How to foster innovation and big thinking | Eeke de Milliano (Retool, Stripe)

Process by definition is variance reducing. Like you're introducing it because you worry that the variance in your org is too high. Like you want people to sort of meet a certain standard. And the cost of that is obviously while you're reducing the standard and bringing folks up to the average, you're also bringing other folks down to the average. And oftentimes the folks you're bringing down are your hire performers, your most creative thinkers, the folks who like, you know, I think actually don't really need process to do their best work. And so that I think is always the tension that you have with process. And obviously, like one of the reasons why companies introduce process much more and more as companies get bigger is because it's just, it's harder to sort of like get all these folks who kind of don't need process. You actually want to reduce the variance.

Welcome to Lenny's podcast, where I interview world-class product leaders and growth experts to learn from their hard-won experiences building and scaling today's most successful companies. Today, my guest is Erika Demiliano. Erika is head of product at Retool. Prior to that, she was a longtime PM at Stripe. She was actually one of their very first PMs, where she helped build some of their foundational products like Stripe Checkout, Stripe Connect, Stripe Radar, and Stripe Chargeback Protection. I had a total blast chatting with Erika. We covered Stripe's internal culture and what makes it so unique and innovative, how to foster and protect innovation at your own company, what is the right amount of process by stage of company, how to build Talent Portfolio and so much more. I am really excited for you to hear this episode. And with that, I bring you Eka de Meliano after a short word from our wonderful sponsors. Today's episode is brought to you by Miro, an online visual whiteboard that's designed specifically for teams like yours and mine. I have a quick request. Head on over to my board@miro.com/lenny and let me know which guests you'd love for me to have on in 2023. And while you're on the mirror board, feel free to play around with the tool. It's a great shared space to work closely with your colleagues to capture ideas get feedback, and iterate quickly and easily on anything you're working on. For example, in Miro you can build out your product strategy by brainstorming with sticky notes, comments, live reactions, voting tool, even an estimation app to scope out your team's sprints. Your whole distributed team can come together around a wireframe and draw ideas with a pen tool, or even put mocks right into the Miro board. And with one of Miro's ready-made templates, you can go from discovery and research to product roadmap to customer journey flows, Final mocks, you get the picture. Head on over to miro.com/lenny to leave your suggestions. That's M-I-R-O.com/lenny. This episode is brought to you by Notion. If you haven't heard of Notion, where have you been? I use Notion to coordinate this very podcast, including my content calendar, my sponsors, and prepping guests for launch of each episode. Notion is an all-in-one team collaboration tool that combines note-taking, document sharing, wikis, project management, and much more into one space that's simple, powerful, and beautifully designed. And not only does it allow you to be more efficient in your work life, but you can easily transition to using it in your personal life, which is another feature that truly sets Notion apart. The other day, I started a home project and immediately opened up Notion to help me organize it all. Learn more and get started for free at notion.com/lennyspod. Take the first step towards an organized, happy team today. again at notion.com/lennyspod. Lenny, welcome to the podcast. Thanks so much for having me, Lenny.

It's absolutely my pleasure.

A bunch of people have told me that I need to have you on this podcast. And actually a big thank you to Sneer Kodesh, who I believe is a colleague of yours, who gave me a bunch of interesting questions to ask you.

So I hope you're ready. Awesome. Very ready. Sneer is the best. So yeah. Excited.

So you're currently head of product at Retool. But before that, you spent a lot of time at Stripe. You spent six years at Stripe. And so before we get to Retool, I actually want to spend a little time on Stripe. To set a little foundation, could you just talk about some of the things you worked on at Stripe and maybe some of the things you're most proud of during that time? Yeah.

So when I joined Stripe, I joined Stripe in 2013. It's pretty small at the time, around 50 people. And Stripe didn't have any product managers at that time. and I think Stripe was kind of famous for that. And it really wasn't because, you know, we were anti-PM in any way. It was mostly just because we were building this product for developers. We had a lot of really talented engineers who were essentially doing the PM job. So I actually joined Stripe as the first account executive, which I think is pretty funny now that I've seen, you know, real account executives do their jobs, because I, I really just had no business doing that job, but. It also really wasn't your typical sales job in that most of Stripe's volume was inbound. So I was really just spending all of my day talking to customers, helping them figure out how Stripe might work for their particular business flows and payments models. And a lot of it was, I think, you know, what PMs do, it's just like talking to customers, understanding their pain points and really helping them figure out how the product might do that. So to this day, when people ask me, hey, like, what's your advice? for going into product management, I'm like, well, you should go get a sales job. I think it's pretty valuable. But it's kind of a long story to explain that at the time there was this one sort of business model, the two-sided marketplace business model that was really, it was just becoming huge. And yeah, you can even really sort of imagine that now, but back then, like Airbnb, Lyft, Uber, all those marketplaces, it was quite a new way of doing business. So I was talking to all these customers and they had pretty complicated payments needs. So, you know, it's like one payer putting in funds and it's getting paid out to a recipient. So an Airbnb guest and an Airbnb host or multiple payers putting funds in and it's getting paid out to one recipient, like a GoFundMe campaign or one payer putting in money and it going out to multiple recipients, like, you know, with ClassPass where you pay the subscription and then it gets paid out to a bunch of studios. And then on top of that, all these platforms and marketplaces started having all these pretty complex regulatory and compliance requirements that were pretty different country to country. So it's just this really complicated financial infrastructure problem and Stripe actually didn't really have the right solution for it and no one else in the market did either. So at this time, this Stripe engineer, Brian Krause, started poking around in this problem. because I was talking to so many of these customers, we started talking about this problem together and poking around. And that actually resulted in us launching this sort of evolution of this product that we had at the time called Stripe Connect. That was easily, I think, you know, one of my favorite products I worked on. Not only it ended up actually being really quite significant for Stripe's business, but also because, you know, it was the first product. And I think that's always pretty special. And then the second product that I think of very fondly during my time at Stripe is this product Stripe Radar. Stripe obviously processes a lot of payments wherever there is money, fraudsters will follow, and there were certainly some merchants who were just particularly vulnerable to payments fraud. So someone's using stolen credit card information to essentially purchase a good. And if you're not in payments, this is gonna maybe sound kind of shocking, but if a merchant processes a payment from a customer, and that customer used a stolen credit card, the merchant is ultimately on the hook for those funds. So it can be detrimental. Like the merchant who's trying to run a real business is just losing all this money because there are all these fraudsters who are buying stuff from them online. So I really liked working on that product. So the product at Stripe was essentially we built a bunch of machine learning models to try and predict when a payment was fraudulent, and then we built a product on top of that to help customers understand why we were blocking payments and help them sort of write their own rules around that too. That was really fun both from an anthropological perspective because we were kind of hanging out in corners of the internet where you wouldn't usually go if you were doing only legal things. So we were trying to understand who these fraudsters were. But it was also really cool from a product perspective because of all the, I think, kind of complicated product questions around how humans should interact with AI and models. And I imagine a lot of the folks in the sort of latest movement in AI are thinking a lot about that too. It's like, how do you explain a model to humans? How do you let them interact with it?

