Pod Insights The Twenty Minute VC (20VC): Venture Capital | Startup Funding | The Pitch 20 Product: Marty Cagan on The Four Questions of Great Product Management, Product Lessons from Marc Andreessen, Ben Horowitz and eBay's Pierre Omidyar & The Difference Between Truly Great Product Teams and the Rest About this episode Marty Cagan is one of the OGs of Product and Product Management as the Founder of Silicon Valley Product Group. Before founding SVPG, Marty served as an executive responsible for defining and building products for some of the most successful companies in the world, including Hewlett-Packard, Netscape Communications, and eBay. He worked directly alongside Marc Andreesen and Ben Horowitz at Netscape and Pierre Omidyar at eBay. In Today's Episode with Marty Cagan We Discuss: 1. Entry into the World of Product From Engineering: How Marty first made his way into the world of product, having started life as an engineer? What does Marty know now that he wishes he had known when he started in product? What are Marty's biggest tips to anyone making the move from engineering to product? 2. Lessons from Marc and Ben at Netscape and Pierre @ eBay: What are the single biggest lessons Marty took from working side by side on product with Ben Horowitz and Marc Andreesen? What did Netscape do right? What did they do wrong? With hindsight, what would Marty have done differently? How did Marty break all of his rules by working with Pierre Omidyar? 3. Hiring a World Class Early Product Team: When is the right time to make your first product hire as a startup? What is the right profile for that first product hire? Senior or junior? If you go for the junior hire, how do you structure the rest of the team? If you go for the Senior hire, how do you structure the rest of the team? What are the single biggest mistakes startups make when hiring their first in product? Does Marty prefer someone with or without expertise in the domain you are in? 4. Mastering the Onboarding Process: What is the optimal onboarding process for all new product hires? How can leaders ensure that product hires see and understand all areas of the business? What can product leaders do to proactively impress in the first 30-60 days? What are clear red flags that a new product hire is not working out? How long do we give them?
Transcript 10,924 words · 2 speakers
Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.
0:00 Speaker 00 It's one of the most obvious differences between great product companies and the rest, how they think of their engineers. We say if you're just using your engineers to code, you're only getting about half their value.
0:11 Harry Stebbings? Welcome back. This is 20 Product with me, Harry Stebbings. 20 Product is the monthly show where we dive inside the world of product, speaking to the best product leaders to unpack their tips, tactics and strategies to scaling product teams today. And what a show we have in store for you today with one of the OGs of product management, Marty Kagan. Marty is the founder of Silicon Valley Product Group. Before founding SVPG, Marty served as an executive responsible for defining and building products for some of the most successful companies in the world, Hewlett Packard, Netscape Communications and eBay. He worked directly alongside Marc Andreessen, Ben Horowitz at Netscape, and then Pierre Omidyar at eBay. And I want to say huge
0:52 Harry Stebbings? thank you to Avid at Ontario for some fantastic questions and suggestions today. Really did make such a difference. But before we dive into the show today, this episode is brought to you by Mixpanel, offering powerful self-serve product analytics. You know, something we talk about a lot on this show is how startups can build the best possible product. Relying on gut feelings is an expensive way to find product market fit, and that's a price many VCs aren't willing to pay. That's why Mixpanel will give you $50,000 in credits when you join their startup program. $50,000 in credits. With Mixpanel, startups find product market fit faster, helping you take your company from MVP to most valuable player. Access your real-time insights with the
1:33 Harry Stebbings? help of pre-built templates and know that at every stage, Mixpanel is helping you build with confidence and curiosity for free. Apply for the startup program today to claim your $50,000 in credits at mixpanel.com/startups with an S. If you don't qualify for the program, Mixpanel has pricing plans for teams of every size, from startups to enterprise. Grow your startup like you've always imagined with Mixpanel. And next, we have Stitch. Stitch is your all-in-one platform for authentication. From a modern reboot of traditional passwords to passwordless options like email magic links, one-time passcodes, biometrics and
2:10 Harry Stebbings? Web3 logins, you can quickly and easily integrate Stitch to power your authentication use case. Our SDKs allow you to get up and running quickly with the option to use Stitch's pre-built UI components or fully customize the experience to fit your brand's look and feel. For maximum flexibility, their direct API gives you the tools you need to build out authentication flows to fit your exact needs. Make user and customer onboarding seamless to boost your conversion and retention rates by making authentication as frictionless as possible. And Stitch have raised over $100 million from the likes of Co2, Thrive, Benchmark and Index. And their customers range from individual developers to the fastest growing companies in tech like Bitcoin.com.
