The 10 traits of great PMs, AI, and Slack’s approach to product | Noah Weiss (Slack, Google)

1. 0:00 – 4:22

   Noah’s background

   1. NWNoah Weiss0:00

      We have this mental metaphor that we talk a lot about, getting to the next hill. The actual wording is: take bigger, bolder bets. I think teams can often get lost crawling up that hill, not realizing that there's a huge, incredibly beautiful range behind it. We've, over time, created kind of new teams from scratch that incubate in a new area before the areas mature. So we did that with a lot of these kind of native audiovisual products, like Huddles and Clips, early in the pandemic because our customers were demanding it from us. And I think in the AI space, we're trying to hear from customers, "What do you wish Slack could do if it had these new superpowers?" And let's incubate a couple teams, prototype, give them space to run and pilot, and then get something to launch that's amazing, blows people away. That's kind of been the formula that we've seen.

   2. LRLenny Rachitsky0:43

      (instrumental music) Welcome to Lenny's Podcast, where I interview world-class product leaders and growth experts to learn from their hard-won experiences building and growing today's most successful products. Today, my guest is Noah Weiss. Noah is chief product officer at Slack, where he spent the last seven years. Prior to that, he was head of product at Foursquare, which is near and dear to my heart, as you'll hear at the top of this episode. Prior to that, he was a PM at Google and at Fog Creek Software. And in our conversation, we cover the 10 traits of great product managers, how to work effectively with strongly opinionated and product-minded founders, what Noah has learned about working effectively with AI in a product over his last 15 years at Google and Foursquare, and now at Slack. We talk about a process called complaint storms that helped Slack build better product, plus what he's learned from Slack's self-service business plateauing back in 2019, and how they turned it around and what they took away from that experience. Also, how he thinks about competition with Microsoft Teams and with Discord. Also, a bunch of new dad advice, which I found very helpful. This was such a great, in-depth conversation about all things product and leadership, and I'm really excited for you to hear this episode. With that, I bring you Noah Weiss after a short word from our sponsors. This episode is brought to you by Sidebar. Are you looking to land your next big career move or start your own thing? One of the most effective ways to create a big leap in your career, and something that worked really well for me a few years ago, is to create a personal board of directors, a trusted peer group where you can discuss challenges you're having, get career advice, and just kind of gut check how you're thinking about your work, your career, and your life. This has been a big trajectory changer for me, but it's hard to build this trusted group. With Sidebar, senior leaders are matched with highly vetted, private, supportive peer groups to lean on for unbiased opinions, diverse perspectives, and raw feedback. Everyone has their own zone of genius, so together, we're better prepared to navigate professional pitfalls, leading to more responsibility, faster promotions, and bigger impact. Guided by world-class programming and facilitation, Sidebar enables you to get focused, tactical feedback at every step of your journey. If you're a listener of this podcast, you're likely already driven and committed to growth. A Sidebar personal board of directors is the missing piece to catalyze that journey. Why spend a decade finding your people when you can meet them at Sidebar today? Jump the growing wait list of thousands of leaders from top tech companies by visiting sidebar.com/lenny to learn more. That's sidebar.com/lenny. This episode is brought to you by Superhuman. How much time do you spend in email each day? How about your team? You may not realize this, but your email tools are wasting your time. Superhuman is blazingly fast email for high-performing teams. Built to work with Gmail and Outlook, teams who use Superhuman spend half the time in their inboxes, respond to twice the number of emails, and save over four hours a week. That's over a month of saved time per year. With Superhuman, you can split your inbox into streams for VIPs, team members, and emails from your favorite products to reduce context switching and make sure you never miss an important email. You can set reminders if you don't hear back so that you can follow up and never drop the ball on an email thread. You can also work faster than ever before with powerful AI features like writing, editing, summarizing, and even translating. Join the ranks of the most productive teams and unleash the power of Superhuman. Try one month free at superhuman.com/lenny. That's superhuman.com/lenny.

2. 4:22 – 7:23

   Advice for new parents

   1. LRLenny Rachitsky4:23

      Noah, welcome to the podcast.

   2. NWNoah Weiss4:25

      Thank you for having me. I'm excited to finally get to join you. Been a longtime listener.

   3. LRLenny Rachitsky4:29

      I feel the same way in reverse. I'm really excited that you're finally on the podcast. And I don't know if you know this, but this is actually gonna be the last podcast I'm recording before I go on pat leave. Uh, this is, this is gonna play while I'm on break. And coincidentally, you're actually just returning from pat leave, is what I just learned.

   4. NWNoah Weiss4:46

      Yeah.

   5. LRLenny Rachitsky4:46

      And so let me ask you a question. What advice do you have for someone about to enter the beginning of baby life from someone that is exiting that and going back to work?

   6. NWNoah Weiss4:55

      So first off, I mean, obviously, congratulations. You're about to go on-

   7. LRLenny Rachitsky4:58

      Yes, man.

   8. NWNoah Weiss4:58

      ... a roller coaster of emotions, sleep, and everything else. You know, I literally went back to work two days ago, so I think my maybe advice about being a new parent is better than my advice about being a PM right now. Here are the three... My wife and I wound up coming up with, like, three maxims that we want to be using throughout the first two months to keep ourselves grounded. So first one, I would say a little bit better every day. No matter how many books you read, you know, no matter how much Emily Oster you consume, there's nothing like actually doing it. And it's a physical thing, being a new parent. And so getting a little bit better every day, giving yourself permission to be like, "That didn't go great, and that's okay." That's number one. Number two, don't over-extrapolate from the early days. Like, they are... You know, the fourth trimester is a real thing. These babies come out, they are not fully baked. They can't even support their own heads. So if you try to extrapolate and think the next 18 years are gonna be like the first 18 days, it's gonna be sobering. So, like, keep that perspective. It gets... They develop so much every week. That's part of the fun. And then the third thing, which I got advice from this from a good friend, is, like, s-You got to fully get into it as a parent. There's nothing that replaces actually, you got to change the diapers, you got to do the feeds. When they're up, even if they can't talk, you got to talk to them. You got to like listen to what they're saying and just be fully kind of present in the moment. I kind of realized for myself, and then like basically a full digital detox. You saw how long it took for me to reply to your emails. Uh, I was like put all the devices away, just kind of be fully with our daughter, Willow, and our family. And I felt like it was so much more rewarding, and I feel really connected with her now after just a couple months. So it's a crazy time. You're gonna love it. It's gonna drive you mad at times as well, and that's all okay.

   9. LRLenny Rachitsky6:47

      All right, we're gonna be pivoting this podcast into a parenting podcast. This is awesome advice. I wrote everything you just said on this little Post-it as you were talking, so I'm gonna put that up in my, in our nursery and see, (laughs) see, uh, see how it all goes. One thing that's tough about my career path in this weird life is I don't get, uh, a nice, uh, pat leave, you know, paid pat leave from a big company, so I've actually been working on stacking guest posts and podcasts ahead of my leave so that I can actually, as exactly as you said, just get fully into it.

   10. NWNoah Weiss7:14

       Smart.

   11. LRLenny Rachitsky7:14

       So I have six awesome ... Yeah. So I have awesome guest posts coming. All these podcasts are backlogged, so I'm hoping it all works out.

   12. NWNoah Weiss7:21

       That's the smart way to do it, yeah.

3. 7:23 – 11:33

   Lessons learned from Foursquare

   1. NWNoah Weiss7:23

   2. LRLenny Rachitsky7:23

      On a totally different topic, you were head of product at Foursquare. And I don't know if you know this, I actually built a startup on Foursquare's API. It's a company called LocalMind. And for folks that don't know about it, the way it worked is it basically let you talk to someone checked in on Foursquare anywhere in the world if you were thinking about going there. So you could be like, "Hey, is this bar fun right now? What's happening there?" before you actually show up. And we ended up selling the company to Airbnb, ended up not being a big problem for enough people, and that's how I ended up at Airbnb, but it was, it was like quite magical and the API was amazing. And so I guess just say, I just want to say thank you for building an awesome product and awesome API.

   3. NWNoah Weiss7:58

      Thank you for being a developer on top of the ecosystem. I mean, it's interesting with Foursquare, we'll talk about this I'm sure later, I feel like I have more lessons learned and more scar tissue from the, you know, crazy up and down of, I don't know, what, what it w- was it 2010 to 2015 roughly, and I, I think there's something actually where you learn more from the things that don't fully work out or don't quite achieve what you want it to achieve, and you actually have a feedback loop where y- you get a lot of negative signal about like, "Okay, that didn't work. That didn't work. What can I actually learn and take away from that?" So it's still great. I still love using Foursquare. I think, you know, we got caught in the Death Star of Instagram's ascent, you know, back in 2012, 2013. But uh, you know, I hope a product like that exists forever in the future and I'm glad you got to build the company and landed Airbnb through it. Uh, it's a great story.