So both those things were really, really neat. I actually use Stripe for my newsletter. It's how folks subscribe. And when I log into my Stripe dashboard, and there's so many products, I don't even know what many of them do, but I've often seen Radar being pitched to me as something I should pay for.

Nice.

And I feel like I should probably turn it on. That sounds really useful. One thing that I want to dig in on, so you said when you joined there were no PMs at Stripe. And how many PMs were there when you moved into product at Stripe?

That's a great question. I want to say maybe like three or four.

Wow. Incredible. And I know you said Stripe is kind of known for being really late to PM and kind of like, I don't know if I want to say anti-PM, but there was like a lot of sense. Why do we need PMs? We have amazing engineers who can decide what to build. Is there anything you can share around just that general philosophy of Stripe and was it effective? Was it good? Because a lot of companies are anti-PM and I think they always point to Stripe and Snap is another example of like, we don't need PMs. Look how far these companies got without PMs. Is there something unique to Stripe that allowed them not to have PMs? I think it was like 200 employees probably by the time they had their first PM.

I think we actually had our first PM at, I want to say, about 100 employees.

Okay.

But yeah, no, it was late in the game. And actually, maybe just sort of to pattern match, Retool also didn't really have PMs for a while. And I think actually both in the Stripe case and the Retool case, the thing that both of these companies have in common is you're building for developers. The people who are building the product are the customers in a lot of ways. So like, they get it. You know, there's, there's really no, no reason why you should have sort of some, in some ways, like a middleman trying to figure out like, Hey, who's the customer and what is it that they need and what are their pain points? And I think the moment at which it became clear, like we really do need PMs at Stripe, and I think we felt the same way at Retool is your customer base starts expanding, you start having different kinds of customers, you need to understand sort of the whole market, all of the ICPs that you're going after, and the organization gets more complex. I think Stripe, gosh, in particular, it's just an extremely matrixed organization and business because every time you launch a product, you're having to think about well, how do we do this in other countries? It's different financial infrastructure. How do we think about the legal side, the compliance side, et cetera? And I think PMs could really kind of bring that whole story together and make the whole machine move. And, you know, in addition to sort of understanding the customers and the value prop and like what the overall strategy was.

That's a really good reminder of engineers and designers can do the work of a PM, but often they don't want to. It's like there's a lot of not fun. parts to it. They're like, oh, I wish we could have someone here do these things. And PMs enjoy that work. They're good at it. And so eventually engineers and designers realize, okay, I see. I see why maybe we should hire PM.

I say this often actually to folks who are thinking about going into product management. That's awesome. Just be really, really sure about what you're signing up for. It's kind of the same with a lot of people want to become a manager. But just so you know, being a manager is like, hey, you're doing performance reviews a lot and you're in one-on-ones a lot. You have to really love that kind of work. And I think in the product management case, it's like you're constantly working across a lot of different teams. You're trying to influence a lot of people who don't report into you directly to do a bunch of stuff. And if you love that, that's great. But you've got to be sure you're signing up for it.

Yeah, that's a really good comparison. When I first became a manager, I was an engineer actually, and I was managing engineers. And then when I moved on to something as an IC engineer again, I was like, I will never manage again. no fun at all. Why would I want to do this? And I imagine people get into product thinking they're gonna have all this control, power influence, and then they're like, oh my God, this job is so freaking crazy. What did I sign up for? It's so hard.

Yeah.

That reminds me something that I heard you did at Stripe is you wrote their internal culture guide. It was like a quick start guide to culture at Stripe that I think was maybe shared with early employees. And if that's true, I'm curious. what it is about Stripe's culture. If you could just, I don't know, briefly share just like what makes Stripe so special? Clearly it's one of the most successful companies in the world in history. I know there's been a slowdown with the market. Everyone's slowed down a bit, but it feels like Stripe has been incredibly successful, continues to innovate like crazy, hires incredible people, people that are starting amazing companies. So I guess back to my question.

