Pod Insights Lenny's Podcast Making and Leading Products People Love (Scott Belsky) | Lenny & Friends Summit 2024 About this episode Scott Belsky shares profound insights from his journey building Behance and leading product at Adobe, revealing the human elements that drive successful product development. His framework combines practical wisdom with deep understanding of user psychology.
Recorded live at Lenny and Friends Summit on October 24th in San Francisco
Transcript 0:06 Speaker 0 Hey everyone. All right. Here's the plan. In 30 seconds, I want to share a little bit of kind of the laboratory of products that I've been exposed to. Then I want to share an updated curated selection of the insights that I hold dear as I navigate product decisions and help some of you as well. With the hopes that maybe two or three of them stick in your mind and become things that you can talk about with your teams and consider in your daily practice. And then finally, I think there is a thread throughout all of these that I want to touch upon at the end about the very human aspect of our discipline in a world where we're oftentimes talking about the non-human aspects that I think is really important to touch upon. And so really quickly, you know, my passion has always been to organize the creative world at work.
0:52 Speaker 0 Whenever I told people I wanted to do that, they were like, yeah, good luck with that. But uh that was that's been the passion. It started actually as a physical product of all things. Um it evolved into books and conferences and a company ultimately called Behance. Behance is a platform of 56 million creatives around the world showcasing their work, and you can sort through the work by you know, the by the person who did what for what agency with what tool, um, in what discipline, in what color. It's the organize, it's organized in the creative world at work. We joined Adobe in 2012, and then I was faced with all these new challenges and things that I always considered broken as a customer of Adobe, and the idea was let's go and you know try to fix some of these things. I thought I'd be there for a few years, ended up being there for three years, leaving, coming back for another seven years, overseeing creative cloud, bringing our products to the to the web, to mobile, products like Lightroom Mobile, products like Adobe Express, um, and then have now you know been able to work with a lot of startups over the years as a very early product advisor and seed investor as they sort of navigate some of their product challenges and and and write about it a little bit as well.
2:05 Speaker 0 30 seconds or less. So now I want to talk about some of these essential insights. And the way I've been thinking about um these types of insights in our in our nomenclature is what is it, what does it mean for all of us to sort of see the matrix as a product leader? Um, what are those moments where something just kind of clicks and those become cardinal uh insights or perspectives or um just lenses that we uh that we take that make all the difference in our practice? And uh and this is something I've been thinking about for a few years, but I've lately uh and over the last month, thanks to the forcing function of this conference, tried to put together uh some of the ones that I think are most important to think about. And so we'll ask this question to kick off. You know, what is the equivalent of seeing the matrix as a product leader?
2:53 Speaker 0 And I'll start with the most obvious, and then we're gonna get a bit more nuanced and uh and dynamic here going along. But data is a compass, not a map, right? The idea that we ultimately need to determine what mountain to climb, right, with the intuition we have with our understanding and with our empathy. Um, and yet data is ultimately what helps us climb the mountain we choose. And but this is something I actually often come back to in a discipline that's so driven by data these days, is to remember that data is to help me get from the bottom to the top of the mountain, but not necessarily to choose the right mountain to climb. And so I just want to put that up there very first, because that's something that I that I think is very important. All right. Another thing is to seeing the matrix is that recognizing that with new products, you need to prioritize and optimize for the problems that you want to have.
3:45 Speaker 0 Now, what do I mean by this? When you're building especially a new product or launching a new feature, what are problems you want to have? You want customers to be using these products or features to this to the extent where they start to come to you and ask for things that power users would want. That's success. What you don't want is for them to get stuck in the onboarding. What you don't want is for them to realize that because of security issues, they can't even use the product. And so I've started to think about everything on a product roadmap in the standpoint of things that you, you know, and this is a time scale here. You need to optimize for shallow value before you optimize for the deep value. And it's a sort of contrarian view because you would argue that we should build products with this great deep high willingness to pay value from the start.
4:34 Speaker 0 When in essence, if most if any of our products you know get any more than 30% of the people at the top of the funnel through, they're world-class products. So why not optimize for all the things that help people discover the shallow value quickly? aka feeling successful quickly, which is in some ways the definition of product-led growth, before they discover that that deeper value. And in some ways, then when you're looking at a roadmap and a list of things to debate prioritization around, thinking about well, what do we want people to be requesting? Let's not do those things. Let's instead do the things that could conceivably become obstacles. And recognizing that whatever the MVP is, I think one mistake we often make is we forget how much gravity an MVP has. Whatever you put out, your instinct as a team will always be to iterate from there.