2:53 Harry Stebbings? And get started today with $1,000 in credits. Sign up at stitch.com/20product. That's stitch, S-T-Y-T-C-H.com/20product. And finally, there is a common adage, data is more valuable than oil and gold. This got me thinking about my personal data and what I'm doing to protect myself online. And I thought, well, I wouldn't leave my front door whilst I'm on holiday. So why don't I protect my personal data and privacy whilst I'm online? This is why I decided to team up with the good folks at NordVPN. I downloaded their app, which is super easy to use, and within seconds, I'm completely protected whilst online on six of my devices. It gives me complete peace of mind, knowing that all my personal data, bank details and
3:37 Harry Stebbings? privacy is completely safe from falling into the wrong hands. I can even access streaming services, this is my favorite, in other countries by switching my virtual location. Hello, US Netflix! And it costs the price of a cup of coffee in London every month, a small price to pay for premium cybersecurity. And of course, I have a special discount for my listeners. To grab your exclusive NordVPN deal, head on over to nordvpn.com/20vc to get a huge discount off your NordVPN plan, plus four additional months for free. It's completely risk-free with Nord's 30-day money-back guarantee. But that's enough from me. So now I'm so excited for this incredible episode with Marty Kagan at SVPG.
4:23 Speaker 00 You have now arrived at your destination.
4:26 Harry Stebbings? Marty, this is such a joy to do. I'm sure you know, but I've loved your writing for quite a while now. But thank you so much for joining me today.
4:33 Speaker 00 Thanks for inviting me. I'm flattered to be on your show.
4:36 Harry Stebbings? That is very kind of you. I want to start there with a little bit of context. You are the OG of product management, as many have cited. How did you come to be the OG of product management, Marty, in a swift two to three minutes?
4:47 Speaker 00 My interest has actually never been really product management. My interest has always been product teams because I've never seen a product come from a product manager. I've only seen great products come from product teams. I started as an engineer for 10 years and I was lucky I was in a great place, HP Labs, and I was working on tools and products for other developers, other engineers. So very fun 10 years. I had a great time and that was during the time where they were in their prime. They were known as the most consistently innovative company in the world at the time. And that's why I went there and I had a great time. I did get interested there in more than just engineering. I started asking the questions about who decides we should build this particular product?
5:31 Speaker 00 How do they know that's going to be a good thing to do? I started asking those questions beyond the little swim lane of a typical engineer. And I learned about product, I learned about design and I learned about go to market. I learned about all these things. My lucky break happened actually because I had done all these products for developers and Netscape was really the original internet company. And I got to work for Marc Andreessen. I was, there was already a browser team, there was already a server team, but I was for platform and tools. In other words, again, doing products for developers. In this case though, was the internet, new platform, instead of just Windows. That was a dream job, honestly. I'd still be there if it was still going as a company.
6:13 Speaker 00 But after about five or six years, we lost the browser wars to Microsoft in your internet history. During my time at Netscape, I got to meet the founder of eBay, Pierre Omajar. And I really liked him. I loved his vision for the company. eBay was probably the first real app in the internet era, the child of the internet. I was brought on to be the original head of product and design. Really, he was the original head of product. That's the truth. But he was a founder and they had got some real traction, really because of his innovations.
6:44 Harry Stebbings? Can I ask you a really unfair question, Morty? I'm just intrigued. How does Marc Andreessen and Pierre Omajor compare as leaders slash bosses?
6:52 Speaker 00 Two great people, but totally different. Marc Andreessen is literally, I think, the smartest person I've ever met. It took all of my mental energy just to keep up with what he was explaining to me. And I was so impressed because he was really the epicenter in those years. People don't realize how incredibly valuable this was. He was meeting with the engineering teams every day. He was meeting with prospective customers every day. He was meeting with startups every day. He got to assimilate all this knowledge about where the industry was really going. And I think it helped him to see the future way before the rest of us. I learned way more from him and actually from Ben Horowitz, who was a peer. I learned more from them than they would ever realize.
7:37 Harry Stebbings? Are there one or two really big things that have really shaped how you think and operate today from Mark and Batten?
7:43 Speaker 00 Ironically, the lesson that I took away the most, and I remember him saying this over and over, was the most important thing is to know what you can't know. Isn't that essentially a lesson in humility? It's like, you need to know what you can't know. Because that is really the essence of product discovery. We can't know. We can't know if our customers are really going to buy this. We can't know if, how hard it's going to be to build until we actually go figure it out. Honestly, that has had a huge impact for me. I genuinely believe it made me a much better product person when I realized he was right. Because I had grown up in the era where product people were expected to have the answer. That's not only not right, but it misses such an important point.
8:28 Speaker 00 Ben is a different character. He's one of the best product minds I'd ever seen. Ben, I'd say, one of the things I really got from him was this sort of tough love. Give it straight, but give it from the right place. He, of course, was lucky enough to be coached by Bill Campbell and you could really see the impact of that. You could see him coming through. So anyway, those were great people. Pierre was really a different kind of person, but I loved what he's trying to do for the world. You meet startup founders all the time and some of them just have that burning fire that's so addictive. That's my favorite thing about working with startups. It's just contagious. And he had that. He saw that the internet would enable something
9:11 Speaker 00 that's never been possible before. And I don't usually talk about this, but the truth is, in my whole career, I've always focused, my interest is really on product teams, helping product teams, mostly engineers. But I made an exception with eBay. It was the one time in my career, because the customers were not developers, the customers were buyers and sellers. And the reason I made that exception is because of Pierre, it was such a persuasive vision. After eBay, I wanted to get back to, at that point, I was pretty tired too, several big rides. And I wanted to get back to what I really loved, which was working with startups. I started SVPG as just a way to work with a bunch of startups. My timing
9:51 Speaker 00 was incredibly lucky, because if you remember back then, we're talking like 2000, 2002, that sort of timeframe. A lot of the early startups were getting going and a lot of the people either came from Netscape, or they came from eBay, or they came from PayPal. And so I knew a lot of these people. I knew the early Google. I knew early Amazon. I knew these people were calling me up and saying, hey, the stuff that you used to share with us at Netscape and at eBay, can you share that with our team? So it was a perfect timing, really.