What is it they think that makes Stripe's culture unique and special? Yeah, that quick guide to Stripe's culture. I think we wrote it maybe when we were around 150 people or so, and we actually wrote it to share with candidates because we, yeah, the idea was like, look, we have like, we're kind of opinionated about how we do things here. Most companies struggle to describe what their culture is. Like how does a fish describe water? But it's worthwhile getting a sense of sort of the things that we feel opinionated about and that you might like or you might not like. And actually, when we put out the guide, our metric of success was that more of the right people would apply to Stripe and fewer of the wrong people. And there was a whole section in there that was like, Hey, we work pretty hard here. And that is certainly not for everyone. And hopefully you're excited to come and work really hard with colleagues who really care, and that's kind of the reward. So that's maybe one example. On Stripe's culture and what's made it so successful, I've really been thinking a lot about were there one or two pivotal points or decisions for Stripe that really set it on its path to success? And I don't think there are actually. Every time, well, if this hadn't happened, I can always reason my way into, I think these other things would have happened and Stripe would have kind of ended up in the same place. So I think what Stripe was actually particularly good at, and by the way, you should take all this with a humongous grain of salt because I haven't worked there since 2019, but at least my experience when I was there, what Stripe was really good at was just making not just one or two big decisions, good decisions, was making a lot of really good decisions all the time, big or small. And that I think was quite cultural. There was this, Just humongous respect and enthusiasm for thinking. Like, that was just like a, it was such a part of the culture that was, you know, one of the values was think rigorously. Like every meeting was very much like, Hey, how do we, how do we think about this thing from first principles? Like no one would ever say, for example, it's a best practice to do X. Like people be like, well, why? Like they, you know, you have to kind of go a few levels deeper. So that was, that was, I think one piece. the other piece, and a lot has been written about this already, is Stripe had a very strong writing culture. All communication was along from writing business reviews, strategy memos, product reviews. It was just, there was a lot of writing that happened. And I, you know, I, I very strongly believe this. I don't think you can be a good writer unless you're a clear thinker. And if you couldn't write well, I think it was actually pretty hard to be successful at stripe, at least in the early days. So I think Stripe just cultivated a lot of really Good writers and, you know, by sort of input into that, a lot of really good thinkers. And then I think the last bit Stripe is really good at was not overthinking decisions. Cause that's kind of the flip side of this sort of like really rigorous approach is that you like lock yourself into a room, like can't come out for five days and you have to like be good at making a lot of decisions and right decisions quickly. And one of the things that, that we would always sort of kick around a lot was this like idea of is it a Trapper decision? which I think is an Amazon concept. It's like the, if you make this decision, is it one door or is it two doors? And can you come back from this decision? And I thought Stripe was actually really good at being rigorous, back to the rigorous point, about what actually was a trapdoor decision. So I think a lot of people, for example, think pricing is a trapdoor decision. But actually, it's really easy to grandfather your existing users into some existing pricing model and change pricing for future users. if you believe that the product's going to be successful, your early users are only going to be a fraction of that pricing model. And if the product's not successful, then who cares if you change the pricing model? I think that's a really good example of people think that's a trapdoor decision, but actually you can move much more quickly on that decision than you think. And then decision that was definitely felt to us very much like a trapdoor decision. I think Stripe took a long time being really rigorous about was titles. I think Stripe's kind of famous for not really having titles and I think it was right to be rigorous around that decision because you can't, like, once you've given someone a title, you can't really take it away. So, you know, you better be damn sure about, you know, the title you want to give someone.

Yeah, the titles are hilarious. I put together a career ladder doc of all company leveling names across all the big companies.

And it's ever like a lead.

Product manager. Product manager, product manager, product manager.

Yeah.

And there's like, A lead here and there. Like, yeah, like VPs are basically product manager or product leader, something like that. Hilarious. You talked about this idea of first principles. People talk about that often as we want to be first principle thinkers. We're going to think from first principles. How did Stripe actually operationalize that, implement that? Was that just like founder top down, continue questioning people and that trickles down? Or is there anything else they did to kind of instill that mentality?

I definitely think it's founder top down. They were really good about hiring people who thought that way too. So it's just all that stuff. It just really, really trickles. And actually, to this day, when I'm preparing like a memo or like a board deck or something, I imagine in my head, I'm like, what if I had to present this to Patrick? And like, my ideas just get so much better because it just like sort of, I know, like you can kind of think about sort of the questions they would ask. so I think that was one, you know, I think the other one was just to serve this writing culture piece. Like, people would just like, once you start writing things down, you realize, like, hey, that actually doesn't make a lot of sense. And then I think the, the third thing actually was there was just always a lot of questioning about, like, the status quo. So if something had been done in an industry for a long time, people would always be like, well, why was it done that way? And, I mean, I think this is kind of actually how stripe got its start. Right? Like, I think when stripe started, if you want to set up a merchant account, it could take weeks. And everyone just assumed that's what it took. And of course, John and Patrick were like, well, it doesn't actually make that much sense why it should take that long.

You shared some of the values. I don't know if they're just called values, core values at Stripe, but are there other ones you can share? You said one is think rigorous or don't ever think.

Yeah, what are the other values? Yeah, I think rigorously was one. And I think they might have actually sort of updated these on the website recently. So my recollection is very likely out of date. and Stripe would call them operating principles actually, which I think is actually better because values is almost like, you can't really tell people what to value. Like everyone has their own value set, but you can tell people like, Hey, here's, here's how we operate. Another one that I loved was move with urgency and focus. It was just really this idea that you're kind of your, your biggest compatriot and your biggest, I think, enemy as a startup is time. So like you just, you have to move like really, really really fast on it. Customers first was the other one, or users first, as Stripe would refer to customers as users. So those are some of my favorites.

When I think back to what made Airbnb special in a similar way, the values, we call them core values, I think were actually really, really important and turned the company into what it ended up being. And it was actually there around the time they created these values. I'm curious at Stripe, do you have any recollection of just how they came to these Operating principles, because I imagine founders are listening, really, how do I.

How do I come up with these for our company? I'm pretty sure Patrick wrote them. Yeah, I think they came from Patrick and John wrote them. The other actually, the other value that I love, I don't think they have it anymore, but it's so good to the operating principles, micro pessimists, macro optimists. Wow. Yeah, it was just like this idea that like, look long in the long term, we expect the curve to go up. Like we believe, like, you know, we very much believe in the upper trajectory of just about everything, but like, on the sort of day-to-day decisions on the minutiae. We're quite sort of critical. How do we make sure this thing works? And we think about all the ways in which it doesn't work.

So zooming out a little bit, we've been talking about Stripe. Retool also is a very first principles, innovative company. Something that I think of when I think of Retool is during the wild market days of long ago last year, I know Retool is very conservative in how they raised money and the valuations they raised at, which looking back ended up being a really good idea while everyone's raising at like $100 billion. I think their last round is like a few billion and it's a very reasonable, conservative way to think about fundraising. And so I guess just thinking about innovation in general, why do you think some teams are able to run like innovation machines continue to put out new products, disrupt industries while other companies kind of plot along, stay conservative. Is there anything you've seen, something you bring to teams you work on to help foster innovation and big thinking?