5:22 Speaker 0 And so to also make sure that while it can be the minimum viable version of most things, the thing that distinguishes your product the most needs to be done extraordinarily well before you put it out, because it's hard to iterate your way out of not getting that right. Another thing is about realizing that we should only do half of what we want to do. I mean, this was probably the greatest mistake we made in Behance, which is now a far simpler product than it was actually when it launched like well over a decade ago. When we had groups, we had something called tip exchange, we had something called circles, we had something called work in progress, and we we realized that every time we killed one of those things, the engagement on the core feature that actually mattered most, which is just people publishing projects to their portfolio went up.
6:10 Speaker 0 And so we started going on these killing sprees. And I think that was one of those moments when I realized that it's I think our practice in product is much like the the bonsai cultivating master, where you're constantly pruning those beautiful branches, those things that are just taking flight, you're actually cutting to enrich this kind of core. And so that that notion, that voice in your head of half, half, half is somewhat seeing the matrix as a product leader, knowing that is that is what compounds our likelihood of product market fit. Defying the product life cycle is something I want to spend a few minutes on because we have to obsess over this. And and what do I mean by the product life cycle? It's as simple as this. Users flock to simple products. We love simple products, both as enterprise customers and as individuals.
6:58 Speaker 0 Those users, you know, all sort of inundate a product, and what does the product do? It takes those users for granted by saying, oh my goodness, now we need to start to monetize. Let's add power user, let's let's listen to power users and see what they want, let's add features. Um, and uh and before you know it, the product becomes more complex, you're focusing on adding now that deep value, and what happens users flock to simple product again and again. And I feel like this is almost what fuels the Silicon Valley life cycle of so many industries of product is simply this you know, strange flywheel or I don't know, or cycle, in other words, you know, that we go through. And so, how do we keep products simple is probably one of the most important product questions that we always have to be asking ourselves.
7:44 Speaker 0 And I think a big part of this is always ensuring that the product is working for the newest cohort of new customers, recognizing that every new cohort's different. You know, early cohorts are more willing and forgiving, and then later cohorts are not willing and forgiving. They are there sometimes against their own wishes, and they better get a very simple experience and a different experience in the onboarding. So I think that the realization here is the recognition that every customer of any type of product, I would argue, even in the enterprise, in their early stage of using the product is lazy vain and selfish. They're lazy in the sense that they don't have time to learn something, right? And they don't have tolerance for onboardings that are bad, instructional videos, more than one line of copy.
8:30 Speaker 0 And so part of the solution here is to show rather than explain how to do something. And you know, we've we've explored with some of those ideas, for example, in Photoshop with sliders to just and and tool tips that just basically show people in the product how to do something. The only thing I think better than that is to actually do it for the customer. And this is where templates and chat UIs and presumptuous defaults you know come into play. Um, and and we've tried to actually build new features on top of features in Photoshop, for example, to just do things for you as opposed to having to teach you how to do them anymore. Um moving on here. Oh, there you go. So my favorite saying, I think in this whole space is uh by uh my friend Dave Morin that the devil is in the defaults. And really, I I I can't emphasize this enough.
9:17 Speaker 0 Oftentimes we think about the default state of our product in the final mile of shipping it, when in fact it might be the only state of the product that most of our customers ever experience using it. And so to really argue and think about what that default experience should be is an absolutely crucial thing to do. Vain. Products need to make us look good and quickly, especially in the early stage of using it. A big thing about this is what I call ego analytics. How does the product make me feel good about myself in my job in my life? You know, we've always seen the hearts and likes and appreciate buttons and everything else on social, but it's interesting when you talk to people who work at companies like Instagram, what they will admit is that the engagement in a product like that actually goes up after the user posts something, which suggests That maybe people are actually going back to the product more often to see who saw their content than to see other people's content, which is the more intuitive reason why people would go to a social media site.
