTube staff meeting in 2008, actually one of the toughest questions we were answering was this question we called the Modern Family question. And- and, um, it- it may sound small, but it actually was kind of... It was very perplexing. And the question was pretty simple. The question was, you know, people... If you looked at our search traffic at- at YouTube, one of the top five queries basically every week was for a show called Modern Family. And Modern Family was number one show on television at the time, you know, by far the most popular. And we were, you know, second- second biggest search engine in the world, behind our parent company, Google. So search traffic was very important. There was one big problem. People would come search for Modern Family, one big problem, we didn't have Modern Family on YouTube. And so we gave them some pretty crappy responses. And the question was, you know... Oh, and- and- and, by the way, abc.com had decided to post every episode of Modern Family live on their- on their- on their website, which nowadays is kind of typical, but in 2008 that was not typical. And there was... It was kind of a big gamble they made, and they're gonna post all these- all these shows and so on. So the question was, you know, should we answer the que- the query Modern Family by linking off to abc.com? Do- do we link out or not? And it... The company basically divided. And so there's like half the- half the company, mostly the product team, advantage team, so on, kind of aligned around a viewpoint that was, "That's what the user wants, link them off to abc.com. You know, we're now owned by Google. Google tries hard to prioritize, do right by the user, the rest will follow." Like, "It seems like the right thing to do, so let's do that."And then the other half of the company, most of the business functions, sales, marketing, especially content partnership, said, "Please, please, please don't do that. If you do that and you start linking off to all these other places, nobody's ever gonna put good content on YouTube and we're just gonna get the stuff that doesn't deserve to live anywhere else. And that's, like, not a very good path to be." And you can imagine that, like, those two mindsets, it's almost like it was, like, good versus evil debate, which is like, do right by the user or the business. So like, these are all, like, like, almost impossible to solve problems. And this would, like, happen meeting after meeting after meeting. "What? What? Did we make a decision yet, Modern Family? What are we gonna do?" So we had this offsite where we said, "We're gonna, like, spend the whole day and we're gonna figure this problem out." And we went up to this, this hotel in Half Moon Bay and the ex- the executive team all sat down and I was asked to frame the discussion, go collect everybody's opinions and collect all the data and just, like, ground it all in facts. And then we're gonna have a discussion and we're gonna reach a decision. And so the night before, I'm sitting and thinking about like, how, you know, how the hell am I gonna have this discussion not just be like a shouting match of, you know, uh, of, of this good versus evil position? And I happened to read a, uh, an analysis that was being done by a different team at Google, the Google Shopping team, where they were facing this interesting challenge of, um... They were in this deep fight with Amazon and they were getting their butts kicked and they were trying to figure out why. And it... The walking in theory was, you know, the Google Shopping team's view was, why would anybody ever pick going to Amazon? Like, you could come to, to Google and we had indexed all of Amazon and the entire internet. Why would you ever pick Amazon? And the feedback that was coming back from users was they would... They'd say, "I picked Amazon because I value consistency over comprehensiveness." They would say things like, you know, "I, I really value that. You know, I go to amazon.com and I understand how the reviews work and how the ratings work, and I know that the returns work the way I want and I understand how shipping is gonna work." And, and it just felt consistent. I know it doesn't have everything, but it has enough and I value consistency over comprehensiveness. And so the night before this, this YouTube meeting, I decided to reframe the question and say, "Let's not have the discussion about linking out at all. We're gonna start by having a theoretical discussion about in a decade from now, is the video market more likely... The online video market more likely to be about consistency or about comprehensiveness?" And that is a question that you can have a very reasonable debate over. Like, what are the... What are the reasons why a market evolves towards consistency over comprehensiveness? So... And we basically have this discussion and we all came to the conclusion, this market is gonna value consistency over comprehensiveness. And by the way, I think, I mean now almost 15 years later, I think we were right. I mean, I think the... If you go look at the, the... I mean, video market obviously has exploded. There's so many great video properties out there. None of them are comprehensive. There is no one-stop shop for a, uh, w- a place where all, all the video exists. And so I think we were right. But what, what happened was by answering that question, the link out question all of a sudden became super easy. Like, we value consistency over comprehensiveness, we definitely don't need to link out. In fact, we should make a whole bunch of other decisions as well. So we went and, you know, at the time we used to... This is the, the days of flash players and so on. We embedded other people's players on YouTube. We stopped doing that as well. Probably the most famous decision we made was with the iPhone. As I mentioned earlier, the i- the iPhone, when it first shipped, had no app store. And so they built all the first, first few apps, including YouTube. Um, and so here we were a few years later, iPhone most popular phone on the planet, and the YouTube app on the iPhone was built by a team at Apple. And they had not been able to keep up with what we were doing. Almost half the catalog didn't play back on, on the iPhone. They were missing a bunch of features and so on. So I drove down to Cupertino and I sat down with, with Scott and Phil and said, "Hey, we're gonna have to take back the, the YouTube app." And they, they said, "I don't understand. Why would you do that? Like, you're... You have default distribution on..." You know, as far as they were concerned, the most important operating system in the world. "Why would you do that? You're gonna have to, like, rebuild all this from scratch and, you know... Uh, just seems like a really bad choice." And I said, "No, no. It's like actually quite an easy choice. We value consistency over comprehensiveness. We would much rather be on fewer phones with a more consistent experience than be on all of them with an inconsistent experience. So what... This is the choice we're making." And it worked out fairly well. So this decision, the, the Modern Family question, ended up becoming named as the example for a term called Eigenquestion. So Eigenquestions, it's