Yeah, I actually think about the extremes of that question. So not, you know, why are some teams innovative, but like, why isn't every team innovative? Like, I feel like no one wakes up in the morning and thinks like, yeah, today I want to work on like boring incremental stuff. But most teams do end up working on like pretty incremental I always wonder, like, what is it that's stopping folks from being creative and, and thinking bigger? And I think it comes down to like a couple of things that, you know, companies do sort of unwillingly or not even sort of realizing it. And one is this fear of failure. Like, I think leaders want the upside of innovation, but they're not really willing to deal with the cost of innovation, which is like, look, if you're going to swing big, you will invariably stumble sometimes. So, you know, I think if you want to mitigate that, you have to start shining light on failure. That's, that's really the only way. Like, you have to start normalizing it a little bit. And obviously, like, if an individual or if a team is, like, consistently failing and not learning that you need to, you know, you need to sort of deal with. But I think sometimes it's, it's good to fail. So, you know, like, and, and when, when you do feel it, like, use it as an opportunity, like, don't squander that moment. to not learn. But I think one example is instead of calling something a postmortem, call it a retrospective. So that it's kind of a positive thing, like, hey, we're learning from this thing. And then I think the other way to kind of mitigate the fear of failure is you have to figure out how to give folks more at-bats. Because obviously anything that takes one year to ship and you haven't gotten any customer feedback, the stakes of that are just going to feel so, so high. any, like, if you get it wrong, it's, it's gonna be devastating. So you have to figure out how to lower the stakes. And I honestly, I think that's, in some ways, it's kind of easy. It's like, look, don't put too many resources against these kind of like bigger swings. Like, have them be small teams. And then also, like, just get customer feedback as quickly as possible. Like, don't wait until the thing is perfect. And that way you can kind of, yeah, like, limit the, limit the risk. So, yeah, I think that's fear of failure. That's, that's definitely one thing that stops teams from being innovative. There's kind of a, I think a very practical one. Sometimes teams are just getting bogged down by really urgent work. There's just too much tech debt, there's too much product debt, bugs, instability. It's like massive hierarchy of needs. There's just no way that they're going to be able to focus on the enlightened, bigger creative stuff if they're just heads down dealing with incidents all day. So if that's the case, yeah. diagnose it and get your team out of that. And then I think the last reason why teams aren't always that innovative is because I think thinking big is really hard. And to some people it comes pretty naturally, but for most of us sort of, you know, mortal souls, it's just really, really hard. And it takes time. And when you're at a startup and you're just grinding day in, day out, you're just like, you're treading water, you're just like trying to make it through the day. taking the time to really think about the strategy and where things should go and get creative. It's pretty hard. So I call this, you have to give teams permission to think. So create these moments in your company culture, in your overall business processes where you're asking people, you're literally saying like, Hey, this is part of your job to think bigger. So at a retreat, we have these these team charters and we do team planning. And at the bottom of every team charter, we have a section called Think Bigger with 20% more time. What would you do that isn't on this list already? And then another really neat tradition that we had at Stripe, we have a Retool now too, is this thing called Crazy Ideas. So at the beginning of every year, David will send out a blank doc to the org. and it's titled Crazy Ideas. And the prompt is, Crazy ideas are ideas that we shouldn't obviously do. There's a 90% chance that they make no sense. But in the 10% chance that they do, they will make, you know, a 10 to 100x difference for the retail business. And then it's literally just a call, a request for crazy ideas. And the org loves it. It's amazing. The energy around it is so cool. And it's not just product ideas. Different ways of, of how we should run our organization or like, it's really cool marketing ideas in there. So it's just that doc is awesome. And whenever I have like a down day, I just like scroll through that doc.

I know that you launched three different products this year, which you want to talk about, which maybe came out of this big thinking, but a couple more questions just to dig into some of the stuff you just shared. One is this big doc of awesome ideas. What's come out of that? Cause, you know, I think of like hackathons and people have all this energy and it's exciting and then they get all these cool things and then they don't go anywhere. do things come out of this? There's a lead in actual ideas that people. Yeah.

Yeah, yeah, yeah. So every year that we send out the doc, we look at the past year, the doc from the past year, and we're like, okay, did we do anything on this list? And like consistently we do anywhere between.

Like three to eight things on the list. Wow.

Yeah.

Is there an example of product that.

Launched that came from that list that comes to mind? Actually probably at least one, if not two of the sort of three products that we launched this last year were on the crazy ideas list at one point. Like a year ago, if you'd ask me, what is Rachel's product? I would have told you, Retool helps you build internal front ends really, really fast. But if you were trying to schedule a query to run at a later point in time or have something trigger or run an ETL task, you couldn't do that in Retool. So that was actually one of the crazy ideas and that ended up becoming Retool workflows, which we launched just last year in November.

Okay. We're definitely going to chat a bit about that. I want to go back to the two other Embrace failure, make it okay to fail. And then two is don't let people get sucked into urgent stuff and have time to focus on important stuff. Is there an example of just like how to help people embrace failure? You talked about retrospectives. Is there anything else just like that you found works either as a tactic, as a process, as a framework, just to like help people get out of that? Cause you know, you hear that a lot, like embrace failure and then People are like, oh, but I failed and you suck. So yeah, is there anything else along.

Those lines you found effective to create that feeling? To me, it's always in the follow-ups. So like, you know, have people talk about the failure in like these sort of public forums where usually you talk about the successes. So like have someone actually like write a note that's like, Hey, here are all my learnings from this thing that we did and sent it to the org. Every tool we have a shipped at email list. Like if you ship something, you know, you'll listen to that. Have them send that email to the org. And it's just kind of an awesome way to celebrate or have them present at all hands and share what they learned. And it almost always sort of results in, I think, a really positive twist.

At Airbnb, I think for a period there was an award for the biggest failure, like project in the feature.

Really? That's so cool.

Yeah, like a trophy or something. Yeah, it was like a short-lived idea. Yeah, yeah, yeah. And then on the second bullet point of urgency creating Giving people a chance to think longer term, is there anything there that you found to be actually effective to create that culture and give PMs and product teams a chance to think longer term and not just be stuck in the fires that they're dealing with?

There's a couple of top-down things and I think a couple of bottoms-up things. On the top-down, I think sometimes you just have to be willing to fund someone with something. They come to you with an idea and you're like, yeah, take an engineer and go do it. And you just have to give folks that sort of organizational buy-in. Actually, I think I was thinking back to that engineer, Brian Krause at Stripe, who started poking around on this marketplace model. And I was thinking, I can't remember a moment when someone formally said to him, this is now your job. I think he just kind of went and looked and went and dug into them. And at some point, I think a manager was like, all right, this is now your job. But yeah, I think you have to be able to sort of fund folks this time and give them that. Hackathons, I think, are pretty good for that stuff. It is a moment for folks to take a step back. And then I think more than anything in people's planning processes, I really like asking folks the 20% more time question or the other question is like, look, if you doubled the team today, what would you do? That shows up in our planning processes as well. I think it really helps people break out of this I think you kind of end up planning towards the sort of the team that you have, not the team that you should have. So, like how folks can break out of that process. Awesome.