10:17 Speaker 0 And so that's an example, I think, of ego analytics of merchandising to the customer what's happening. You know, and certainly in the creative world, we've seen that when customers can see who likes and who appreciates who views their content. We made the decision to let everyone appreciate work, even if you weren't logged in, because of course, you know, it's giving that gratification to the customer that their work is getting appreciated. Why would you want to lock people out of expressing that and receiving that feedback mechanism? Progress begets progress. There are a number of studies. There's a study by this woman named Teresa Mobley at Harvard Business School about motivation in organizations, but I think it applies to customer basis as well, where people make more progress when they feel like they're making progress as product leaders.
11:05 Speaker 0 We need to merchandise to our team the progress they're making to get their continued commitment. And similarly to customers, helping customers understand the progress they're making through our products actually makes them make more progress. And it's important to keep in mind. And then the last piece is just around this selfishness. People need to feel like they're quickly benefiting from spending a little time on this, because most customers at first carelessly engage with the product. You know, Pinterest, before it was a discovery engine, it was simply a very immediate utility to store your bookmarks. And uh the idea of people suddenly using it for other things beyond was was more of a hope uh than an actual reality in the beginning. Getting paid faster, right, was the value proposition of Stripe when other I'm sorry, when with Square, uh, when other value props didn't work about the terminals, but they realized that small businesses just wanted to get paid faster.
12:01 Speaker 0 It was like a very selfish need, and they didn't really care about the point of sale system and everything else like that. But that was the value prop that that worked. Similarly with Stripe, the two lines of code you could argue is capitalizing on the laziness uh of a developer who was saying, hey, you know, I want commerce, but I want to have to work with merchant providers and all of this spaghetti of uh of work. And so this notion of two lines of code was actually something that really pierced through uh that that that industry. Which goes to suggest a question that I always grapple with personally around copy, believing that effectively copies part of the product experience and needs to be born from within a product team. It's part of the overall end to end, and most companies are not structured that way or will even outsource copy to some degree, and that's never really an intuitively made sense to me.
12:53 Speaker 0 Because we're very we're very skeptical of long-term promises that that products make, and we need to make sure that we're merchandise the value we're getting now that has to be nailed. Uh, and I think that it goes also into the the defaults and the types of capabilities that we surface in our products in the first mile experience. Okay, so moving on to another, you know, what does it mean to see the matrix as a product leader? Recognizing that you need to get customers to talk about their problems and not your product. And I don't know about you, but I've held actually only a few focus groups uh for behance very in the beginning, and everyone told us they didn't need another social network, they had deviant art, they had um Facebook, you know, they had um they had LinkedIn, and they were like, last thing I need is another place to put my work.
13:41 Speaker 0 And then we said, well, what are your what are you suffering from? And that's where we understood that people didn't get attribution for their work, their portfolios were always out of date, and they weren't reaching new potential people that would hire them. And so always making sure that you focus on the problems, which goes to say that I think that empathy, empathy for the problems our customers are are suffering with outperforms passion for the solutions. And if I were to characterize some of the entrepreneurs that I've backed over my years as a seed investor that I was most inspired by that were not successful, it was because they were so governed by their passion for the end state of what the world needed to be, and they never really had that empathy that was the difference between that 30 degrees of being product market fit off because they didn't really understand what that customer was going through on a daily basis.
14:35 Speaker 0 Even though they were saying this is what they wanted, they really needed to get the credit from their boss or whatever the case might be that actually drove the utilization of the product. Another insight is that the most powerful and affordable driver of product growth may not actually be what customers want. A great example of this was the early days of Slack. I mean, our team was using this product called Hip Chat from Atlassian at the time that Slack rolled out, and then our team started to use it only because of the animated GIF capabilities and uh sending a random one to your colleagues and not knowing what it was gonna be, and then it became a place where people would say, Hey, I'm going to coffee. Does anyone want to come with me? And then before you know it, it became a utility that people were using every day.
15:22 Speaker 0 Novelty does precede utility, and we have to sometimes remember that when we are planning and prioritizing roadmaps for our products, it's very easy to be reasoning-oriented of what we know the customer's saying they want as opposed to the things that they'll get excited about, at least initially. Um, which brings to a moment, you know, or just uh a point on product-led growth. People don't rave about products doing what they expected. They only really rave about products doing what they didn't expect. In Tesla, you know, no one was in the early days saying, I just love how this car functions as a car as much as they were saying, oh, this cool rainbow road Easter egg where you like do this on the dashboard and touch this button and this happens. You know, these were these moments of delight that the customer did not expect that drove the exchange, the the discussion, the people that were raving.