10:23 Harry Stebbings? You said there about loving working with startups and that just giving you the energy that drives you. I want to start on one aspect that I think is really interesting. And as you said, in particular, product leaders are focused on secondary risk and not primary risk. And so I wanted to start on nomenclature. What is a secondary risk and what is a primary risk?
10:43 Speaker 00 And I'm really talking about product slash company founders there, startup founders. It's usually one of the founders that is the product leader. And of course, there are all kinds Of risks when you do a startup. One of the tools I like actually is goes by various names, but a business model canvas, a startup canvas, a lean canvas. The reason they're nice tools is because they make you face your risks. They spell it out. There's all kinds of risks. There's how are you going to make money? There's, do you have a effective go to market channel? Do you have a unique value proposition? But what I find is that most founders gravitate to the risks they're more comfortable with. So for example, if the founders are like fresh out of business school,
11:29 Speaker 00 they will gravitate to the business model. They're super excited about how they're going to monetize. And they're already excited about their spreadsheets and how much money they're going to make. And I'm like, okay, yes, that's a risk. But you know what? It's not the primary risk. My point to them always is, look, all this other stuff is only relevant if you can come up with a product that people are willing to buy. Full stop. Everybody acknowledges that, but too many of them think of it as like an exercise left for the reader. Or worse than that, they think they know. They have this idea. They think they know what that is. In fact, I just a couple of days ago, I did a call with a founder.
12:12 Speaker 00 And by the way, I think this founder is quite a smart person coming from the right place. But he had, out of frustration, asked for a little time because he had been pounding away for three years and they couldn't get the product market fit. So we had this discussion. He showed me the stuff and literally they had not changed their MVP in a significant way in three years. He figured, I know this is a great idea. There must be a market somewhere. They were trying different pricing strategies. They were trying different like premium strategies. And I'm like, look, dude, the problem is you don't have a product. Your product is not valuable. That's the issue.
12:52 Speaker 00 They had basically been dancing around every other risk and not tackling this head on, which is unless the product is right, you've got nothing. It's worth double clicking on one part of that, which is when we talk about getting the product right, there's really two sides to that. The first side is, does anybody have this problem? Does anybody need this thing? It turns out, though, that's the easy one. People spend all this time on that one, again, because it's easier. But they spend all this time. That's not hard, honestly. We have very good tools. Most of the time, there's no doubt. All you have to do is look at the competitive marketplace and see how much money people are paying for solutions in that space. Most of the time, it's not rocket science.
13:35 Speaker 00 But if it is a little different, we have very good tools to test demand. The issue is rarely demand. The issue normally is, the solution that you've come up with is not significantly better than the alternatives that are out there. And if it's not significantly better than the alternatives, people won't switch to your product. And then, none of the other risks, none of the other cool marketing strategies, none of the other cool sales models, none of the other cool monetization, none of these things matter, because people don't really want your solution.
14:08 Harry Stebbings? You have to speak to customers, and you have to put yourself in front of them. When you're going through the product discovery process, what are the critical questions that we need to answer in determining whether our solution is right with the customer?
14:20 Speaker 00 I do not know how to do product without intense immersion with customers. Everybody says that today, but usually when I describe what that really means, they're going, "Oh, wow, I need a lot more time for that." To your question, and you'll see why you need to go to customers. In product discovery, there's really four questions that we always have to consider. Now, sometimes it's easy. Any one of these questions might be easy, some are hard, you never know. But we always have these four questions. The first one, is it valuable? In other words, will people buy it? Or if they've already bought it, or if it's a free service, will they choose to use it? That's the value question. Usually, that's the hardest one.
15:01 Speaker 00 Second, assuming they want to use this because it does something that they need or they want, can they figure out how to use it? That's the usability question. There's very rarely an Excuse to do bad on that one unless you just don't realize you need product designers. So normally, if you have the right skills in the startup, that one we can address if you just do the basics. The third, is it feasible? Now, this can be a very big one, especially with things like machine learning technologies. Feasibility means, do we know how to actually build this solution? Do we have the skills on the team to build this solution? Do we have the technology stack available in order to build this solution? Do we have enough time? All that is feasibility.
15:46 Speaker 00 And the fourth risk is, can this sustain a business? This is viability risk. In other words, is it legal to actually do this? Can we afford to pay for it? Can we actually get this to market? Do we have the marketing mechanisms, the sales mechanisms? Do we have the channels set up so that we can effectively get this to market? These are all viability questions. A product to succeed needs to be valuable, usable, feasible, and viable. Any one of those, of course, can kill a product if they're not.