I really like the think bigger bucket in your planning docs. Just like, what would you do if you had more resources? Just maybe a quick question there. Is there like a bunch of detail that you ask them to put into.

That, or is it just like a bullet point of big ideas? Just, yeah, folks can just go crazy on it. It's however they wanna take it. Yeah, I've seen folks actually sort of like create entire demos. But I actually think the trick is like less structure in those cases because you don't really want to pigeonhole or make it even that intimidating for folks. Like if someone just wants to write down a few bullets, that's great.

This episode is brought to you by Eppo. Eppo is a next generation A/B 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. does all that and more, delivering results quickly, avoiding annoying prolonged analytics 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 getepo.com, getepo.com, and 10x your experiment velocity. Okay, so you launched three new products this year. Usually companies are lucky to launch one product a year. A few questions. One, just like tell us what those three were in case people are interested and want to check them out. Two is, was that a good idea? Is it good to launch three new products in a year? And then three, what did you do organizationally to allow for this to happen? Because that feels really ambitious and rare. I'm curious what you learned from that.

So the three products we launched were one was retool workflows that, you know, I talked to you a little bit about. It's like, if you want to be able to sort of essentially create a workflow, like schedule an alert, run a task, you can kind of do it in the retool way where it's this visual sort of easier builder of creating these, you know, different sort of workflows, but you can write your own code on top of it as well. Second product is retool mobile. So you could pretty easily build, you know, CrowdWeb Apps with Retool, but there are plenty of folks who don't sit sort of behind their laptop at a desk all day. And, you know, for those folks who had a lot of companies who were like, I want to build like a mobile native app. And so we launched that product. And then the third product was a Retool database. So until very recently, like if you came into Retool, you could connect your data, you know, via your, your APIs, your database directly into Retool. But what we found is that actually a lot of folks either don't have access to their database or they're actually just like they're trying to build an internal tool and they don't necessarily want to store that data in their production database. So we built essentially like we spin up a Postgres database for you and.

You can just access it via a really nice UI and manage it directly in Retool.

Amazing. Okay, back to the other two questions. Yes. Okay. So I think your second question was, was this a good idea? Right. Yeah. You know what? I think in hindsight, if I were to do it again, I think maybe sequencing it would have maybe been slightly better. Just because I think what we ended up doing is we kind of ended up launching or launching the idea behind all three of these products at around the same time, and they all ended up maturing at around the same time. And that was just a lot of headspace from the org, even though actually the teams themselves were not that big. Like, it was just like, you know, we had all these products and we were just waiting for them to launch and waiting for them to launch, whereas, like, maybe if we'd sequenced it, I think that would have felt probably also just more satisfying to the whole organization. But, I mean, it ended up working out. That's. I think that's kind of the crazy thing. It's like, you know, we. We kind of were able to pull off launching these, you know, these three products. And by the way, I should caveat that, you know, the. sort of the, the further I get along my career, the more I realize I'm just kind of building on the shoulders of giants. And a lot of these ideas, et cetera, were very much in the works and, and were happening before I came along. But yeah, it was really neat to see how we, we got that all over the finish line in a year.

And then the last question there is just, what do you think you did to allow for this? Because it's pretty rare you can build so much. And I know the team's not huge, right? Like how many people work at Retool, roughly?

Today we're around 300 people.

How do you structure an org to build three and launch? I didn't realize they launched around the same time. It's like I'm picturing Elon Musk launching three rockets at once. How do you allow for that to happen?

Yeah, I'm sure our product marketing team felt that way. Yeah, a couple of things. We started really small with all of these initiatives. So I think I mentioned we really had one or two people working on on each of these products for the first six months. So it was one engineer and one designer, one engineer and one PM. And they really didn't get funding until it was clear that there was something there. So they just, those teams just, they spent a ton of time with customers, spent a ton of time building, a ton of time prototyping. And it was kind of the moment where I was like, okay, there is a bear there that we started sort of bringing more people onto the team. So it was the first piece. Second piece is that we really treated them like startups where like Ritual's the VC and like Ritual funds with like resources and Ritual's existing customer base, which is obviously quite valuable because you kind of have all these customers you can market to and promote to. But then the teams really had to prove out ROI, either in engagement or eventually revenue in order to be able to move forward. And then the third thing, and this one actually, I think it can be quite controversial, is we were really deliberate about keeping these teams separate from the rest of the org, especially early on. Now a lot of them are, they're very much kind of a part of the overall organizational processes. But very early on, they were kind of, they're running on their own. They were, you know, meeting quite independently with like one or two or three folks from the leadership team. and they were also just quite separate from the product itself. Like, I think Retool Mobile is actually a really good example where we had a lot of debate about whether or not we should build Retool Mobile out of the core web app builder because a lot of the primitives are the same. And we eventually decided that we were gonna run it as a separate team because we wanted the team to be able to move really, really quickly. And we didn't want it to get bogged down in what are just the realities of running a product that has product market fit like. You know, bugs, yada, yada, et cetera. And I think that was absolutely the right call because Retail Mobile actually has quite a different target customer, which we only really realized maybe like halfway through the project. And I don't think we would have been able to sort of really understand that or even like sort of like get broader in that kind of thinking had we sort of been stuck in the core Retail product. But yeah, there's a cost to that too, which is like, okay, now we have these two products and we have to, I think obviously the strength will be, how do all these products interact with each other and how they build on top of each other? So we have to go and invest in that now. But I think it's totally worth it because your team can just move more quickly, be more independent and think more independently too.

In the time that you worked at Stripe and Retool, I know we chatted before we started recording. You think a lot about what is the right level of process for companies at different stages. And I'd love to just hear your thoughts because I know I know that's a challenge a lot of companies face. How much do we put in now? Do we get inspiration from big companies? Do we try to stay lean? Something here maybe actually run into, I'll just mention briefly that there was a huge resistance to process for a long time. And so the product team was just kind of a little crazy for a bit. And then they're like, okay, we need product development process. We need deadlines and specs and, and that helped a lot. So, yeah. So let me ask again, just how do you think about the right level of process for a stage of company and what have you learned there?