   4. LRLenny Rachitsky8:52

      Looking back at Foursquare, do you think there was a path to building a massive consumer app type business or is that just never gonna work out? And I know they went in direction of B2B data sort of business, so I guess was there, was there a path or was it just like, "No, that was never gonna work out"?

   5. NWNoah Weiss9:09

      It's tricky. I mean, I'm not gonna do like a 30-minute postmortem 'cause I would probably bore everyone, but I've thought about this, we've all thought about this a lot kind of on the early team there. I think, you know, the, the biggest probably lesson learned frankly is that, you know, we were really close with the Instagram folks early on. They were like big developers in our platform. They used the Foursquare API before they were bought by Facebook. And I think in hindsight we were a little bit mistaken to believe that the idea that the atomic unit would be a person talking about a place that they're at, and you had to have a physical place to tie it to, versus a person sharing a moment or an experience that they're having in the world, and sometimes that might have a place connected to it. I think that one change in framing on what you would say like the customer actually wanted to do, that probably was the thing that took us away on, on the social side. I think on the more kind of local discovery side, it's actually what people want to be using the product much more for over time, getting kind of personalized recommendations and getting tips when you go to a place and all the push notifications. But I think there, again, it wa- it was kind of hard to stay ahead I think specifically of Google because they had, you know, a billion plus Google Maps users distributed on Android and iOS. And even though they might only, you know, take a couple years, eventually they would wind up replicating a lot of the functionality, and then I, I think kind of it was hard to regain that momentum. So you know, so much of this stuff is luck and timing and just coincidences of history. I think there was a path. I think in the end we kind of lost our social sales and then Google was able to catch up on the utility side. And now the company's built a really valuable kind of B2B API company which offers a story. I mean, Slack is in some ways a pivot obviously from a consumer company to a B2B company. But yeah, that, that's my mini postmortem on what could have been with Foursquare.

   6. LRLenny Rachitsky11:00

      It's interesting how many consumer companies pivot to B2B, 'cause turns out that's where the money ends up being.

   7. NWNoah Weiss11:06

      Yeah, and I think the feedback someone can get from, are people willing to pay for the product that you're building is so much faster than, can I build a large scale consumer business and one day hope to have enough reach to then slap ads onto it? That's a much more of a kind of try to hit a home run and hope it works out, but you don't really know if you're doing it along the way. Uh, so yeah, I think B2B is a, easier to have a more incremental successful business than pure consumer.

4. 11:33 – 14:14

   Working with strongly opinionated founders

   1. NWNoah Weiss11:34

   2. LRLenny Rachitsky11:34

      Okay, so speaking of Foursquare, Dennis Crowley was the CEO and founder, a very strong product-minded founder. I know you've worked with a number of very strong product-minded founders including Stewart Butterfield, Dennis, obviously we just talked about, maybe others. I'm curious what you've learned as a product leader working with very opinionated founders. And I think this is interesting not just as like a product leader working with very product-minded CEOs, but also as a first PM at a startup, you're often put in this tough spot of just like the founder's just telling you what to do and you have to go build it, versus having a lot of say and agency. So I'm curious what you've learned about working and being successful in that position, which is often really hard.

   3. NWNoah Weiss12:15

      ... say to folks in general, if you're joining a company and the CEO does the role that is your functional area of expertise, it's probably the w- area where you'll learn the most, 'cause they're hopefully world-class at it, but also one where you'll, you'll be the most frustrated at times because you're gonna feel like you have less agency. And so you should just know that going into it. If you go to a company that's run by a former marketer and you're in marketing, they'll probably wanna have a lot of say and influence over that. And I think just going into it knowing that is good. I, you know, looking back I would say probably two main things stand out of what, what's really worked with both Dennis and Stewart. Not just for me, but I think for the teams that kind of work with them as well. The first is, I think as much as possible, I think maybe we'll talk about this a little bit later as well, is kind of getting to the point where you have alignment on the principles for what it means to build a great product at that company. Not just about like the intuition and taste and gut, but how do you distill that into principles that become the language of the company so that everybody else can start thinking through a similar frame or a similar lens when you're designing product? Because otherwise they can feel a little bit kind of Goldilocks every time. A team builds something, they take it to the CEO, the CEO's like, "No, not quite right." Again, "No, not, not exactly that." And then you don't have the language to actually have a more constructive review. And then doing that at the level of strategy as well. I think the product founder CEO is always gonna be the holder of the vision for the company. I'm sure at Airbnb, I, I imagine Brian was very much like that as well.

   4. LRLenny Rachitsky13:40

      Absolutely.

   5. NWNoah Weiss13:41

      And I think it's actually great to say, okay, the overall vision for the company isn't the responsibility of any one team. Have everyone buy into that vision, but then to have space for teams to be able to actually do creative work, do explorations because you know that it's aligned with that high level vision. So if you can get that alignment and you can get those principles as the common language of what great software looks like, I think you can have a really good working relationship. And then the other bit I would just say is I think when to involve the founder CEO in a project is really important.

5. 14:14 – 16:53

   Thinking of involvement on a U-shaped curve

   1. NWNoah Weiss14:14

      And the short version I, I think that works the best is almost like a U curve where the X axis is time and the Y axis is level of involvement. I think you wanna get the founder CEO really involved early on, especially if it's a big new project, to make sure that there's strategic buy-in, you agree on the principles for how you're gonna approach it, you agree on the goals and the anti-goals. Getting that so then the team can run and explore. And then I think at the very end you want them to really be bought in that, did you build something that's up to the quality of our company? Is this something that's going to delight customers? Like literally taste the soup. What's missing in it? And I think at most companies that have a maniacally kind of customer-focused founder, if you don't do that last step, it's gonna be much more painful after you launch because they weren't part of that co-creation of the team. Uh, and, and so I think that kind of formula winds up working pretty well if you throw in that kind of alignment on principles and, and, and vision.

   2. LRLenny Rachitsky15:08

      That U shape sounds nice in theory, but I often imagine you get to that final step and this founder is like, "What the hell is this? This is not at all what I was hoping it'd be." Is there an example of that that comes to mind where you maybe went through that and then it's just like, no, that did not work out the way we expected? And if not, no problem.

   3. NWNoah Weiss15:26

      Yeah. I, I mean I think that does happen. The U shape is maybe the end of the U is kind of like the level of engagement and often that last level of engagement, that's where there's actually the most rapid refinement that you're doing.

   4. LRLenny Rachitsky15:37

      Mm-hmm.

   5. NWNoah Weiss15:37

      And I think what's important there is that hopefully you're refining in code and you're not still at like static design mocks because using the software is so different than looking at what the software will, uh, you know, visually appear. And so I think what we would wind up doing with, with Stewart at Slack for example, is like we would get the entire development team, engineers, design, product, user research, and Stewart together in a room and we'd kind of almost do like a bug bash together. And the idea was like, we'll do it all together, we're trying to make the best product possible. Making great software is really messy and we're all trying to kind of clean up the mess together. You know, sometimes you might find things like, okay, this entry point really isn't working. Maybe we have to move this entry point. That's maybe a bigger change. But I think often what you'd find is just all those bits of polish and refinement and doing the little delightful things that might otherwise be missing to kind of raise that craft bar and do it in a real collective way so it doesn't just feel like the team says, "We want to ship." And the founder says, "No, it's not ready." Ideally as a group you're saying, "We want to get it to a bar that's going to delight our users and here's the gap from where we are today to what we want to ship." I think that mentality winds up (laughs) being a lot more constructive, but that's not always easy

6. 16:53 – 19:32

   Principles at Slack

   1. NWNoah Weiss16:53

      to do.

   2. LRLenny Rachitsky16:54

      You talked about creating these principles, which is an awesome approach of just like creating guardrails for the team so they kind of think the way the founder and the head of product think. What are some examples of principles you have and had early on, maybe at Foursquare or Slack?

   3. NWNoah Weiss17:07

      I mean, Slack I think is where we kind of enshrined them much more 'cause we scaled the org so much, uh, more that we needed principles. And I think for us they were really about unpacking just the mission, which for Slack is making people's working lives simpler, more pleasant, more productive. That's the mission of the company. The question is, how does software help do that? That's what the principles are there to answer. So for us, we've got five core principles. They've, they've largely stayed the same. Some of the language has changed over the last couple years, but at least for the last four or five years we've had these. So the first is be a great host, which is all about kind of that level of craft, the relentlessly saving people steps. If you're, let's say a host at an Airbnb, it's like putting clean towels on the bed so no one has to wonder, are these for me? Like, that type of foresight.