17. 53:00 – 59:05

    Eigenquestions: What it is, how to use it, and examples of it

    1. SMShishir Mehrotra53:00

       not a German game show. (laughs) Uh, it's not... It is a made up word and it's named after a math concept called Eigenvectors. And, um, you know, the math is not really necessary, but for people who are curious, you know, go back to linear algebra. Eigenvectors are in a multidimensional space, they're the most discriminating vectors of the, of the, uh, vectors in that space, the dimensions of that, of that space. It's a concept that gets used a lot in machine learning and, and, and so on. But actually the math doesn't really matter. The, the Eigenquestion, the simplest definition of an Eigenquestion is the question that when answered also answers the most subsequent questions. And it's a very simple idea that when you sit down and you say, "Hey, here's all these questions we gotta answer. How do we all usually rank them?" Like, sometimes we rank them just by, you know, what order we came up with them. Sometimes we rank them by importance, like which is the most impactful decision we're gonna make. But this methodology says don't rank them that way. Rank them, like you said, Jessy, think about leverage, same, same idea. Rank them by which ones would eliminate the most other questions from the list. So you take that list and you said, "Should we link out to Modern Family? Should we own the, the, the YouTube iPhone app? You know, should we do..." Those were all, like, really hard questions to answer. But it turns out if you answer just one question, you know, do we value consistency over comprehensiveness? The answer all the others. They, they all, all of a sudden become very simple. And so this, this idea of Eigenquestions became part of our vocabulary, became a clear ritual for, uh, for YouTube that, you know, what is the Eigenquestion here? It is for Coda as well and sort of spread to other places. But that, that's the basic idea.

    2. NANarrator54:26

       Amazing. What a... What a baller move with Apple. And, uh-

    3. SMShishir Mehrotra54:30

       Yeah. Pretty scary move. (laughs) Yeah.

    4. NANarrator54:32

       (laughs) But I was just gonna say, I, I think YouTube's probably in the top five, 10 most downloaded apps, so it worked out.

    5. SMShishir Mehrotra54:37

       You know, the... We'd, we'd go up to that meeting and I bring along our, uh... The, the, the product manager for the iPhone app, a guy named Andrey Darnacheff. And he's the one, like, having to...... explain what we're gonna do and so on. And it's like a hard, contentious meeting. And as we're leaving the meeting, he says, "Hey, uh, can I get a selfie with Phil and Scott?" (laughs) And I'm like, "Andrey, what are you doing?" Like... (laughs)

    6. LRLenny Rachitsky55:00

       (laughs)

    7. SMShishir Mehrotra55:00

       And so there's this great picture of him, like, with us asking for, you know, take this back. And he was clearly, he was very starstruck. I mean, it's like, there's a lot of people who view it as like, Apple would do a better job of building a YouTube app than us. Like, what? Who are we to tell them to not do that? Um, of course, you know, in retrospect that was, that was a silly way to think about it.