16:20 Harry Stebbings? On the valuable side, I think the biggest problem that founders often make is they're building features, not products. I see this a lot in people management solutions. And it really puts into question whether it's valuable enough. Are you going to switch off something for another feature? I constantly question my own definition of features versus product. What is the difference between a feature versus a product? And how should I think about that?
16:42 Speaker 00 At one level, it's very superficial, the difference between a feature and project. And a lot of times, we'll just have a definition around the level of effort. So that's not really helpful. The way I try to get people to think about it, it's the difference between output and outcomes. So the difference is, features are just output. We can crank out features faster than we've ever been able to crank out features. So if you notice, though, the percentage of companies or products that do better is not really going up. So it's not about cranking out features. We call those feature factories. It's just, that's all they do is optimize for features. What it's really about is solving problems for customers and for your company. And that's what we mean by outcomes. So when we talk about a product... We mean solving problems for real people.
17:25 Speaker 00 And again, could be meeting a need like in the entertainment space or the gaming space. But you're doing something valuable as opposed to just a bunch of features.
17:35 Harry Stebbings? When we look at those four, you could butcher the process or not do a good process through many different ways. What makes a really great execution of product discovery versus average?
17:47 Speaker 00 Usually the clock is ticking. There's a certain amount of runway there. That is true that some companies have had much longer runways in the last few years than I'm used to. And I don't necessarily think that's a good thing because people did a lot of really dumb things because they had so much pushing off the getting to product market fit. Because they could buy customers, things like that. But the clock is ticking for a startup. So the difference between a good company, a bad company, you don't really need to use product discovery techniques. If you have essentially unlimited time and money, you could do, we used to call this the old Microsoft model. And to their credit, they would pull it off.
18:26 Speaker 00 You just assume you're going to spend three, four, five major iterations and you're taking about a year for each iteration. So that's why it was normal for Microsoft. Their first couple of years were terrible versions of the product. But by the third year, fourth year, it was starting to get pretty good. Most of us do not have monopoly profits to fund that stuff. We have to be smarter. And that's really the difference between good discovery. Because even what Microsoft did, that's product discovery. It's just very slow, very inefficient. But that's what it is. You're trying out an idea, you're seeing if it's valuable, usable, feasible, viable. If it's not, you take another shot at it. In teams that really have constrained resources, by the way, this is not just startups.
19:11 Speaker 00 Because even in a big company, it's not like they're running out of money, but they do run out of management patience. We like startups and I tell them, look, here's the reality. You will probably need to do between 50 and 100 attempts before you get to product market fit. So you either need a boatload of money and time, or you better learn some skills. It's not that hard to learn the skills. We have great techniques today. This is the part I love about product. The techniques have got so much better over the years. So we can do 50 to 100 attempts at getting the product market fit in months instead of literally decades.
19:51 Harry Stebbings? That's an interesting one because I often think about how long is the right amount of time to get enough data? You need to put it out in market, get determining customer feedback. How long is enough time to know when the writing is on the wall? You don't want to pull it back too quickly and you don't want to leave it out too long.
20:07 Speaker 00 This is where the skills really make a difference. And also, first of all, I should say, you just tapped into a very big topic in product. So we could talk literally all day about this. But the truth is there are quantitative and qualitative techniques. The quantitative techniques are about collecting the data. That's actually the definitive set of techniques. That will tell us for sure if something is working or not. But the qualitative techniques, first of all, they're much, much faster. And second of all, they do something very different. They tell us why our product is not working. The quantitative techniques can tell us it's not working, but that doesn't really help us that much. It just says didn't do what we need. Now what we need to do is figure out why and the qualitative techniques do that.
20:51 Speaker 00 And good product teams are doing a combination of both all the time. So the short answer is most of the time we can figure out in hours that this is not right. In fact, it's not unusual to do a dozen iterations in a day. Now, when you're doing that, you're not putting it out into the market with 100,000 people using it. That's not what you're doing. But you are putting it in front of real users and customers and trying to understand two things. Number one, could they use it? Could they figure it out? Could they do what they need to do? And most importantly, number two, would they use it? Would they really buy this? And of course, we have a whole bunch of techniques. For the first, that's pretty straightforward. We've been doing that for years. That's called a usability test.
21:35 Speaker 00 For the second, a value test, that's more involved because there are lots of different ways to measure value. It depends if we're talking B2B. If we're talking a consumer product, there's a lot of different factors there and a lot of different techniques because of that. But the bottom line is, as soon as you're convinced as a product team that this is not there yet, you iterate. You don't wait a month for some statistically significant results and then a report.
22:02 Harry Stebbings? On the qualitative that I'm just stuck on, I speak to a lot of founders and like, we spoke to customers and they told us they would buy it, but they haven't bought it. And I'm like, how did you ask the question?
22:12 Speaker 00 I know, this is product 101. And also, to be honest, we don't really care what people say. You have to realize the reason we don't care is because they are customers as much as we love them. They don't know what's possible. They don't know what's possible. So, no, we're not. That's why there was a great quote from Steve Jobs. He would hold up his iPhone and say, "You can conduct 100 focus groups, you are never gonna get an iPhone."
22:35 Harry Stebbings? So then does qualitative feedback even matter?