   4. LRLenny Rachitsky17:57

      That's actually a value at Airbnb exactly. It's, uh-

   5. NWNoah Weiss17:59

      Oh, really?

   6. LRLenny Rachitsky17:59

      It's actually be a h- be a host at Airbnb is one of the four core values.

   7. NWNoah Weiss18:03

      Right. So maybe we borrowed that or someone was inspired by it, but be a great host s- sounded aspirational.

   8. LRLenny Rachitsky18:08

      (laughs)

   9. NWNoah Weiss18:08

      I love that.

   10. LRLenny Rachitsky18:09

       Yeah, yeah. It's a little (laughs) bigger.

   11. NWNoah Weiss18:11

       Th- there's a famous, uh, user design book called Don't Make Me Think, which we stole the title of for our, our next principle. And that's really just about...... as people building the software, you know how it works so well, you care about all the nuances and intricacies, and you really want your, you know, users to love it as much as you do. But often actually, that kind of owner's delusion that someone else will care as much about the software that you built as you do prevents you from actually making something that's simple, comprehensible, understandable. And so one of the core tenets, 'cause Slack is pretty complex under the surface, is how do we actually make people not have to think? How do we not reinvent the wheel if there's existing design patterns to use? How do we actually, uh, you know, wind up designing for people who come from many different backgrounds and we kind of cater to their needs in ways that don't make them have to customize it too much? There's a saying we, we also have, which is, "More clicks can often be okay." You know, you'll often have an, you know, optimization, you know, experimentation circles, like, oh, every click, remove it. But I actually think in a lot of software when it's not transactional, helping people understand what they're doing, giving them confidence, helping them have trust in the steps, we've seen that that can actually be a better experience. So, like, that's another example of don't make it stressful. Help people chill out when they're using the software. That's the idea behind that one.

7. 19:32 – 25:11

   Implementing ML, AI, and LLMs

   1. NWNoah Weiss19:32

   2. LRLenny Rachitsky19:32

      Shifting a little bit, I know you guys have been working on a bunch of AI stuff at Slack, and I believe you've been working on AI-related stuff for, for many years. I think at Google you worked on a lot of AI-ish related products. I feel like a lot of people are just getting into this and trying to figure out, how do we integrate AI and ML and LLMs into our product, and how do we not just waste our time chasing things? So, I wanted to ask you just in your time working with AI over the many years you've been doing it a- and share a little bit about what you've been d- doing there, what are some things you've learned about how to be actually effective and build valuable products and not just kind of fall for the shiny object issue and, and trap?

   3. NWNoah Weiss20:11

      I mean, it's almost 15 years ago now, but I was working at Google in Search on what later became called the Knowledge Graph, so this idea of building kind of a canonical repository of kind of information about people, places, things in the world, and relationships between them. And back then, it was a lot of the same idea. It's gone through the techniques and gotten a lot more mature. So we used natural language processing to extract all this information from the web and try to build this kind of database of facts. The idea then was, could you take queries people have, like what are the tallest mountains in Europe or what are the most popular beaches in Southern California, and be able to actually get answers and not just temporary links. Uh, I think the thing that's really changed and super exciting, you know, in the last six, 12 months with LLMs and ChatGPT and everything else is the idea that now you can take not just knowledge about the world but actually have natural language generation, where suddenly the, you know, computer can kind of talk back to you in a way that feels extremely human. And then the creative e- you know, applications of that are pretty massive and exciting. So, that's kind of, I guess, the lineage there. I think from over the years back at Google, at Foursquare we did a lot of personalization and recommendations. At Slack we have, you know, search and ML that's kind of been fused to the, the product. I think a couple things come out as kind of, I guess, maybe principles that we've kind of used over the years. Back then at Google, one of the big ones was that the promise of the UI has to match the quality of that relying data, which is to say... And I think this is actually one of the failings of the various LLMs right now is, uh, they all appear supremely confident even when they're completely hallucinating.

   4. LRLenny Rachitsky21:53

      Mm-hmm.

   5. NWNoah Weiss21:53

      And I think that's going to be something that people are gonna have to work on a lot, which is to figure out how to be not so faultless, to acknowledge when you're not sure, because otherwise it undermines the trust people have in the system. Using a lot of transparency about where the data comes from so people can actually build credibility in the tool is really important. And then I think making sure that as you're designing the products that you have virtuous cycles that are naturally part of the project experience where you can get training data as a byproduct of people naturally using the software, and then can make the model that you're building behind the scenes smarter, more accurate, and more predictive.

   6. LRLenny Rachitsky22:28

      Mm-hmm. Yeah.

   7. NWNoah Weiss22:28

      So, you know, the classic example of that would be Netflix back in the day. Their, their rating system, they actually had a feedback loop from their customers to then make the system better at predicting. Uh, and I think people are still trying to figure out what does that look like in this world of LLMs.

   8. LRLenny Rachitsky22:42

      Something I hope that you're all building at Slack is a way to ask a bot questions based on all the conversations in the Slack. I've been looking for that product for a while now.

   9. NWNoah Weiss22:51

      I can safely say we have a lot of prototypes internally where we were playing with this.

   10. LRLenny Rachitsky22:54

       Mm-hmm.

   11. NWNoah Weiss22:55

       And I think... It's actually funny. As a aside, in one of the original Slack, I don't know, product vision decks back in 2014, there was our kind of whole strategy in four parts. And then part number four, which was a joke at the time, was then do magic AI stuff on top. And I know we didn't even know what the state of AI would be by the time hopefully companies had their collective knowledge in Slack, and now we're finally at the period where, uh, magic AI stuff seems finally pretty amazing and pretty magical. Uh, so yeah. We're, we're doing a lot of prototyping internally and, and also trying to work with the ecosystem around as well 'cause there's so many companies doing amazing work in this space, so that if you work at a company where you have so much knowledge in your kind of Slack channel repository that you can suddenly get amazing leaps in productivity to help you better do your job, because that knowledge is in Slack but it's sometimes hard to reach. And I think these technologies can make that possible.

   12. LRLenny Rachitsky23:50

       This reminds me of something Gustav, the CPO and CTO and co-president of Spotify, shared that they always have a deck and a vision of just, like, a play button within Spotify. You just play and all magic happens, and it's the best music and thing, exactly what you want to hear, and just how that isn't actually possible, and it's still not possible. And so t- exactly to your point, you have to like really think about how does act? Like how close is it to the reality? And if it's not actually there... Like, he's like, he was saying how like, "We'll pick two songs that are correct out of 10 just 'cause we don't really know exactly what you want to hear right now." And it's just there's no point in trying to design that right now 'cause it's not actually gonna be, uh, delivering on the promise.

   13. NWNoah Weiss24:27

       Right. Yeah. I think our... I love that. Our version of that has always been that...... you open up Slack and, suddenly, instead of having to read through dozens of channels or find all these mentions, that magically Slack could just tell you in the order that you would care about kind of a summary of all the interesting things that have happened. And then let you dig in if you want to. Like, your very own kind of, like, personal chief of staff who knew everything that you cared about and read everything that you could read. Uh, I don't think that's gonna quite be possible anytime soon, but I think, like Spotify heading towards that North Star, you wind up developing, uh, I hope a lot of really compelling product experiences along the way.

   14. LRLenny Rachitsky25:02

       Yeah. Man, the more I think about it, the more, the more amazing opportunities exist in Slack. It's, it's like all texts. Amazing. Okay. There's a lot of cool stuff coming, I imagine.

   15. NWNoah Weiss25:10

       Yes.

   16. LRLenny Rachitsky25:10

       I can't wait.

8. 25:11 – 26:59

   How Slack structures AI teams

   1. LRLenny Rachitsky25:11

   2. NWNoah Weiss25:11

      Yes.

   3. LRLenny Rachitsky25:11

      On that, on that topic, how do you think about creating teams within Slack and AI? Specifically, are you, like, recommending each team think about how AI can make their stuff better? Or are you dedicating, here's the AI team and they're gonna work on stuff and you guys just keep shipping what you're shipping and keep moving your metrics?