    8. LRLenny Rachitsky55:17

       Wow. I would do the same thing. (laughs)

    9. SMShishir Mehrotra55:20

       Yeah. (laughs)

    10. LRLenny Rachitsky55:20

        That's amazing. Who's this person? 'Cause that's awesome. I love that as a leader you bring the, the PM of the team working on it versus just, you know, the big, the big shots at the top.

    11. SMShishir Mehrotra55:29

        Yeah. Andrey, he's, uh, he's now a founder. He's, uh, he started a company called Optik which is basically building content ID for NFTs, which is a, you know, v- much needed thing in the Web3 world. But yeah, I mean, the, the team building it, I mean, it was, uh, that meeting I brought him, I brought my partnerships lead, and I brought the eng lead that was covering the area too. And, uh, yeah, I think that some of it is, you know, if I had to be honest, some of it is, like, they really wanted to come and meet with Apple. Some of it was like, you know, for my own sake, like, I kinda wanted some backup. (laughs) That's the... I was about to make this kind of bold ask. And they could've been, uh, I mean, to Apple's credit, I mean, they could've been, you know, pretty bad about it. I mean, they could've, like, not allowed us on the store, and, and, and so on. And they, they said, "Okay, well, you know, we don't like it, but we understand your choice. You have to know that you're gonna start from zero. We're not giving you a single download, uh, uh, for free. You're gonna have to start from zero, and we will brick the current app by the agreed-upon date." And the negotiation was, "Can you please just tell those people that there's a new app?" And so that's what we negotiated out of it. And they, they eventually did that, and that was, uh, that, that was, that was fine. And in the end, YouTube is now, you know, one of the top download apps on, on iPhones, I think. I mean, e- it was like six months after launch we had like 80% share, and like almost like e- e- everybody downloaded the app, vis- a- and, and so it, it kinda ended up not being that much of a comprehensiveness, uh, uh, uh, choice. But it was a, yeah, clearly hard decision, made much easier by asking the right eigenquestion first.

    12. LRLenny Rachitsky56:55

        Wow. Speaking of eigenquestions, are there other examples of eigenquestions that come to mind to make this even more concrete in people's minds? I don't know if that's the right way to frame it, or is it more just when you have a list of questions, look for the one that'll answer the most? How do you, how do you kinda operationalize this concept?

    13. SMShishir Mehrotra57:12

        You know, uh, there's lots of them. I mean, for Coda the eigenquestion, um, the sort of most conceptual eigenquestion for Coda was, you know, we use a line a lot for Coda, that Coda allows anyone to make a doc as powerful as an app. You can reverse that statement and say, "Allow anyone to make an app as easily as a doc." And those two sound similar, but they're not. They're, they're actually quite different statements. And so our, our most commonly debated eigenquestion is, you know, are we more committed to being a doc or being an app? And, you know, which way do we want people... If people are gonna misunderstand Coda, would we rather them perceive it as a document or perceive it as an app? And we decided on doc, which is like, actually... And the way I, the way I cemented that decision when we made it was I named the company that way, Coda is a doc backwards. I said, "Well, well, it's all, it's a, it's a, this is, this is gonna be a, we're definitely not revisiting this one. This is a, uh, Coda is a doc first." That's a good example. I mean, another one, and I, by the way, I would say eigenquestions is a, is a term that, um, you know, a guy could

    14. NANarrator58:13

        Yeah.

    15. SMShishir Mehrotra58:13

        ... resonance and so on. But it's a hard technique. I mean, it's not, it's not always easy to know how to do it. And one of the things, you know, I get asked a lot is, like, is it a skill you can learn? I absolutely think it's a skill you can learn. I mean, it's a thing that once you observe it and get better at it, you can learn it. But it's not, it's not easy to learn it. And one of my observations about learning skills like these is you wanna learn them in non-pressure-filled environment. Like, to use a- an analogy, you know, if you were trying to learn a sport or learn an instrument or so on, imagine if you never did practice. Like, every time you played basketball was in a real basketball game, and every time you played the piano was in a recital. You know, you probably would never get better. And I think one of the troubles with a concept like eigenquestions is we tend to only practice it in real world scenarios that are high stakes.

    16. LRLenny Rachitsky58:58

        Right.