22:38 Speaker 00 Oh yeah, because that's not what we're doing. We're not asking them, "What would you like in the product?" What would people that don't know what they're doing, they'll often ask things like, "So pretend you're the designer, what would you change?" No, that's totally missing it. What you can do, in fact, there's a great Elon Musk. Of course, he's very good at product and his argument is with qualitative feedback, your goal is to find all the things that are wrong with the product. So that's the difference. You're not looking for affirmation, which is what these naive founders are doing. You're not doing that. You're actually the opposite. You're going after all the reasons they would not use it. That's much more helpful to us. What are the reasons they wouldn't use it?
23:20 Speaker 00 Now, that doesn't necessarily tell us what to do. We have to simulate all this together, that what we're seeing, and then we put together a model and if we keep working at it, there's no guarantees in this, you could never get there. Some products are never gonna happen, but the idea is to dramatically accelerate the time to get to the point where we actually have something that resonates.
23:41 Harry Stebbings? How many do you need to hear that from? And what I mean by that is, there are some products where I say, "Oh, I need this for it to be useful." And the truth is, I'm a super user or a very different type of user, as some cohorts of users are. How do you have enough data to know that's a broad enough swathe that we should listen versus it's an isolated case?
24:02 Speaker 00 The general heuristic here is don't go there. In other words, when we do qualitative feedback, we're not looking for a number. When we do quantitative feedback, we are. So the quantitative, that matters. And there's a function of how much time are you willing to wait to get the degree of confidence you need to be able to either have evidence or statistically significant proof? That's where we go, but that's different. We're talking qualitative. We're not like saying, oh, we got six people that didn't have an issue. We're good to go. That's not what we do. What we're looking for with every user is what are the reasons that they would not use it? Now, we're also crafting a value test to see if they're serious about it.
24:44 Speaker 00 Because most people, especially when you sit down with them and they're not going to tell you, the joke is they're not going to tell you your baby is ugly to your face. They're going to say, what a cute baby. But the truth is, they're like, yeah, I'm never going to use anything like that. How do you get the truth out of them? There's a lot of ways we have. One way is to see if they'll actually pay for it. Literally, see if they'll pay for it. Pull out a credit card. You've probably heard of this idea of a non-binding letter of intent to buy. That's one of the techniques. To pull out a payment card is another technique. To be able to commit to time is another technique. To pay with your reputation is another technique. Book Inspired, I wrote that book a few years ago.
25:24 Speaker 00 Inspired is meant to describe the major techniques that teams need to do good product work. There's a lot more techniques out there, but those are the ones that everybody needs. And it describes several of these tests you can do. Another test of doing to measure this is the Shawn Ellis test, which I'm sure you've heard of.
25:43 Harry Stebbings? When we compare quantitative and qualitative, there comes this question of is product today more art or science? We've never had more metrics. We've never had more tooling. Have we lost the art? How do we think about that?
25:53 Speaker 00 I would say product development today is a great blend of the two. What I just described, the qualitative with the quantitative, that is a beautiful blend of the two. If you force me to pick one, I would definitely say it's more art than science. Part of the art is deciding when to count on the science, when to use the science to win. And to really do that and really know, look, this is very risky. This is very important. I don't want to commit the resources to building this until and unless we have statistically significant evidence, this is going to succeed. That's a very reasonable statement. I work with teams where that is exactly what you want to do. For other teams, it's, we don't need that. We don't need that. We just need some confidence that we're not going in the wrong direction.
26:38 Speaker 00 That's all we need. We can get that faster. This is the kind of judgment that product teams are making every day. And I would argue that judgment is much more art than science.
26:48 Harry Stebbings? I loved it. Scott Balsky actually said on the show, intuition is what takes you to the mountain. Data is what takes you up it. I listen, it was a brilliant saying. And I do want to ask, I think in terms of building the teams themselves, you said there, when we said about the primary risk, this is for founders. At some point, the founder has to start hiring the team. And I think there's just a lot of confusion around when is the right time to hire your first product people? How do you advise founders? And how would you advise me on when is the right time to hire your first product people?
27:17 Speaker 00 There's different schools of thought on this. I'm a big believer in the founder as the product leader. I personally never invest in a startup unless at least one of the founders is a proven product leader. Now, of course, the equation, the calculus is a little different if neither or none of the founders are a product leader. First of all, I don't have a lot of confidence in that company. But second, you're going to have to bring products expertise on much, much sooner. But if we can assume in this discussion, we have at least one founder that's a proven product leader, then it actually becomes quite clear. And honestly, it's a function of the number of engineers more than anything. That's the heuristic I use.
27:57 Speaker 00 And that's because at a certain point, there is no way for that founder to be close enough to all of the engineers to be able to make the daily decisions on the product that founder needs to make. Partly, this depends on just how many hours is this founder willing to invest in this stuff? But realistically, once you get past about 25, 30 engineers, the teams are saying, I can't get enough time with the founder. Just can't get enough time. We're going too slow. We're waiting. We're making bad choices or no choices, which is worse.
28:29 Harry Stebbings? That seems like an awful lot of engineers, twenty five before you make that higher.