16:13 Speaker 0 And so it's important. Now I oftentimes wonder, you know, how do you actually get a team to prioritize in a roadmap session about building a high performance car, something like Rainbow Road? But but I think what you have to recognize is that your job is to get your customers to discover and adopt your product. And if novelty does precede utility, if people do rave about things they didn't expect, how do you prioritize those things? Another insight that I've had that I really hold strongly as a product leader is whenever I review any part of a product, any screen, I'm ultimately asking myself three questions in every moment when I see something new. I'm asking, how did I get here? What do I do now? And where do I go next? To me, this is almost like the ultimate object model test. It's knowing whether or not the customer who's usually carelessly using a product because they're busy and they're distracted and whatever else, if they do they know the comfort of how they got to where they are, or is your product doing the equivalent of dropping them off in the desert somewhere without a map?
17:15 Speaker 0 And so constantly asking this question of orientation boils down to whether the actions are clear or the actions are confusing. Are you pushing enough emphasis in the direction that they're most likely to go? Are the defaults correct? You know, I find that these sorts of questions became a bit of a holy grail uh for me and a number of folks that I work with, uh, which is why I wanted to share them with you. Another insight that A plus designers, you know, are ultimately this cheat code. And I have to admit, and if I look back to my career as a product leader, the things that I am most proud of were the result of a partnership with designers that made all the difference. Part of the reason is this power of this prototype, right? Being worth a hundred meetings. People going around a table debating and discussing whether two products could connect or whether some new feature makes sense, or whether some MA opportunity could actually connect to the user's experience, and then a designer suddenly shares a prototype of what it looks like, and it's like magic.
18:15 Speaker 0 You know, it is that hot knife through the butter in the sense that everyone either agrees or disagrees, but suddenly it prompts decisiveness. What an example of the power of design represented at the table. Another insight that I've learned was kind of humbling, you know, for those of us who are also have engineering responsibility is that perceived performance can matter just as much as actual performance. I remember the early days uh when Pinterest started to scale in like 2011, 2012, and the team was like racing to keep this product up and running. There was this big roadmap, you know, uh of how to get scalability and performance and improve performance of the of the site loading. And then the designers came up with this idea to basically load the color of the pins before the actual pins.
19:01 Speaker 0 And suddenly overnight, customers were like, oh my gosh, Pinterest is so much faster. And it wasn't. There was no change under the hood other than a design change. Another example of having designers at the table to solve what was actually an engineering problem, but was really a user experience problem. And then in the age of AI, more and more of our choices in every industry are being made at the surface, right? And I think that's another argument that design will increasingly commoditize tech to some degree, and that having this designer empowered and have a seat at the table will be more and more important. You have to prioritize grafting talent as much as hiring talent. This is something that I've learned the hard way many times, right? When you hire some amazing person, but you fail to suppress the immune system of your team that has always done it a certain way to allow this new organ transplant to come on and take hold in order to change the way your organization works.
19:57 Speaker 0 And so you have to suppress the immune system to allow this new, oftentimes senior person or someone with great um, you know, great, a great history of executing, but does things slightly differently than you. And it's amazing to me how we have a huge industry to hire talent, but grafting talent is sort of like, oh yeah, like make sure they get a computer and a login and they'll figure it out. Like so I think that us making sure we graft talent onto our teams very thoughtfully is important. A few other thoughts real quickly on the talent front, because I know that's a theme of a lot of uh this conference. You know, is hiring for initiative over experience. I think a lot of people have learned this the hard way, or in some ways the easy way when they hire someone because they can't they can't afford the experienced people, they hire the people with initiative, and those people just come on and just do amazing, amazing things beyond your expense expectations.
20:52 Speaker 0 It's because they had a history of taking initiative of what matters to them. Celebrating impact over tenure. And when I came into Adobe, they had these tenure awards that I was supposed to do every quarter at my all hands. And I was like, wait, I'm celebrating someone for how long they've survived here as opposed to someone who's come in and made an impact on day and on day two. Like that doesn't make any sense to me. And so I replaced them with the impact awards. And I think that's such an important thing to especially bring in new talent and make them feel like their contributions are valued despite how long they've been there. And then finally, that you have to collapse the talent stack every chance you get. When you do have that opportunity to hire that product leader with a design background, or that designer who was a front-end engineer, you know, or the growth hacker who's also a designer, that conduit that you then collapse is an incredibly effective way to unlock something new in your organization.