   4. NWNoah Weiss25:29

      Maybe the unfair answer is a hybrid of the two, which is to say we have a kind of central machine learning and search team where there's a lot of people who have expertise in this field to build infrastructure that everybody can use. And what we've done is, uh, because Space is evolving so quickly, like literally every month, like, the capabilities are evolving, the risks and trade-offs are evolving a ton. What we want to do is actually kind of spin up a couple different teams that are focused on prototyping using that common infrastructure, but in specific directions that are all a little bit different. So we've got a common ML, let's say, search team, and now we have a bunch of teams that are kind of working in parallel in different kind of customer problems that we're trying to solve using that shared infrastructure. So I, I think this isn't the steady state. I think over time what it'll probably look like is that all the existing product areas, as soon as we kind of know more of the shape of what the technology is capable of, will just have, you know, AI capabilities as part of their roadmaps just like every product team is responsible for their own mobile roadmap. They don't have, they don't outsource it to someone else. But I think today with things are moving so quickly, you actually want a little bit of a more kind of ad hoc, flexible approach to move quickly. And that's what we're doing.

   5. LRLenny Rachitsky26:48

      That's kind of what I've been hearing from everyone I've been asking this question. The search ranking team is, always seems to be the center of all this and then it's a few experiments here and there. So that's an interesting pattern I've been noticing.

   6. NWNoah Weiss26:58

      Good to

9. 26:59 – 30:01

   Complaint storms

   1. NWNoah Weiss26:59

      know.

   2. LRLenny Rachitsky27:00

      I heard that you have a process internally called complaint storms and I'd love to understand what that is.

   3. NWNoah Weiss27:05

      It's something that started, I wanna say, back in 20, end of 2019, maybe early 2020. And the idea a little bit was how do we help as a team look at the software that we build with fresh eyes? 'Cause we've been starting at Slack for a long time, and Slack maybe more than almost any other company, maybe like Figma is probably similar. I was listening to the podcast just earlier today where if you work on Figma, you work on Slack, you also live in Slack and you live in Figma all day so you can become more of a power user than anyone else on Earth. And what we were realizing, especially for people trying to build Slack for the next million customers, the people who have never used Slack before, it was becoming increasingly hard to kind of have empathy for what their usage of Slack would look like. How would they look at it in a more critical way? How would they care less that we cared? And so what we started doing was these complaint stores and the idea was really simple, which is we'd get a team together, uh, often Stewart or myself would also join, and we'd actually start off with other products first, like in adjacent spaces. And we'd say, okay, as a group we're gonna go through, like, the customer journey from the moment you land on the website through, let's say it's a, you know, workplace product, getting your first account going, getting the first couple users on board, getting to the point of value. We're gonna do it on one screen and someone's gonna project and then people are gonna fill in every issue, everything that's confusing, every pain point. Not bugs, but ways in which if you didn't care about this software, you don't work on it, what would actually confuse you? What wouldn't stop you in your tracks? And from that you wind up generating a bunch of amazing inspiration by looking at someone else's product in a really critical way for things you might wanna try in your own product. Once you get to that, then it becomes easier to actually do with your own software, but it, it is a little painful obviously. Sa- same with watching usability tests to, like, look at your own, you know, baby in a way that is, okay, I'm trying to find all the warts, I'm trying to find all the problems. But that's wound up being a pretty great source whenever a team I think either gets stuck or feels like they reached like a dead end in a direction, is doing complaint stores about the product area that they're in or using adjacent products just to get inspiration. Uh, and then I think it kind of unlocks a lot more kind of creative views on the problem space.

   4. LRLenny Rachitsky29:23

      It's similar to a process that I learned Stripe has called friction logging, but I love, I love the nuance here of starting with someone else's product 'cause I could totally see how that makes you feel better looking at your product in real l- like, it's not like we suck. It's okay, everyone's has so much opportunity.

   5. NWNoah Weiss29:38

      Exactly. Yeah. I've heard that from Stripe too. It, i- I think gets a similar place and I think it's th- the doing it, I think the byproduct is that you also get, like, calibration on product taste, product quality, and as a team you kind of develop that together. Again, similar to the principles, it's like how do you get these things that are kind of hard to actually feel collectively on the same page about and how do you calibrate? It's another good way to do it.

10. 30:01 – 32:26

    Slack’s approach to prioritization

    1. NWNoah Weiss30:01

    2. LRLenny Rachitsky30:01

       I'm imagining some PMs might be hearing this and wonder, okay, great, now the founders and the execs have all these things that they want us to fix. I have, like, goals to hit. I got a roadmap. How do you think about prioritizing things that come up in these sorts of sessions for the team and how do they mix and match versus all the other stuff they wanna do? Or is it just like they don't actually have a huge roadmap and this is a way to inform the roadmap?

    3. NWNoah Weiss30:23

       No, I mean, I think more broadly, I think the way that we think about, or I like us to think about our roadmap for any feature team at Slack is that it's a portfolio.And it's meant to be a portfolio that's diversified a couple of different ways, right? I think one is you want to diversify things that are meant to be new capabilities, versus making the thing you've already built a little bit better every day. Similar to parenting. Uh, or they are things that are meant to be risky, that you aren't sure are going to work, but might have a lot of upside. There's things that are kind of known bets, and then I think often you're kind of balancing, are you doing things that are meant to have impact that you're already very confident in, versus things that are meant to learn about a new s- you know, possibility space. And so I think for most teams, this stuff usually will end up tactically filling up that bucket of let's make the existing product a little bit better every day for users. And at Slack we, we have this thing that we call customer love sprints, which is an interesting way teams have figured out how to get this on their roadmap, is it's hard to allocate that work throughout the quarter. So what we want to be doing often is have a team do a two-week customer love sprint, almost like a hackathon, but with that kind of burn-down list of what we think is the lowest effort, highest impact, uh, changes we can make to generate more love from our customers and whatever that feature is. And then pe- people just sprint for two weeks, design, product engineering, and then you have a bunch of things that you celebrate at the end, and the goal is to ship all of them. So this isn't like hacks that you throw away. So that's kind of how we want to prioritizing it off in that kind of work is actually kind of making it this really fun total change of pace throughout the quarter. Uh, to not do big feature work that may take months, but to do all these small, delightful things that you know customers are gonna love at the end. So that's the other way that we kind of figure out how to balance it in.

    4. LRLenny Rachitsky32:11

       I love that. And how often do you do these sorts of customer love sprints?

    5. NWNoah Weiss32:14

       I would think teams that work on very user-facing products do it at least once a quarter. So I think other teams that work on maybe less user-facing might do it maybe, you know, twice a year. But quarterly is a pretty healthy cadence.

11. 32:26 – 34:41

    How delight is baked into the DNA of Slack

    1. NWNoah Weiss32:26

    2. LRLenny Rachitsky32:26

       Wow, I didn't know about that. And that kind of connects to... Slack has always been a very delightful product. I remember early on the animations were so awesome, the, uh, little twirly, I don't know, pounds hashtag thing, and it feels like Slack has always invested in delight. How do you operationalize that? How do you... Is it, is it these customer love sprints? Is there something else that's just like we need to allocate some percentage just to like make things really fun, even though it's not gonna move any metric?

    3. NWNoah Weiss32:51

       I would say it's a little bit the DNA of the company, honestly, which is that the four co-founders were trying to build a massive online role-playing game for many years that was called Glitch, and their kind of background was all in like building delightful, playful experiences. Glitch didn't work out, but you know, there's a whole long backstory. But the short version is a tool they had built internally that they then wound up spinning out a company move from, which became Slack. I think that DNA of we're trying to build a consumer-grade experience that just happens to be for work is really ingrained in the company. It's also a big part of how we hire. I would say certainly the majority of PMs, designers and engineers who joined Slack had never worked at an enterprise software company before. It's not like most people have worked at Oracle or SAP. It's most people have worked at consumer companies or game companies, and so they bring that focus and spirit. And then I would say the, the last bit beyond kind of the principles and the com- complaint storms and the customer love is that we have this amazing team that we call the CE team, the customer experience team, and they're kind of in some ways the team that is doing our scale support, but is most often in touch with our customers. And from the very early days, you know, people used to do CE shifts if you worked in product so that you can actually figure out what's frustrating, what's confusing. And we have a really great kind of pipeline for getting the insights from the CE team of what are the obstacles, the pain points, the most frequent complaints into the hands of the product teams to be able to prioritize, to figure out, yeah, not all these are gonna move a given metric. They might not achieve something for the business. But collectively, I think the way that Slack thinks about competition is we obsess about customers. We build something they'll love enough to tell their coworkers, and the rest kind of take cares of, takes care of itself.