What I would really love from companies is to sort of like this time capsule where you can see what their processes were when they were like 20 people and 50 people and 100 people and 500 people. Because when we were at Stripe, we were trying to figure out our planning process. We actually talked to like Amazon and Atlassian and Apple and like all these companies that we really, really respect and look up to. And I remember, you know, taking down notes from these companies and being like, yeah, like this is awesome, but There's no way that this will work for Stripe. Stripe was so much smaller than any of these companies. So yeah, they were showing us where we had to go, but no idea how to get there. And so yeah, Lenny, maybe you can help us figure out time capsules for companies and what processes make sense.

I am working on that roughly, company by company, but how they think process and about process, and then maybe I'll check in every couple of years.

Cool, I love that. But yeah, to answer your question directly, Process, yeah, it really gives people a bitter taste in their mouth, I think. Process by definition is variance reducing. Like you're introducing it because you worry that the variance in your org is too high. You want people to sort of meet a certain standard. And the cost of that is obviously is like while you're reducing the standard and bringing folks up to the average, you're also bringing other folks down to the average. And oftentimes, the folks you're bringing down are your highest performers, your most creative thinkers, the folks who I think actually don't really need process to do their best work. And so that, I think, is always the tension that you have with process. And obviously, one of the reasons why companies introduce process much more and more as companies get bigger is because it's harder to sort of get all these folks who kind of don't need processing. You actually want to reduce the variance. This is actually a little bit of an aside, but it's kind of relevant. So I'm just going to mention it. You can feel free to let me know if it's too much of an aside. Let's get into it. But I was just listening to this awesome podcast with Russ Roberts, who's a professor, he hosts EconTalk. I don't know if you listened to it. And he was interviewing this guy, Ian Leslie, who's just this great writer and who's just, I think, written this article that's basically something along the lines of what it means to be human in the age of AI. And I thought he just articulated this idea so well, which is we're all really so impressed when we see ChatGPT 3 spit out this sort of piece of writing that feels very human-like. But what we're kind of forgetting is that over the last 10, 20, 30 years, we're actually asking humans to be a lot more robot-like. and that like we're really asking everyone to like standardize in a lot of ways and like we're making people a lot more formulaic. Like if you think about how we ask people to or how we teach people to write, it's like, well, there's five paragraphs and there's your opening paragraph and like there's, you know, you state your topic. And so anyway, I think like the point is like we actually, I think, you know, the more formulaic, the more sort of like mass produced you're trying to make something, the more you kind of quench people's creativity. And I think sort of the further away you get from the really, really high highs, and that to me is kind of the cost of process and rules and templates. So if you're going to introduce it, be really, really sure that you're okay with that cost and give folks escape hatches. So I've started referring to this as the MVP, the minimum viable process. So if I give folks a template, I'm like, look, use the template. But if you want to break out of it, please absolutely do. And start writing this at the top of templates now. It's like, if this doesn't work for what you're trying to explain, don't use it. But just know that this is the minimum viable thing. This is kind of like, we're setting the bar here, but go higher if you can, please. So anyway, that's my sort of long diary drive on process. With all that said, I do think that companies, there's just a set of documents that are really, really valuable that every sort of biggest level of the organization should have throughout its life cycle. And then, you know, I think how sort of involved it is or how long term it is really depends on, you know, how big you are. But, you know, those, those, those documents to me are the charter. So, what is your mission, your vision and your strategy? The goals, what are you aiming to do and how are you going to measure success? And then the roadmap, what is the thing that you're shipping? And I think the whole company needs these, the whole function, so like in product, you need all three of these, and then within each team, you need all three of these. And I've kind of noticed two things about this framework for myself. One, I've actually noticed that teams often work their way from the bottom up versus top down. They start with a roadmap. They're like, what are all the things we're going to ship, especially early on? And then they kind of work their way into a charter. But I think it's really, really worth it to start from the top down. And then the second thing that I've noticed is that your time horizon really shifts as you get more mature. So if you're very early on, you don't have PMF, you should have a charter, but your charter should be three months in the future because you can't look that much further. And if you're kind of a company that's humming and going, your charter is probably, the horizon is maybe a decade. So.

Wow, there's so much there. I could go in so many different directions. One thing I wanted to follow up on a little bit is this idea, such a great point, that process brings the best people down and kind of averages them out to kind of create consistency. And I'm curious if there's anything else you've seen succeed in allowing the best in most innovative I know part of it is probably hiring amazing people, but yeah, is there anything else that's like a scape hatch or was just like, oh yeah, this person just go, go figure this out.

To me, the unlock for organizations is managers for this. Because like you need managers to both detect the high performers and be like, this person doesn't need the process. And then you need managers to give that person air cover to be like, we're honestly, because what you're doing is you're giving them some special treatment and you need to be kind of okay with that. I mean, you also need to be okay with the organizational costs for that. Like Claire, who used to be the CEO of Stripe, would always say, are you willing to break the org for this person? And I always thought that was a really nice framing and you kind of need to decide who you want to do it for too. But yeah, some people are just that good that, yeah, of course we'll break the org for them. They're going to break the org in the best way possible. I love that term.

I think Claire's coming on this podcast. She just wrote a book, right? Is that the same person?

Yeah, she just wrote a book.

All right, we just booked her. Come and maybe we'll spend some time.

That's awesome.

Do you have any other product building philosophies that you found especially useful in your time at Stripe, at Retool, anywhere else?

I have a couple of, I guess, sort of like mental models I use. The first is build for your best user, not your worst user. And what I mean by that is I think it's actually really easy to kind of get or to sort of focus on what if there's abuse of this product or think about all the ways in which the product won't be used well. And then you kind of end up sort of shaping the product in these really weird, funky ways to make up for that. Whereas in reality, the worst users are always like, they should be a fraction of your users anyway, so you shouldn't really, really be building for them. I think a really good example of this is onboarding processes. where you're probably going to be building an onboarding process where you're trying to maybe sort of collect a lot of data or trying to figure out like, Hey, how do I help this user who maybe isn't sort of well suited for my product be successful? Really should just be building an onboarding process for the user who's going to jump into your product and get it immediately. I was thinking about this actually just the other day because we're in this product review and we're talking about this other new product that we're thinking of exploring.

More products.

And we're going- Never ending.

Yeah, exactly.