21:49 Speaker 0 And so I can't emphasize that enough. Whenever I've had the opportunity to collapse the talent stack, it's always surpassed my expectations. You only can expect what you inspect. This is important, right? Each of us has to find the few level ones, the few things that you want to hold yourself directly accountable for. And if you hire other product leaders and you delegate effectively to other people, that's great. So long as you hold yourself accountable to the things that are most important to you by inspecting them rigorously. And this is actually something I've learned from our CEO who had to delegate almost everything, but he always has a list of a few things that he goes so deep on inspection, and I uh, you know, I feel like that's a that's a best practice for all of us. That you level up when you learn to know when to amp up this volume of your own intuition.
22:39 Speaker 0 And this goes to that earlier point around the data. You're gonna be thrown data constantly. You're gonna be told best practices from other people coming in, maybe people like me and others, and yet at the same time, knowing when to tune into that feeling over the thinking that you have from the more academic side, knowing that conviction should outweigh consensus when you have faith and trust in somebody who feels suddenly super strongly, you know, that's when you can make a non-popul decision that differentiates you in the market. So learning to gain that con value, that conviction over consensus at times, to lead with curiosity over criticism. When I am frustrated or disappointed, what I challenge myself to do in product meetings is just to ask in a barrage of questions, to tune into my curiosity on what's off, what's wrong, and just go poking with questions and questions and questions to the point where the team can suddenly you know realize something's wrong, or I realize that I was wrong.
23:36 Speaker 0 And then I think, you know, la last couple points here, but great product management these days in a world where so much of product management is being offloaded to compute, and we're starting to get these new this new generation of agents that complement us as managers, which is both exciting and scary. But I think that more of our jobs will become less about how to do something and more about what to do. And if you think about this dilemma of the you know what to do versus the how to do, it's kind of the it's kind of this new era where it's more about taste than about skill. As skill becomes offloaded to compute and and AA agents are assisting you, it's really your intuition and your taste that is ultimately going to make some of the most important decisions in your organization and in your product.
24:22 Speaker 0 And those the what to do is a human discipline. You know, models are trained on the distribution of data in the middle. But great, great decisions and new creative solutions to problems always fall outside the distribution of what's been done before. Otherwise, they would have been done already. And so remembering that our job is now to start to think about outside the distribution of what's been done before is I think such a new critical practice of this thing that we all call product. And finally, that the best new products in the world, they really do take us back to the way things once were that we long for, but with scale and efficiency. And if you just think about a few examples of that, you know, we want our we want to be known in our in our towns. We just like you know, 300,000 years before the industrial revolution, we were known until we became anonymous.
25:11 Speaker 0 And so maybe things like marketing tech at its best can help us be known again. Hopefully, we know how we're known or why we're known. We want scarcity, LinkedIn is about networking instead of job boards. It's about the old way of doing things where it was through recommendation and referral as opposed to these, you know, anonymous job boards. AI-driven interfaces in some way take us back to natural language, which is the old way we used to do things before before these graphic interfaces. And so I think it's just a great litmus test. Um, and then finally, finally, product, right? This is the thread. Product is mostly ultimately this people discipline. These product decisions that all of us make, they need to be boiled down to this deep understanding and empathy for natural human tendencies, that lazy vein and selfish, that first smile.
25:59 Speaker 0 That to build these products of the future. We must understand the parts of the past that we long for, but with scale and efficiency. And finally, that a product, you know, this is the craziest thing, and I think we all know this in a room. When you use a product, you can kind of tell by using the product who did and didn't talk to each other, right? Did they get along? Did they not get along? Was the copywriter in the room or not in the room? Did the marketing department have to get the product leaders sign off before they ship this thing? The Apple doesn't fall far from the tree. So thank you so much for your time and and looking forward to seeing you at the rest of the conference. Thank you.
Previous Seth Godin's best tactics for building remarkable products, strategies, brands and more Next How To Win Friends & Influence Decisions (Julie Zhuo) | Lenny & Friends Summit 2024 Page updated August 25, 2026 · transcript and summary are machine-generated