    17. SMShishir Mehrotra58:58

        And so one of the, uh, one of the things I encourage people to do is to practice eigenquestions in completely almost frivolous situations.

18. 59:05 – 1:03:11

    One of Shishir’s favorite *retired* interview questions

    1. SMShishir Mehrotra59:05

       So I, I have an inter- interview question I ask, which I think maybe we'll get to this a little bit la- uh, uh, later as well, but it's a very, it's a very simple question, but it's a, and, and it's a coded eigenquestion test. And the question is, um, a group of scientists have invented a teleportation device. They've hired you, Lenny, to be their sort of business counterpart, bring this to market, product kind of reward. This question actually works well for any role. But say you could be a product manager for this thing, bring it to market, um, and what do you do? That's, that's the whole question, right? So, so, so you, people, usually people will start asking a bunch of questions and say, "Well, tell me more about this device. Like, what does it do? How does it work? And, you know, is it, uh, is it big? Is it small? Is it fast? Is it, like, does it dis- disintegrate things or not? Does it need a receiver and a sender? Does it, you know, is it safe?" Like, all these different questions come out. And at some point I'll just let those questions come out, and at some point I'll say, "Okay. Nice, uh, nice job generating all the questions. Turns out these scientists, they kind of hate talking to people, and they're kind of annoyed by all your questions. And so they've decided that they will answer only two of your questions. And after that, they expect a plan. What two questions do you ask?" And interestingly, all of a sudden, like, the sharp, you know, product managers, engineers, so like e- in basically every role, they very quickly find what are the two, one or two eigenquestions on this topic. And there's no right answer, but I'll tell you, like, one of my favorite ones is, is a product manager said, "Okay, if I had to as- ask two questions, the two questions I'd ask, one is, is it safe enough for humans or not?" And that was a, like a very, like, crisp way to get to, like, just safety, how reliable is it? He didn't ask how reliable it is, how many bits in the middle

    2. NANarrator1:00:46

       (laughs)

    3. SMShishir Mehrotra1:00:47

       ... of this thing. Like, "Just tell me. Is it safe enough for humans or not?" And the second one is, "Is it more expensive CAPEX or OPEX? Is it more expensive to buy them or to run them?"And then he took those two questions and he said, like, "Guess what those two questions, I can form these quadrants." You could say, oh, it's safe enough for humans and it's cheaper to... They're very cheap to buy, but expensive to run. Then you probably run them like human fax machines. Like, you put them everywhere you can and you say, "Hey, look, it's expensive to use, but, like, you all have the ability to teleport anywhere you want. And this is, this is how we're going to run it." On the other hand, they're very expensive to buy, but very, um, but cheap to run. You probably have to place them very strategically. In which case what you'd probably do is we place airports, because, like, airports are pretty strategically placed in places where people are trying to, trying to get, get around places. If it's not safe enough for humans, then you've got a whole different class of use cases where you go value what goods are transported in very costly ways and, you know, people come up with like, you know, do you do the most expensive things or do you do like, you know, uh, is, is teleporting, you know, people's replacement hearts? Is that like a, like a really demanding thing? So, so but these two questions kind of, kind of get to the, the, the heart of it. The question's totally made up. Like, no, no teleportation device exists, or at least not yet. And I find that people's ability to learn the method is significantly higher if it's low stakes. That question, by the way, if you ask a kid that question, the, you know, "Hey, new teleportation device, you know, you get, you get to ask two questions." Almost every kid will, like, quickly get that you have pretty good Eigenquestions. Again, kids are incredibly good at simplifying these things down. It's actually a skill we, like, remove from ourselves. Like, I'll see, I'll hear candidates tell me things like, "Well, I guess I would ask them what, what size it is." And like, why would you ask them what size? What, what, what decision is that gonna allow you to make, to, to know what size it is? And, and you know, sometimes they can explain it, but sometimes not, and don't get hired. But the... (laughs) But actually, the, the thing I'd say about it is, there are Eigenquestions kind of everywhere. I mean, so you could, you could take any product out there. I'll, I'll do it with my kids a lot, and they'll say, you know, uh... I was just riding with, with my younger daughter and she said, you know, "How come there's three gas stations, like, in the same corner? Like why, why, why do, why do people do that?" It's a, you know, uh, that's a really, that's a really insightful observation. What's the Eigenquestion? How many places has he asked the question? And there's like a var... You know. Uh, and you can take, you can almost take anything and say, "What is the question that really drives, drives his answer?"