And we're kind of going down this path of like, oh, well, you know, if this gets really big, like there's going to be all this abuse of the product. And Anthony, our founder, was like, wouldn't that be an amazing problem to have? We're like, yeah, that's a really good point. So we kind of, you know, put.

That on the back burner. That's an interesting approach because usually you're trying to optimize a flow, get more people through it, which are the least good users. You're saying you found more success just like focus. Is this more initially focus on the users that will understand this and be excited about it and make that work really well and then kind of over time build on that.

Yeah, totally. Yeah, sure. If you're like at the end, you're like trying to optimize, et cetera, absolutely. But in that sort of early product development stage, it's like, it's just not worth it to be too focused on that.

Sweet.

That's one. The other that our head of design, Ryan, I think always kind of reminds us of is build the scooter, not the axle. So, you know, if you're trying to build the minimum viable product for a car. Don't sort of build just the wheels and the axle, build the scooter first, and then from there you build the bicycle and the motorcycle and then the car. And it's always just such a good reminder. You always kind of want to start building the whole thing, but really try and think about the slice that gets the customer to complete value on a smaller thing first. So with Ritual Mobile, for example, there's just so much you could be building there. And we decided very specifically, like, hey, we're only going to build this for companies that have field workers and they would need to do inventory management. It's a very specific slice, but it helps get through from something that was actually viable beginning to end.

Got it. It's an approach for MVPs. Basically make something simple and functional, not just something barely, not actually working, but getting you to some dream product eventually.

And then the last one is, this idea of 70-20-10 split investments. I really think that a lot of product management can sometimes be reduced to funnels and portfolios. So the 70-20-10 investments model in my head is just like 70% of your building time should really be going to your core product that has product market fit. 20% of your time should be going to strategic initiatives that aren't core, but they're strategic to the company that you know you have to do them. and then 10% of your time should be going towards bets.

That's actually exactly the same heuristic I've always used. One question there is, how do you think about maintenance and bugs within that?

Yeah, to me that falls squarely in the 70%. So yeah, core product tech debt, the stuff that you kind of the maintenance mode type stuff. And core products, obviously, you're also doing a bunch of new stuff there too, but yeah, no more than 70% of your resources should be going there.

What did you say the 20% was?

Strategic initiatives that aren't your core, but you just know based on your strategy, you have to do these things.

Got it. So the way I break it up is 20% I think is where I put bugs and maintenance, and then 10% was like big, big ambitious bets. So it was similar ratios, but different things go into the buckets.

Cool.

But let me ask a similar question. How do you just at Retool and maybe even at Stripe think about Finding time to do maintenance and bugs. Do you build it into roadmaps? Do you set off time to just fix all the bugs? I know it's probably an evolution and goes back and forth, but any thoughts? Yeah.

This, I think, is it's really one of the trickiest parts of product management in my mind. We don't have a company-wide sort of process on this. It's pretty team-specific. Some teams do like Friday bug bashes. Other teams are just kind of you know, as products roll out, we'll, we'll kind of work on them as they go. I was actually speaking to a product manager at a different company who mentioned that they just spent the last 18 months doing basically just product polish and bugs. And I, yeah, I think they landed in a place where they just, they had to do it because they just had so much debt. But, you know, luckily we're not there yet. But right now we let most of the teams just kind of do it, do it themselves and, figure out what makes sense for them.

Okay, awesome. I wanted to go back to something that I've heard about Retool that you all are really good at, which is PMs being really close to customers. And I'm curious what you've done there or what the team has done that it's been really effective.

Well, I do think every good product team figures out how to sort of get close to customers, but just based off of my observations from Retool and what I've seen at other companies are, A couple of things that are more pronounced at Retool than I've seen in a lot of other places. The first is we have a lot of PMs who used to be in customer facing roles at Retool. So, I obviously am a big fan of that. I think it's, there's just nothing like really understanding the value of your product at the moment where a customer actually has to put money down. And so I really like that about, you know, PMs who have had those conversations with customers and they really, really get it. Second is because the virtual product is so technical, and I do think this just depends on what product you have. Like, our PMs are really very technical. Like, everyone has, you know, either a CS degree or has done some sort of engineering in the past. Third is we use Slack very heavily to, to talk to our customers and interact with them. So we get this by, I think, hundreds of Slack channels with customers every time, you know, where testing a new product or we're kind of a new customer who's somewhat large comes online, we will work with them in Slack to get their off-the-cuff feedback just back and forth. It's really nice. We just have this direct line to them, which is awesome. And then the fourth thing we do is we use Retool to build Retool. So our product roadmap lives in a Retool app. And the app that we use to, you know, to have feature flags for particular features is a Retool app. So PMs are just in the product all day long, every day. They're their own customers in a lot of ways. And that, that really helps as well.

Wow. I didn't know that. That is very cool. So you built like a task management product using Retool.

Oh, yeah. Well, we build many, basically all of our internal tooling happens in Retool. Like our PMs, all their team dashboards. all of their, you know, all their, yes, their task tracking is in Retool. Submitting linear requests or bug requests happens in Retool and then goes to linear. So it's just, yeah, it's how we run the company.

You haven't been able to replace linear yet. That's cool. Big fan of linear. Okay, two more questions and then I'll let you go. One is around product like growth, which I don't wanna get tooed into. We talk about it a bunch on this podcast. But it's interesting that Stripe was very product-led growth. It was just self-serve PMs, engineers, whoever, just started using it. It grew and became enterprise-y down the road. Retool, on the other hand, I think people think it's product-led growth. I imagine it's actually sales-led mostly. And so I'm just curious what you've learned about the difference between building product and leading product teams within a sales-led org versus a product-led org.