12. 34:41 – 38:04

    How Slack thinks about competition

    1. NWNoah Weiss34:41

    2. LRLenny Rachitsky34:41

       Speaking of competition, something I wanted to ask you a bit about. So early on, Slack was competing against this product called HipChat, and that was actually what I used at our startup (laughs) . And w- we loved HipChat. It was so hilarious, just these memes everywhere and their billboards were amazing. But then Slack ate their lunch. Later on, I'm just kind of thinking out loud, Discord feels like that was the big threat, and now Microsoft Teams obviously. I'm curious just how you think about competition and even just what you've learned about working in a space where there's a lot of competition and thinking about that long term and even short term.

    3. NWNoah Weiss35:12

       Yeah, I mean each of those is kind of like an interesting mini kind of lesson learned about those. And I think the through line for all of them, I would say is still the, the maxim that we have internally, which is we're customer obsessed but competitor aware. Uh, so I think it's a little bit different. I think some companies are like, I don't know, Uber for example, I think was notorious, like competitor obsessed and yeah, they tried to kill their customers when they could. I think HipChat, I don't think Slack sought out to like kill HipChat. At Foursquare we used, I think it was called Campfire back in the day from the 37signals people. So there's a whole generation of those products and I think Slack came along and I think they had a couple of innovations. One was they had a great mobile experience that synced across every client, search actually worked, and then they brought a lot of the best parts of like consumer messaging, uh, into the workplace like with emoji and reactions and all those bits. And I think it turns out that, you know, if you're 10x better on a couple of those axes, then you can see a huge change in behavior. And so I think that's what happened with that move from like the HipChat, Campfire to Slack world. Discord's interesting. I mean, we keep aware of Discord, but it is so much more focused on the kind of consumer... Originally it was gaming, now it's more community space. And I think at Slack the lesson I would have, uh, I think we learned in a good way is we've always really been focused on groups of people who are trying to do work together, and that winds up being a completely different audience to build for than communities. And so I think that focus has been really helpful and I think...... Discord is amazing and many people love it. And the people who use Discord certainly use it in a very different way than people who use Slack at work. I think Microsoft obviously has become, over time, the biggest competitor there. I think, you know, the origin of Teams really was a defensive move for them to protect Office, because Office is an incredible, very profitable kind of monopoly in the, in the productivity space. And so, I think when they built Teams, it was more of a kind of covering their flank versus Slack kind of on the ascent. I think as Teams has evolved over time, it's become much more of a, a video conferencing product that competes with, like, Zoom and Google Meet. Uh, the people who use Teams use it completely different than Slack, where you live and breathe in, in channels and work in kind of workflows...

    4. LRLenny Rachitsky37:24

       Yeah.

    5. NWNoah Weiss37:24

       ... all day long. And I, I think what we've seen there too is that, like, a lot of our customers, they happily use both. Most Fortune 500 companies have either an Office subscription or a Google Workplace subscription. And all of those customers who use those, also use Slack. And we like to say that Slack is this connective tissue that makes all the rest of your tools that much better. So, I think there, we've kind of taken very much an open ecosystem and platform approach. And, you know, we've just been focused on how do we keep building the best version of what Slack can be as a new category of software for our customers. And, you know, staying aware of our competitors, but really obsessed on one of the new ways that we can delight our users as the years go by.

13. 38:04 – 41:40

    Building a culture that takes big bets

    1. NWNoah Weiss38:04

    2. LRLenny Rachitsky38:04

       So, Slack is kind of a big-ish company within now a, let's say, a big company. But it feels like you still are launching really interesting stuff. You launch Huddles, uh, Clips. There's this AI stuff coming, sounds like. I'm curious to what you have done at Slack to enable these sorts of zero to one bets, and what you've seen is important to allow for sort of, for innovation along those lines.

    3. NWNoah Weiss38:26

       I, I think maybe we're all a little self-delusional 'cause I think everyone who works at Slack likes to think that we're still at a small startup. And I think keeping that spirit alive, honestly, culturally has been a big part of it. You know, I think going back to the principles early on, one of the ones that we didn't talk about, literally one of the, the actual wording is, "Take bigger, bolder bets." And the idea there is that it's really easy to fall into the trap of just constant incrementalism. You know, the concept it's like a feature team and you have like a KPI and you feel like your whole life is measured by that single KPI going up 1% a quarter, and then you kind of lose sight of what's beyond the horizon. And so we have this kind of mental metaphor that we talk a lot about getting to the next hill. And the idea is that if you're in a mountain range and you're maybe in a little valley, you can kind of see what's right in front of you, but you have no idea how tall the mountains are behind. I think teams can often get lost kind of crawling up that hill, not realizing that there's a huge, uh, incredibly beautiful range behind it. So, take bigger, bolder bets, get to the next hill to see what the horizon looks like around you. That's kind of how we think about it strategically. And then I think structurally, the way that we've approached it is that we've, over time, created kind of new teams from scratch that incubate in a new area before the area's mature. So, we did that with a lot of these kind of native audiovisual products like Huddles and Clips early in the pandemic because our customers were demanding it from us. They were like, "We love living in Slack all day, but we feel disconnected from our teammates when we can't be in the same physical place. Like, what can you do to help us?" And that's where that came from. And I think in the AI space now, it's a similar thing, which is what we're trying to hear from customers, like, "What do you wish Slack could do if it had these new superpowers? And let's incubate a couple teams, a prototype there, and then figure out what can get to real product market fit." And I think when we have those teams, I think it's important to just give them space to run, to give them kind of a, get-out-of-jail-free card for maybe the normal process of, you know, okay, our planning quarterly reviews and make it feel something that is like the pace of learning is what matters. Like, how fast are you prototyping? How fast are you learning from users? And then getting to do that publicly and, and pilot and then get something to launch that's amazing, blows people away. That's kind of the formula that we've seen.

    4. LRLenny Rachitsky40:43

       This episode is brought to you by Vanta, helping you streamline your security compliance to accelerate your growth. Thousands of fast-growing companies like Gusto, Calm, Quora, and Modern Treasury trust Vanta to help build, scale, manage, and demonstrate their security and compliance programs, and get ready for audits in weeks, not months. By offering the most in-demand security and privacy frameworks such as SOC 2, ISO 27001, GDPR, HIPAA, and many more, Vanta helps companies obtain the reports they need to accelerate growth, build efficient compliance processes, mitigate risks to their businesses, and build trust with external stakeholders. Over 5,000 fast-growing companies use Vanta to automate up to 90% of the work involved with SOC 2 and these other frameworks. For a limited time, Lenny's podcast listeners get $1,000 off Vanta. Go to vanta.com/lenny. That's V-A-N-T-A .com/lenny to learn more and to claim your discounts. Get started today.

14. 41:40 – 44:51

    Rituals at Slack

    1. LRLenny Rachitsky41:41

       One of the things I love learning about from product teams is their unique rituals and, uh, traditions. And I'm curious what's maybe the most interesting or unique or fun or funny ritual or tradition on the product team of things you all maybe do regularly?

    2. NWNoah Weiss41:56

       One of the things that we do, which, which is kind of always a little bit funny, I mean, it's more of a, like a emotional thing rather than a practical thing, is that at all hands we'll often wind up taking, like, specific tweets that people had about their product. And you know, Twitter, people say the craziest things sometimes. Uh, and sometimes they're like really heartwarming, like customer love, but often it's just the, the meanest, most frustrating complaints that people have. And it's honestly meant for us to just have a pulse on like, "We're a..." People are actually saying and feeling in the wild. And not taking it too seriously, but keeping that sense of, you know, I think that the distance you have from your users as your user base gets more and more diverse and larger, I think can, can make it harder to actually s- develop the product because you're not designing for yourself anymore. And so I think all the ways that we help keep people grounded in like what are actual users actually saying, um...... that's, that's one big way. And then the other that reminded me of, which is actually probably better maybe delete that last one because it's kind of boring. Um-

    3. LRLenny Rachitsky42:55

       No, it's great. We're not deleting nothing.

    4. NWNoah Weiss42:57

       Fine. Uh, you know, user bel- so I'm a big believer in you wanna be data, you know, informed, but you don't wanna be so data-driven that you actually don't have a pulse on what real people feel when they're using your product. So we're really big into user research, not only as it gives you the answer, but it helps at least pose a lot of questions for you when you watch how someone actually uses the software. And historically, it's really hard to get PMs, let alone engineers, to actually like attend user research sessions. And so what we wound up doing, this must've been in the pandemic when we first went remote, is, you know, now you can dial in to usability sessions. And to make it really interactive for the team, what we would do is have people live kind of, like, in a thread write their real-time thoughts of like, "Oh, so painful how they used that," or, "I can't believe they missed that." Or, "Well, that gave me this idea from seeing how they were doing that to do this other thing." And so then you wind up having the PMs, the engineers, designers, and the user researcher all in one Slack thread, like live responding and reacting to a usability session. And then suddenly, that thread becomes actually the best kind of source of truth for the research report that then gets written up. But I think most importantly, it gets the team almost like the complaint storms, but actually watching someone else do it, like, in the shoes of an actual human being trying to use the thing that you thought was so brilliant and yet has all these flaws. And it's humbling, it's filled with humor, and also it's, I think, really constructive for the teams to do it that way.