28:33 Speaker 00 Realize it's not like twenty five flat engineers were all are considered equal. What they are, they're naturally going to group around product teams. If you have twenty five engineers, you've probably got a minimum of three, probably four, five product teams. Each one of those will have a very senior engineer. We usually call them a tech lead, but whatever they're called. It's those people that have to have that very close relationship with the founder. What is that? Five, four, five? That's not too many for them to do. And then that tech lead, of course, is talking in detail with all the other engineers about how they're going to divide and conquer on the work.
29:13 Harry Stebbings? Marty, I need your help on this. You are my angel investor in this hypothetical situation. Very thrilled, by the way, you were invested in my company. We've discovered now twenty five hundred is that kind of tipping point. I've never hired this product person that we need now before. How do I structure the hiring process for them? How many meetings do we have? What do I want to achieve in each meeting? Can you help me literally from scratch?
29:35 Speaker 00 Probably the most important, actually. Yeah. And I have this chat with a lot of founders like you about, let's talk about what you really need. What kind of person are you looking for? Is it literally just your capacity is out? Or maybe you're much more the visionary and you need somebody that knows how to make this machine work a lot faster. These are some pretty different responsibilities that you're talking about. We want to be very clear on the kind of product person that you're really looking for. And then we can get to work. Yes, for sure. Most companies, in my opinion, don't do a very good job on the whole hiring process. They're just very random. Another big thing we need to talk about is what is your philosophy on this?
30:17 Speaker 00 Would you rather hire, say, university graduates and develop them into the kind of people you need? Or do you say, no, I want people that are been there, done that, and I'm willing to pay a lot more, but I'm paying for that experience? We have to have that. Discussion. Both of those can work.
30:34 Harry Stebbings? I find it really hard. I think if you're a really early stage startup, the best people are not going to leave the incumbents or the fastest growing companies. Your job as an early stage company is to find the diamonds in the rough before they're diamonds.
30:46 Speaker 00 Don't get me wrong. A lot of people have no choice but to do that. I also believe this is what's great about startups. The biggest reason somebody great would join your company is because of your product vision. You are trying to do something that inspires them. Look at all the people that go work for Tesla or SpaceX despite crazy everything because they are so passionate about what those companies are trying to do. This is our number one tool for attracting strong product people is having a very inspiring and compelling vision. Now, sometimes the founders need a little help articulating that. They really can't get it out clearly in a way that can sell people on an interview, the right kind of people you want to have.
31:32 Speaker 00 Here's the thing, though, on hiring that's most important. If you decide, for whatever reason, you're going to hire junior people, then the worst thing you can do is do that and not have somebody in place, best sense manager, to develop them. If you just hire junior people and they have no coaching on an ongoing basis, you're just throwing good money after bad. That's exactly what happens in so many companies. So the prerequisite for hiring junior people, and I want to be clear, I love this model. And this is the most scalable model I know. Look at Google. Look at Apple. They do a ton of this. But they have the mechanism in place. Microsoft too.
32:13 Speaker 00 The mechanism in place, their managers know their number one responsibility is to coach and develop these people.
32:20 Harry Stebbings? And let's run with this, though. We've gone for that model. How do I know the best ones? And how do I structure an interview process to determine the quality of these young, but potentially brilliant candidates?
32:32 Speaker 00 Yeah, and this is where it really is different. If you're hiring experienced people, that's one interview. If you're hiring, by definition, inexperienced people, that's a Very different one. You're looking and obviously when you're looking for experience, you're looking for whether they can prove to you they've learned about the business, they've learned about customers, they understand the techniques. To me, that is much more straightforward. When you were hiring, instead of experience, you're hiring on potential, which is what we're talking about, right? Based on potential. Then you're making a bet on really more innate traits. And then you're looking for things like, is this person got a strong work ethic? Does this person have a strong sense of curiosity?
33:14 Speaker 00 Is this person that gives up very easily when they run into roadblocks? There's a lot of ways to test. I'm not a fan, by the way, of tests in terms of people will have these assignments that they're doing. I'm not really a fan of that, but how do you?
33:29 Harry Stebbings? Why not out of interest? Cause that's like a tangible thing that people like to do.
33:33 Speaker 00 I know. They love simple tangible things, don't they? And there's all kinds of examples of people failing tests that were just the dumbest. In my experience, they're not very indicative of what you're really looking for. And let me put it this way, I still haven't found that test that I think is any good. What I do think though, is there are traits that we're really looking for and we have good ways of drilling into that. It is true though, that the risk for an inexperienced hire is higher. And so I always tell that, first of all, I tell the founder, please don't do that unless you have like a director in place who knows it's their job to coach and develop those people. Don't do it. Otherwise you're just setting them up to fail.
34:18 Speaker 00 But if you've got that person, we've got that strong director level person there that can do this. Then I tell them that they basically have three months to get those people to competence. That's what it usually takes is about three months to get them to competence.
34:35 Harry Stebbings? Before we get to those three months, sorry, I'm interrupting you, but I'm.
34:38 Speaker 00 No worries.
34:39 Harry Stebbings? I had David Lieb on the show, former product lead for Google Photos. And he said, actually, he had better experience with people who had never done that domain before.