    4. LRLenny Rachitsky1:03:07

       I love that. Do you, do you actually still ask this question?

19. 1:03:11 – 1:15:20

    How to evaluate talent, a story about YouTube and breaking down PSHE

    1. LRLenny Rachitsky1:03:11

       Because you're sharing it and all the answers.

    2. SMShishir Mehrotra1:03:13

       No, I don't.

    3. LRLenny Rachitsky1:03:14

       Okay.

    4. SMShishir Mehrotra1:03:14

       And I have a new one that I can't share. But it's a... And we've written about it. In fact, one of the big debates about publishing the Eigenquestions thing is, i- in order to bring this to life, I needed to answer your question. Like, how do I, how do I test this? How do I practice this? And like, it's much easier. Like, nobody can repeat the YouTube one because they did... You know, nobody has that choice sitting in front of them, so it's kind of a useless... It's, it's entertaining, but it's a, as a teaching tool, it's kind of useless because you can't, you can't really go reinvent history and decide

    5. LRLenny Rachitsky1:03:39

       Yeah.

    6. SMShishir Mehrotra1:03:39

       ... consistent versus comprehensive.

    7. LRLenny Rachitsky1:03:41

       Have to sacrifice one.

    8. SMShishir Mehrotra1:03:42

       We sacrificed one, yeah.

    9. LRLenny Rachitsky1:03:44

       So pulling that thread further and kind of dive a little deeper into, into evaluating talent and product talent. I hear this is one of your superpowers. And so I'd love to learn from you and what you've seen around how to evaluate talent. So we talked a little about interview questions you ask. Um, so maybe we could either go in that direction or just, like, what do you look for in people that you're hiring, interviewing that maybe other people aren't?

    10. SMShishir Mehrotra1:04:05

        I have a technique for it. Let me show... Uh, I'll show a quick picture and-

    11. LRLenny Rachitsky1:04:08

        Ooh, YouTube plug.

    12. SMShishir Mehrotra1:04:10

        Yeah, YouTube plug. You can... Uh, I mentioned it before the call. You can, you can put video on Spotify now too, but the, the, uh-

    13. LRLenny Rachitsky1:04:16

        Spotify plug.

    14. SMShishir Mehrotra1:04:17

        Yeah. (laughs)

    15. LRLenny Rachitsky1:04:17

        All your platforms that you've worked on, now can both play my videos.

    16. SMShishir Mehrotra1:04:21

        That's right. Um, so I'll talk through this diagram that has two axes; scope, this acronym, PSHE, and this line. So I'll stop sharing and describe it and we can come back to it. But I'll tell a little bit of this technique that sort of changed how I think about evaluating not only product talent, but it actually turns out you can use the same set of rules for evaluating basically every role. But I'll tell it from how you asked it, about product talent. So 2011, Larry Page took over at Google and he made a bunch of changes to the company, mostly quite positive. And one of the ones he did was he, he moved us from being a functional organization to being a business unit organization. We called them product units, but they're all for the same thing. And there were eight product units at, at Google. You know, YouTube and search and ads and Chrome and maps and so on. And it was very positive. Uh, it's, like, hard to believe that we were already like, we were like 20,000 people, we're still functional, like all of engineering, all of product still reported into the CEO. Just seems like totally crazy with the breadth of products that we cover. One of the downsides of it was, like, in any functional to business unit switch is you lose some of that, that what does the function need? And in particular things like what is a good product manager was a question we were at risk of losing. So a group of... At the time I was running product for, for YouTube. The group of the, the eight product leaders around the company got together and said, "Hey, we need to, like, keep some level of, um, consistency amongst how we think about what's a great product manager or it's all gonna diverge and it's not gonna mean anything anymore." And, um, actually as a fun aside, we did a ritual that I've repeated a few times but it isn't done often enough is, uh, we said, "So who's gonna, like, drive this process?" And we did an election. So I, I don't know why we do elections in, like, the public world but not in the private world, but it's actually, like, quite effective. We used to do them at YouTube where we would, like, we would elect into certain roles and, like, you got, you got a one year term, you gave a little speech. You're like

Episode duration: 1:31:48