    5. LRLenny Rachitsky44:27

       I was gonna ask where they actually s- share these thoughts and in Slack makes a lot of sense.

    6. NWNoah Weiss44:31

       Yeah, I mean, it turns into a r- report at some point, but literally just link back to the original thread and then you have like, you know, a hundred people's reactions as, as the, you know, report is kind of ongoing.

    7. LRLenny Rachitsky44:41

       If only there was a AI tool to summarize all of your thoughts.

    8. NWNoah Weiss44:45

       We've got a prototype for that. Hopefully, (laughs) it'll work well enough that actually it'll be useful for customers too.

15. 44:51 – 52:01

    How Slack revived their self-serve business

    1. NWNoah Weiss44:51

    2. LRLenny Rachitsky44:51

       You tweeted once about how, I think maybe around the time you joined Slack around 2019, that the self-service business of Slack basically plateaued, and it wasn't clear why. I'm curious just what that period was like, and how did you kind of get to the bottom of what was going on and turn things around?

    3. NWNoah Weiss45:09

       Yeah, it was actually a couple years after I, I joined, but it was a point when I was kind of focused on the self-service business because we had this period with Slack where I would say maybe 2014 to 2017 where it was almost all self-service, and it was just growing like gangbusters. And then we started spinning out the sales team and an enterprise team. We started focusing mostly on that. And I think we kinda, you know, we still have a team that was working on self-serve, but it was primarily the company's focus was all driving enterprise deals, kind of getting to that next level of maturity. And then in 2019, I think we started to see that when you look beneath the surface, the, you know, fundamentals of the self-service business weren't looking as healthy as they used to be. I think kind of the biggest thing as we kind of dug into it was a little bit to what we were talking about earlier with the motivation and complaint storms is it was getting harder to understand what the next generation of Slack customers really wanted from the product. And whether you're thinking about this as crossing the chasm or moving from like kind of early adopters, the needs of kind of the more majority or later adopters, I think we're at that point where not every technologically sophisticated company on Earth was using Slack, but most were. And we were getting into a market that customers just had different needs, they had different levels of, uh, sophistication. And so we did a lot of user research. We looked at all these cohort curves which you can imagine. Suddenly they're like, huh, they're, they're not as healthy as they used to be. Like, what's going on? And I think, you know, we got that, got a bunch of insights from it, but I think really what we wound up changing about how we were operating was instead of to continue to try to optimize the things that had worked over the last couple years, we said, okay, let's kind of throw the whole roadmap away. And instead, let's come up with a bunch of hypotheses about what could be new levers that could actually help based on the insights that we now have about the next set of customers, and we're gonna try to quickly learn which of these levers are real and which of these are just totally off the mark. And, you know, we kind of had to say for the next like six months, we're probably not gonna drive any impact at all. It's only gonna be about learning. But at the end of that, hopefully we wind up finding a couple different levers that had years of room to run, and that's what wound up happening. We wound up kind of doubling the rate of our new paid customer growth in the year, in the couple years after that, and kind of re-accelerating the self-service business. And I think it really came from stepping back, being humble, not feeling like we deserve to have every company on Earth sign up, and then figuring out how to optimize for learning so that in the long term, you can get the impact, but knowing that for the next couple quarters, we're gonna sacrifice impact for the sake of learning. And I think that was a good muscle to build, but it was definitely not easy to do at the time.

    4. LRLenny Rachitsky47:49

       Well, this story begs the question, what are the levers that worked? Whatever you can share.

    5. NWNoah Weiss47:53

       One of the big things that we wound up focusing on is what we talk about as in, as comprehension desirability. So the fundamental challenge I think for new users or new teams using your product, once you get past the kind of tech early adopters is, do they comprehend what this thing is for? Do they understand how it works? And then desirability is, why should they care? You know, most people at work are not like, "Hey, you know what I wanna do today is start using an entirely new tool and convince all my coworkers to get on board." That is not like part of your job. Your job has goals and measurements and everything else. So really deeply understanding that and how do you push on that in that new user experience? It sounds maybe a little ludicrous, but Slack has always had a, a freemium product. Obviously, there's a free tier that you can use, but we had never actually figured out a trial strategy where we actually gave you a taste of the paid product. Either were on the free tier or you have to pay for the paid tier. And that wound up being one of, I think, the ripest veins is figuring out how to give people a taste of the full premium Slack experience so that they would never wanna go back...... and doing that at a variety of different points in the customer journey. And then I'd say the other big thing I would call and point out is we really needed to figure out a new North Star metric for motivating the teams across Slack. At that point in time, we basically had paid customers and then we had creative teams, which is like the very, very beginning, very, very end of the journey. And we did a lot of quantitative research and data science and wound up coming up with this new metric we called Successful Teams, which is a little bit, you know, I, I feel like a lot of companies have this. Like Facebook's, I don't know, lucky number seven or whatever it was, where what we found was that if you could get five people using Slack the majority of the work week to just communicate at all, that would be a successful team. They are gonna be 400% more likely to upgrade over the next six months. And that seems like a very low bar, like five people to use Slack throughout the work week, not even every day. But it turns out that if you can get that level of critical mass, kind of the rest will take care of itself. And we wanted motivating not just the team that was focused on self-service, but all these other feature teams across the company to drive more new successful teams knowing that if we can move that, which is much earlier in the funnel but now a top of the funnel metric, then it would actually drive upgrades and paid customers and thus revenue long-term. And that was a huge kind of turning point for how we rally product teams around something that would actually drive that self-service business.

    6. LRLenny Rachitsky50:22

       Man, this feels like its own podcast just to analyze the things you learned down this journey, and there's so many takeaways here. One is just the importance of an activation metric that is predictive of retention. So it sounds like you landed on five people in a company, like DAU basically for a week, something like that. That's awesome. And then the other interesting takeaway here is I'm actually doing a bunch of interviews with founders of the most successful B2B companies, and interestingly they all, not all, maybe half are like, "I still don't think we have product-market fit." Like they're at like a billion dollars valuation, growing like crazy and they're like, "I feel like I have product-market fit with the current users, but I don't with the people I want." And that's what you're describing, right, is like the new, you stopped having product-market fit with people that you wanted next.

    7. NWNoah Weiss51:07

       I think that's exactly right. I think of like product-market fit as almost like you keep stacking these S-curves where you get product-market fit in a small group and then you suddenly reach like exponential growth because you can crack that whole group, that type of audience. But then you start declining because you start hitting the ceiling of like we've got to, I don't know what it might be, every development team in the US to be using this product. And then you jump up to the next S-curve, which is like how do we get technology savvy teams that aren't developers? Or how do we get people who are, you know, in large enterprises who are outside the US? And these each kind of beat... These each become new curves that you have to build product-market fit for, and I think it's just all a, a huge exercise in like being self-critical, being humble, not, not presuming that you've cracked this thing forever and keeping kind of a very beginner's mindset of what does the next audience need that your previous audience didn't need at all?

16. 52:01 – 58:08

    Slack’s early success

    1. LRLenny Rachitsky52:01

       If you think about the pie chart of what you had to change to make it work, how much of it was like messaging, positioning, onboarding, optimization, versus like product features?

    2. NWNoah Weiss52:10

       I would say maybe 60/40 in the sense that uh, uh the early journey, I mean, uh not just obviously positioning and messaging but like the entire experience of like unboxing Slack if you will with your team. Uh, you know, we, we called it the day one journey but it extended to really kind of like day 30 in re- in reality, and it's a single player and multi-player experience, so it is really complex. But then I think what we realized was you can make that incredible, but if fundamental parts of the product were missing that would make it comprehensible to the next audience, then you're gonna have problems. So like it sounds maybe, uh, impossible to remember but Slack used to not have WYSIWYG message composition. You used to have to use Markdown.

    3. LRLenny Rachitsky52:55

       Mm-hmm.

    4. NWNoah Weiss52:55

       And so making that WYSIWYG was a huge boost. Making mobile work offline so it worked no matter where you were in the world was another big one. All the things about configuring your sidebar notifications so that as you scale the usage of Slack it didn't become overwhelming. Th- those are some of the kind of foundational product investments that we wound up making so that that next generation of Slack customers could get value and, and not be overwhelmed or daunted by it.