34:48 Speaker 00 He's absolutely right, but that's a different question. This is a little bit of a sidetrack. But it's a really important one. There has always been a debate in the product world about do you hire people from the domain? Or do you hire product people based on product potential? And expect that they can learn pretty much any domain. That's the debate. I am squarely on the side that I think he is. My evidence for that is just look across our industry, across time, that all the innovative products, almost all of them came from people and teams that had no real domain experience.
35:27 Speaker 00 The most recent example that just I think is hilarious is look at the frenzy around buy now, pay later. Did you notice that none of the companies that came up with buy now, pay later are the credit card companies? That were doing this every single day for the last 30 years? Did you notice that? Why is that? You probably know Srijast Doshi. He's one of my favorites.
35:49 Harry Stebbings? Yeah, I've known him very well. I've had him on the show. I love Srijast.
35:52 Speaker 00 One of my favorite product people. I met him way back when he was an engineer. The way he phrased this, I give him credit. I love it. I've shared it with lots of people. But one of the things that's critical about product is to be able to understand the difference between domain expertise and domain dogma. And it is very hard when you grow up in a domain to be able to know the difference. You're too close that way. That's my theory that explains why is it that so often the innovations come from people outside of that domain? So now it doesn't mean that you don't go deep with customers. You don't go deep with understanding the financial constraints, the regulations you need to, but you need to bring, the way I usually describe it is you don't bring along all that baggage.
36:36 Speaker 00 I'm a big fan of hiring for product people and I expect them to be able to learn the domain in part of those three months. That's part of the three months. I know I've had to do that. Now, to be honest, most of the time that's pretty easy. There are some domains, like medicine, for example, that's got more to it. But even in those companies, I would argue, because those companies always have, it's sometimes called chief. Domain expert, sometimes it's called chief subject matter expert. But say, for example, they have a physician and every product team can go talk to that physician about like, how does this really work? We're designing, say, a surgical instrument. How does this really work? And they are able to get that extra level of depth.
37:21 Speaker 00 Even in finance, Intuit has these experts around tax. Do you really want to hire product people that are tax experts? Of course not. But do they have to understand personal finance? Do they have to understand the basics of account? Of course they do. I am a big believer. It's not about domain expertise. I say that always because we all know, if you've been doing this long enough, we know exceptions. I know some people that came from the domain that are rock stars. It's just that, for the most part, I would rather have somebody with good product skills that I know can learn what they need to quickly.
37:59 Harry Stebbings? Learn what they need to quickly. Love hearing this, but learn what they need to quickly. Three months. How should the manager structure those three months? And what is the right way to onboard this new person to make sure that they're adequately supported in those three months?
38:13 Speaker 00 Recently, I started answering that question by telling what happened to me when I was on boarded as a product person. And I told it once and people seemed to say, oh, that gives me hope. It was harder for me to get to that competence level than I think most. So my situation was I was an engineer for several years and I wanted to learn product. And my manager, who was an engineering VP, said, I can't. I don't know it, but I can introduce you to somebody who could coach you. And HP was really good about that. So they introduced me to a guy. It was remote. He was in Colorado. I was in Palo Alto. He had started four successful product divisions at HP. He agreed to coach me and the first thing he did, and by the way, I just copied what I learned from him for years. So the first thing he did was an assessment.
38:57 Speaker 00 And I have a tool to assess people. So you start with an assessment because everybody's different, right? You're different than I am and we're all different. I had been an engineer for several years, so the technology side was like my one advantage. Funny enough, I thought my big advantage Was going to be since the product I was building was tools for other engineers. I'm like, I know our customers. And he said, the only thing I know is that is never true. You are never your customer. He was so bothered by my confidence about that, that he said explicitly, I was not allowed to make a single decision for my product team until after I visited 30 customers. And further, he wanted me to visit 15 in the US and 15 in Europe.
39:42 Speaker 00 He made a call to the sales organization. It was a three week business trip. And within about two days, I realized how right he was. I didn't know anything about our customers. I thought that engineers are engineers. And I thought they're how different could they possibly be from us in Palo Alto, sitting on these very expensive workstations. Turns out I was way off. And I went to places like Walmart and FedEx and British Telecom. I learned, no, it's not the same. So first of all, I learned I didn't know our customers at all. But he was smart enough to know that was an area I needed to learn. Another thing he asked me about was, do you know HP's go to market path? And I didn't even know what that really was.
40:27 Speaker 00 And so he said, okay, this is also extremely important. On those same sales visits, he's going to make sure that I learn. HP basically had two major channels, a direct sales channel and a reseller channel. And he wanted me to learn both. He wanted me to learn how it worked. He wanted me to learn what they're capable of and what they're not capable of. And I learned that and I it was all new, completely new to me. The next area he assessed, and this was pretty embarrassing, actually, but I studied computer science. I had zero classes in finance. I regretted as soon as I learned this stuff, I regretted that was the case. And he asked me a few questions about the analytics around my product. And I remember the only one I could get right was he said, do you know what LTV stood for?