I have a couple of insights on that. One actually, I think interesting insight is that teams that have one always want the other. Like whenever I talk to candidates, I feel like you can always tell what their company has because they'll be asking a lot of questions about the opposite. Like if they're probably a growth company, they'll be like, well, how are you guys thinking about enterprise? And then if they are more of a sales-led growth company, they're like, well, how are you guys thinking about your sales or promotion or your product-led growth. My main takeaway just in sort of, and I don't know if I would even serve as the dichotomy of product-led growth or sales-led growth with Retool. I think we do have a lot of product-led growth, but we work with a lot of enterprise customers and a lot of revenue comes from enterprise customers and we have a fantastic sales team. And I think the main thing is that you just have to really figure out you're like interaction mechanisms with sales and that that just has to be so, so tight. And it goes in both directions. There's obviously like how you get feedback from them because they are so often on the front lines talking to customers, more so than product managers even. But it also goes the other way, like if you are going to ship something or if you are going to, you know, put out a roadmap, like how do you make sure that the sales team has everything that they need? The sales and, you know, in Richel's case, the success team and the four team has everything they need to accurately talk to customers about that. And I've always actually found that really hard because it's really hard to figure out the right altitude. If you're giving a presentation on the roadmap, people are either going to feel like it's too high level or it's too low level for folks who maybe are new. So this year we're actually trying something different. We are doing a science fair where each product team has a little booth And they get to sort of stand there and anyone who has questions about the product can come, you know, from the GoToMarket site, can come and ask questions and get demos and go like as deep as they need to. So I'll let you know how that goes, but I'm excited to, to, to experiment with it.

Okay. Are there going to be foam core boards and will there be ribbons?

There, there may be prizes. Who knows?

Oh my God. I can't wait to see a picture of this. Final topic. You have this concept that you call the product talent portfolio. What does that mean? And how have you found that useful in your product leadership life?

Yeah, it's back to the all of product management is portfolios and funnels. I think it's really tempting as a manager to build a team in your image because you understand their skill sets and you value those skill sets. And you're going to be able to detect and assess those skill sets better. But the best product teams in my mind have really figured out how to balance the, you know, the talent portfolio. So instead of sort of having a bunch of PMs who all spike in one particular area, figure out how you can, you know, have, create complimentary skill sets for the whole PM team. So the whole is much stronger. And one way in which I like to do that is I really like to balance product teams with homegrown product managers who really get the product. They've probably been in it. They're amazing culture carriers often. They really set the tone, but they may not have seen product management at other companies and they may not have some of the sort of more conventional traditional product management skill sets. So I want to balance that with product managers who've come from other companies and have done it and can bring a little bit more of that product manager rigor. Even though they don't have that sort of, you know, the core and, you know, the Retool case, sort of the core Retool product management, you know, vibe. So that's one example. I think there's like others as well. Like some PMs are just incredible execution machines. Other PMs are like amazing visionaries. Like you kind of want a little bit of all of them. And you want to also balance within all of your different pillars. So, you know, we have three different product pillars at Retool and kind of leads for all those pillars. And I always push the leads to sort of hire people who don't look like them so that, you know, we have balance everywhere.

I love that. Do you actively, as you're hiring, like, Hey, we're strong in these areas. Here's the person we need here. We want like the specific super strategy mind person. Is that actually how you think about this?

Yeah, yeah. I do a little personal exercise for myself every six months where I like sit Like sort of chart out the team that we have today and like write down all the strengths that we have, all of the weaknesses that we have as a team. And then I tried to sort of hire specifically for those weaknesses.

Amazing. Any last pieces of wisdom or thoughts before we get to our very exciting lightning round? I think that's it on my end.

Okay, we got through everything we were.

I was hoping to get through. So with that, we've reached our very exciting lightning round. I've got six questions for you. Are you ready?

I'm ready.

What are two or three books that you recommend most to other people? Bird by Bird: Instructions on Writing and.

Life by Anne Lamott. It's incredible, so touching, really great. Also, I think product managers should be great writers, so love that. And then the other book that I was going to say, but I'm glad you mentioned Claire, is Claire's book, Scaling People, which is coming out. I had a very small hand in ghostwriting for her a little bit in the first draft, so. I use a lot of the early chapters from that book still in management, and I still recommend, you know, recommend sort of the tactics in there. So I'm excited for her to get it out. Wow.

I've been hearing about ghostwriting as a career recently, and it feels like it could be an option for you down the road. I am excited to read the book. I haven't gotten a copy yet, and I think you can preorder it now, and we'll link to it in the show notes. Nice. What's a favorite other podcast that you.

Like to listen to other than maybe the one you're on right now? Yeah, this one's great. I can't choose between these two. One is Lex Friedman and then the other is EconTalk by Russ Roberts. I really like that both of them are like their agenda is like curiosity, which I love.

Totally resonate there. What is a favorite recent movie or TV show that you've enjoyed?

Is it too basic to say White Lotus?

Cool. That's like the most mentioned one, which says a lot, right? That just says how good that is I love it too. But that works. That works. Season one or season two?

Oh, season two all the way.

Yeah, same. I'm excited for season three. No spoilers. Favorite interview question that you like to ask candidates?

To what do you attribute your success and you can't say luck?

Interesting, because most people look for, do they believe it's luck versus like their innate skill? So you don't even allow them to say luck interesting.

Yeah, because, you know, I think humble people will always say luck in some way. And, you know, I always kind of wanted to, like, did you, like, how self-aware are you basically? And I think, and how curious are you? And I think people have really sort of gone back and reflected on why are they where they are today? Really, really says a lot about how they think about the world.

I love that. What are some SaaS tools that you love or use often at your current company or anywhere else?

Yeah, well, first and foremost is Retool. And then the other SaaS tools that we use a lot, Slack, Gong, Linear, and Full Story.

Awesome. Favorite new product that you found maybe in life, maybe at work? Yeah.

Have you heard of Rewind?

Yes, I have it running right now.

Nice. Yeah, I mean, I think it's really incredible what they've done.

Maybe describe it briefly, just so folks know what that is.

It basically records everything that you're doing on your computer and then makes it really easy for you to search it, which is, it's just incredible. Like, I don't know if you work this way, Lenny, but like, I feel like nowadays I don't really go to like tabs anymore or I kind of directly like search for the thing that I'm searching for. And I think Rewind just fits that user behavior so, so well. Amazing.

Eika, this was so much fun. We got through everything I was hoping for and more. Two final questions. Where can folks find you online if they want to reach out, learn more, see what you're up to? And how can listeners be useful to you?

Definitely find me online on Twitter. It's @EikaDM. And how can they help us? Try out the Retool product and give us some feedback.

Amazing. retool.com, right?

Yes.

Awesome. Thank you again for being here.

Thanks Lenny.

Thank you so much for listening. If you found this valuable, you can subscribe to the show on Apple Podcasts, Spotify, or your favorite podcast app. Also, please consider giving us a rating or leaving a review, as that really helps other listeners find the podcast. You can find all past episodes or learn more about the show at lennyspodcast.com See you in the next episode.