    5. LRLenny Rachitsky53:22

       Maybe one last question along these lines. People look at Slack as kind of the, maybe the first major product-led growth success story and they always look to Slack like, "Oh, we just wanna grow like Slack. Let's see what they did." For people that are studying Slack's journey and success, what do you think Slack did right early on that maybe people don't recognize or don't appreciate enough that founders today should be thinking about moreso versus just like, "Let's just make a freemium product and..."

    6. NWNoah Weiss53:48

       Right. I mean, I think maybe the most telling thing is when Slack started, certainly when I joined still I don't think the word or acronym product-led growth existed, so it wasn't like we were really good at taking this playbook and applying it. I think it was more that whole term of art became a thing as maybe many other kind of freemium SaaS products kinda took off. I, you know, I really think, uh, not to be repetitive, but I, I think the core of it really was building a product that customers loved enough that they would put their own social capital on the line to get their coworkers onboard and that was easy enough to use and get the value from that without ever talking to a salesperson you could put a credit card down, or expense it if you, if you wanted to, for just your team. And I think when people think about this product-led growth notion, I think there are really two very different audiences, and I think Slack was able to crack both. One is when your team is small and your company is small, it's, it is the entire company if you're an SMB, and I think that's almost like S- Slack's sweet spot. When the original pitch deck came to the investors, they said, "Slack is for companies with five to 50 people."... at the time, the biggest company to imagine using Slack was 50 people, because they're like, "Ah, I don't know how this is going to work beyond that. You know, it'll become pandemonium." Uh, obviously, that was the initial, I think, real, real strong product-market fit. But the other bit, which then was what powered the enterprise business, was teams of 5 to 50 people who worked at larger companies. And I think what wound up happening was that you would have teams of those independently at a company like IBM or Disney or Capital One or whoever it might be, or Comcast, discovering Slack, using it for themselves because they thought it would just make their working lives simpler, more pleasant, and more productive, and maybe not even know that anyone else at the company was using Slack. And then by the time we then scaled our enterprise sales team, I mean, truly, the exercise initially was just take customer domain, sort by number of active users, and call them in the order of, of that, which is, you know, "Hey, by the way, you have a couple thousand people actually using Slack at your company. Like, do you want to think about a broader deployment or controls or analytics?" And so I think that that was it, that consumer-grade experience that customers love enough to get their coworkers on and pay for themselves. And then at enterprise companies, like, s- having a bunch of different flowers sprouting so that eventually you could roll out that enterprise-wide kind of deal. And then it was all the tactics. But I think that, that was where it started.

    7. LRLenny Rachitsky56:20

       The way you described at the beginning of, make a product that people want to share with their colleagues reminds me of a ... I was just listening to an interview with Seth Godin, who's a, you know, marketing legend. He, I think he has a new book, so he's on every podcast. And he, he had this really great quote that the products that win are ones that you want to tell your friends about. And it's a really simple concept. And basically, it's like, it's word of mouth is how you have to win. But, but I think that's so true. And s- like, every successful company I talk to ends up being like, "We just want to build something people want to share with their friends." Even if it's growing in some other way, SEO, paid, feels like that's always at the root of it, is you just, like, want to tell your friends about it because you love it. And Slack, I think, is a great example of that.

    8. NWNoah Weiss56:58

       I, I think that's true. And I, I mean, obviously there are categories of enterprise software that that isn't true for, like in security or maybe-

    9. LRLenny Rachitsky57:05

       But even then, I think-

    10. NWNoah Weiss57:06

        ... infrastructure, right? Yeah.

    11. LRLenny Rachitsky57:07

        Like, I think they s- if it's an awesome security product, you're like, "Hey, you got to check out this, like, Sentry or whatever, or Snyk."

    12. NWNoah Weiss57:13

        Yeah. Like, uh, um, good friends with Ivanti CEO, Christina, and I feel like they've-

    13. LRLenny Rachitsky57:17

        Yeah.

    14. NWNoah Weiss57:18

        ... run into those stories where whoever would have thought, like, a compliance company would be something that people raved about to their other startup friends. Like, "Oh my God, you don't want to deal with, like, SOX Compliance? You go to Ivanti. It's amazing." So yeah, maybe that is true. I, I think, especially in this day and age where all the marketing acquisition channels have been so saturated, people are optimizing so much, I think it's really hard to, like, scale a big enough business if you don't have some amount of word of mouth and customer love-driven growth. I think it's hard to just scale it on, like, we're going to just play the cat game in, in hopes that the numbers work out.

    15. LRLenny Rachitsky57:51

        I remember e- uh, Slack rolling out at Airbnb and all the designers getting so excited about it, creating their channels and everyone's just like, "What the hell are they? What is this thing?" And then it, it did exactly what you're describing. It just spread. Everyone's just like, "Whoa, this is cool." And they're all telling each other about how useful it is to them and spread like crazy.

    16. NWNoah Weiss58:07

        I love

17. 58:08 – 1:02:03

    Slack’s pilot programs for testing new features

    1. NWNoah Weiss58:08

       that.

    2. LRLenny Rachitsky58:09

       Is there anything else on Slack that you think would be interesting to share in terms of what makes it a successful product team, product business before I move on to another topic?

    3. NWNoah Weiss58:20

       The only other thing I think is maybe a little bit interesting in terms of how we develop product, and it's, it's really different and it's changed over time, which is that obviously the easiest person to build for is yourself. And the next easiest is people who look almost exactly like you or, you know, have similar preferences and sophistication. And I think in the early days of Slack, that's basically what we did. I mean, it was like really just trying to build for small, technologically-savvy teams. And it turns out, you can build a pretty big business making a great product for them. Over the years, obviously, that's changed. And, and so one of the things I think that we've done, which has worked really well, one, obviously, is we figured out how to do experimentation in a SaaS product, which is not always obvious because the metrics are much, uh, longer term than, you know, you land on a checkout page and then y- you hit checkout. But I think the other thing is we figured out how to scale up getting real customers using Slack in the wild for new functionality. And so we have this really robust program that we call kind of our pilot program, where we have, I don't know, probably thousands of different customers that have all signed different agreements now where we can actually roll out to progressively larger user bases. And because Slack is a multiplayer product, you, you often have to roll out real net new functionality to a whole company or whole team, because otherwise, you know, you can't use huddles by yourself, for example. And then we have a really great program for actually getting feedback from those customers, both through Slack Connect itself, through surveys. And this winds up being kind of the lifeblood of feature teams, where you can, by the time you actually launch a big net new feature for Slack, have gotten so much customer feedback from people actually using it in the wild to get work done, and so much more confidence in what you're building from the metrics and the surveys that we do, that, you know, you can't guarantee it's going to be a hit. But you can be really confident, not because it just worked well internally, which is no longer that predictive, but because it worked well for 1,000 different companies, 50 different countries, in 20 different industries. Uh, and so I, I think, you know, n- early on, SaaS companies don't need to figure that out. But I think as you grow and as you have a more diverse customer base, as, you know, you said all these SaaS founders who said, "Hey, you got to, like, keep reestablishing your product-market fit," I think that is, like, a programmatic way of being able to do that with your product development process that's pretty interesting.

    4. LRLenny Rachitsky1:00:37

       Any tips for how to choose who to include in this group if someone wants to build something like this for themselves?

    5. NWNoah Weiss1:00:43

       I think the two most important things are you want a lot of diversity in terms of industry, company size, location, and so on. And then I think you want to pick people who are actually motivated, uh, to want to be part of the development process and have a slightly higher risk tolerance. Not every company wants to actually be beta testing new functionality that might get removed. So making sure we have kind of a, this, like, champion network that we built of people who...... loves Slack enough that they are willing to put up with a little bit of pain in that rougher period, or are willing to have something that they tried to use and then we decided actually we're gonna kill that feature before we ever ship it to everybody. So, diversity and, you know, pain tolerance.

    6. LRLenny Rachitsky1:01:24

       This reminds me of something else the CTO of Stripe shared of how they build new product, which is they pick a couple of customers that need a problem solved and they just build it w- for them essentially, and with them. And in B2B, generally, it's a lot easier to build something people really want because they are very motivated for you to solve their problem and they're gonna put in the time. And there's like, you don't need thousands of people involved, you just need a couple.

    7. NWNoah Weiss1:01:46

       Yeah. I, I definitely think it's one of those things where if you could do it in a way and, and they say like, "I can't live without it," like the classic like, not like, you know, do you like it? Sure. But can you, can you work without this thing? If the answer is definitely not, you've built something that probably a lot of other companies will want too.

18. 1:02:03 – 1:10:15

    Noah’s famous blog post: “The 10 Traits of Great Product Managers”

    1. NWNoah Weiss1:02:03

    2. LRLenny Rachitsky1:02:04

       All right. I'm gonna shift to a totally different topic which could also be its own whole podcast, but let's just see how it goes. So you wrote this, I'd say famous blog post on product management called The 10 Traits of Great Product Managers. And I wanna just try to lo- go through this list briefly and just see how it goes. I don't wanna s- you know, this could be an hour (laughs) of conversation but let's just kind of run through it 'cause I think it'd be useful for people to hear and I think these are all 100% true even though you wrote this a number of years ago at this point. And just, let's just see what comes up and, and then I have a few follow-up questions on this list.

    3. NWNoah Weiss1:02:34

       These traits are kind of, they came after I wrote this other thing which is like the, the five myths about product management, which are all the things that people think product management is and why they switched the job and they're disappointed by. And then I was like, let me actually write the positive version of this which is the, the things that the job actually is about. It's not a career ladder, it's not like the, you know, here's the structured interview things that you should interview for. But I think it's the actual job of product management. What is it about? What does success look like? And I don't think they're really in a particular order in hindsight, but-

    4. LRLenny Rachitsky1:03:03

       Mm-hmm.

    5. NWNoah Weiss1:03:04

       ... I'll, I'll read them in order. So, uh, live in the future and work backwards I think is very much kind of the idea that as a PM if there's one thing you're responsible for, it's kind of having a longer term vision and time horizon, so how do you carve out time to not just be where we are in the next two weeks, but six months, a year, or two years from now, how do you immerse yourself in that and then bring ideas back, bring inspiration back to the team?

    6. LRLenny Rachitsky1:03:27

       I'm gonna just gonna throw comments out as you're going through them, just add to them. (laughs) So I love that this is like exactly Amazon's like approach of like work backwards, yeah working backwards process. At Airbnb this is actually like th- the main thing Brian Want pushed everyone to do is just think about the idealized product of like a put a magical world where all this is totally solved and then work backwards from that. And then Paul Graham talks about this too, right, just like live in the future and build it.

    7. NWNoah Weiss1:03:53

       I, I definitely riffed off at least the Paul Graham thing because I remember reading that essay of he thinks, you know, everyone thinks that you can get sort of ideas by like I don't know sitting with your co-founder laying in Dolores Park looking up at the sky and like conjuring up the next unicorn or something. Uh, definitely not how that works. You have to actually like immerse yourself in the problem space and try to imagine what the future world looks like and then what's missing for people to get to that future state. So yeah, I, I agree.

    8. LRLenny Rachitsky1:04:20

       I also saw a great tweet by Shreyas the other day about how if you're working at a company with good leaders, they're never gonna be sad that your vision is too big and too ambitious if there's, you know, some reality to it. That often they want that, just like let's go, let's think bigger, let's how do we, how do we change the way we think about the future of all this stuff?

    9. NWNoah Weiss1:04:39

       Yeah. I mean that was, when I was at Google the thing I took away most from any review with Larry and Sergey was they would ask like, "How could we get like 100X the scale?" Or, "How could this work for this, what seemed like an outlandish use case but would like push the team to think much further into the future?" So yeah, I think that's what the founders always want.

    10. LRLenny Rachitsky1:04:56

        That's what Brian Chesky always said too, just like how do we 10X this? What would it take to 10X this idea?

    11. NWNoah Weiss1:05:01

        Yeah.

    12. LRLenny Rachitsky1:05:01

        So, awesome. Okay.

    13. NWNoah Weiss1:05:02

        Okay. Uh, the second one which is maybe obvious but thinking about how do you actually amplify your team? So how do you facilitate ideas? How do you create energy? How do you create momentum? A PM role I think can be a little bit unsatisfying if you're used to it where you create things yourself opposed to you are the one who's amplifying what the work that's being created by everyone else is. So you have to kind of get into that, uh, more of a facilitator mindset.

    14. LRLenny Rachitsky1:05:26

        What I think about here is a lot of teams like don't want PMs on their team or don't like PMs or don't think PMs are valuable. What I find is that just means your PM is not good because if you have a good PM they're just gonna help you do the best work of your life. They're gonna help you clarify things, prioritize well, unblock you, all that stuff.

    15. NWNoah Weiss1:05:45

        Totally. Who, and we should find out who wrote that, um, expression early on of like PMs should be mini CEOs. I think that's the most dangerous piece of advice ever in the history of product management because I think that is how you end up having PMs who try to act like dictators instead of kind of leaders and facilitators 'cause if you're acting like that, yeah, your team can completely reject you and say I never want another PM again. (laughs)

    16. LRLenny Rachitsky1:06:09

        Yeah. Like so many new PMs are just like I'm finally gonna have the power. Finally like if they move from engineering or some other role and then they get there and are like, oh, what the hell?

    17. NWNoah Weiss1:06:17

        Yeah.

    18. LRLenny Rachitsky1:06:17

        I have to convince everyone of all these things I want to do.

    19. NWNoah Weiss1:06:20

        That actually I'm gonna skip it and tell you a different direction of, of the order of this post, but the, the fifth one that I wrote on there was your job is to facilitate the pace and quality of decision-making. And that is very different than you are the person who makes all the decisions. And in fact I think one of the things that PMs struggle with early on is how do you actually get the team to be able to make high quality decisions quickly without you kind of arbitrarily playing tiebreaker all the time? And it's a, it's a soft art to be able to do that but I think that is actually how you have a really healthy team dynamic instead of PMs who want to say, "Okay, now it's my turn to get to make the decisions." It's definitely not what the job is about.

    20. LRLenny Rachitsky1:07:01

        What that makes me think about is I taught a course on product management at one point that I've paused for now of just like, like the core job of a PM is to figure out what's next for every single person on the team and you're this, there's this meme or GIF of a, a dog on a train and he's just laying the tracks as the team is moving forward ahead of them just one step at a time.And to do that, this is such an important part of that, is just help people make decisions and unblock them.

    21. NWNoah Weiss1:07:24

        Totally. And I, uh, I'll kind of combine two of these together. So one is, you do have to have impeccable execution. This is kind of more of a baseline thing. But I've never seen a PM who was, like, disorganized or didn't do follow-up or wasn't clear about expectations or timelines. It, it's not high in Maslow's hierarchy of kind of PM enjoyment, but I, I do think it's, like, a baseline expectation. The thing I think is more enjoyable and probably the most important thing in the long term is focusing on impact primarily to the customer experience, but also to the business. And I think, you know, there's that saying, like, worth solves all problems. I think impact solves all, like, PM issues, which is if the team is consistently building things people love and changing the direction of the business, everything else is just an input. And so I, I think that focus and understanding as, as you're laying the tracks of, like, what direction do you need to go as a team to actually drive that impact? That's probably the single thing that a PM can most control.

    22. LRLenny Rachitsky1:08:28

        I love that. I always recommend exactly that. If, like, if your career is not going as well as you'd hoped or you're not getting promoted, it's usually you're not delivering impact, whatever that means to the company. Like, it may be moving, moving a metric maybe, and building great product that the founders really love.

    23. NWNoah Weiss1:08:43

        Yeah.

    24. LRLenny Rachitsky1:08:43

        May in fact... Can mean a lot of different things. But it's so true. On the execution, executing impeccably bucket, the way I think about that is as a great PM, you need to kind of have this aura of, I've got this. Anytime someone puts something on your plate, it's not gonna fall off. You're not gonna forget about it, you're not gonna cl- let a ball drop. That if you... The more you can create this aura of, like, I got this, the more responsibility people are gonna give you, the more impact you'll end up having, the more people want to work with you and all that.

    25. NWNoah Weiss1:09:12

        Yeah. Ben Horowitz was a board member back at Foursquare, and I, I distinct remember he used to have this saying, very Yoda-like, of, you know, good leaders need to say what they're gonna do and then do what they said. And if they can't, then they need to follow up and explain why. I mean, that's, like, the amendment. And I think that is kind of what good execution looks like.

    26. LRLenny Rachitsky1:09:33

        That last point is so important. Like, you may not be able to do all the things on your plate, but just telling people, "Hey, I'm not gonna get to this thing. Let's reprioritize," is s- such a small thing you could do and really creates that aura of you got this. They're not gonna forget about this thing asked you to do.

    27. NWNoah Weiss1:09:47

        Yeah. You're kind of the, you're the shock absorber for the team. You're the thing that builds people's confidence that things are gonna be running smoothly and you'll get over the inevitable, you know, speed bumps and whatever else. So I'll combine two or three of these that are kind of related or just more skills.

Episode duration: 1:25:33