41:11 Speaker 00 And I somehow knew that stood for lifetime value. And then he asked me, but do you know how that's calculated? And I'm like, I have absolutely no idea. So he realized that I I was totally incompetent around the financial side, which is pretty important when you're responsible for a product. So he said, I'm going to do two things. First of all, I'm going to send you a book, Basic Finance for Business, or something like that. It was not fun to read, but it was straightforward and every page was new for me. The other thing he did that was incredibly helpful was he called a friend of his in the finance organization of my division, and he asked him to tutor me. And once a week, he sat down with me and he said, "Look, you need to understand,
41:56 Speaker 00 if you're going to be product manager for this product, you need to understand every one of the KPIs that's used to judge your product. And more than that, you also have to know which ones are actually in good shape and which ones are in bad shape." I know that the most embarrassing part was, I didn't even know our product was losing money. I was an engineer on the team. Nobody told us this stuff. And I remember going to the rest of the engineers and saying, "Do you know that we're actually losing money?" We were so clueless. But he not only explained to me the meaning of each KPI, but he explained to me, like, just so you know, "This is really bad. This number is not sustainable. If you don't find a way to fix this, you know, the leaders are going to kill this product." And then other ones were like, "This is above average. This is good."
42:41 Speaker 00 So I didn't have any of that subjective sense.
42:44 Harry Stebbings? The thing I worry about is with the idolization of product people and engineers today, because the best are in such short supply, I almost find leaders cosset them. "You just do what you're best at, and we'll make everything around you perfect because you're the special ones." They're not forced to read accounting books. They're not forced to study go-to-market variance. And they're just, they're the special ones. Don't bother the engineers and the special product people. I worry about that.
43:11 Speaker 00 It's one of the most obvious differences between great product companies and the rest, how they think of their engineers. We say if you're just using your engineers to code, you're only getting about half their value. Bill Campbell used to say this, "There is nothing more important than empowered engineers." Tesla, SpaceX, Google.
43:32 Speaker 00 Amazon, Stripe, Apple. That's the foundation of every one of those companies. Empowered engineers. And what really gets me, look at all those companies out there that outsource their engineering. Literally outsource their engineering. When I find a founder that thinks that engineers are somebody you could just do a contract with Accenture or somebody, I'm like, you're not serious about this. Come back if you decide to get serious.
43:59 Harry Stebbings? What's the thing that's moronic is that, oh, it's just for the short term and then we're going to hire in. What great engineers, like, oh, great, I just inherited some substandard stack that some outsourced team just did. So I agree there, but the question is, okay, we love that. We have the go to market knowledge needed. We need the customer base needed. We need the finance side of the business, what drives the business. How do we give them that? Do we put them a week in sales, a week in?
44:22 Speaker 00 My view is you do this assessment, which honestly takes a few hours to do it. The assessment is easy. Then you need to do a coaching plan. Every manager that manages individual contributors is responsible for a coaching plan for their people. At the beginning, that coaching plan is just to get the person to competence. Getting them to the level where they can do their job their team needs them to do. After that, it gets more fun. It's more about helping them reach their potential. In fact, I was taught at that same HP engineering program that my job was to get people promoted, to get them ready to earn a promotion. That's how I was judged as an engineering leader. So that's the coaching plan side. Now, these coaching plans are different for everybody. Sometimes they literally need to spend some time in the field.
45:07 Speaker 00 Sometimes what they really need to do is spend some time on the phones in customer success. It just depends on that person. I am also a big fan of onboarding programs, though, where whole cohorts of new employees go through training programs. These are less personalized, but it's more like, look, I don't care how good you are. Everybody that joins this company, you're not going to know our customers. You're not going to know our culture. You're not going to know our leaders. You're not going to know how decisions are made here. So an onboarding program is a way to accelerate this for a whole. Whole cohort of people.
45:44 Harry Stebbings? What do you think people most do wrong in the onboarding of new talent? Where are you like, oh God, they made this mistake again?
45:51 Speaker 00 The biggest thing I see is that they assume I'm just gonna hire somebody, been there, done that. So they get somebody from Amazon, they get somebody from Stripe, somebody that's clearly experienced what good looks like. But then they assume that just because they were successful there, that they're gonna be successful in our company. And that's not a true statement. There are so many differences. So our job in onboarding is to close that gap. Our job more generally in coaching is to close that gap.
46:20 Harry Stebbings? I do want to ask one final thing before we do a quick fire. It's often I see tension between products and other areas of the org, specifically sales. Oh, we couldn't sell because we didn't have that product capability. And product go, it was not our fault you couldn't sell, it's your fault. How do you think about creating alignment between products and sales? Which is normally a fictitious relationship.
46:40 Speaker 00 Would you rather be in sales, being forced to sell this piece of crap product? Or would you rather be in product where you can actually make it a good product in order to sell? The bottom line is the responsibility for the sales is on the product team. And in fact, if you really care about something like the real meaning of OKRs, which is all about outcomes. It is designed that way intentionally. This is Andy Grove. He intentionally so that if my product is not selling, I need to get my ass out of the office, into the sales organization, sit with them to figure out why it's not selling. Is it because the customers are not seeing it?
47:21 Speaker 00 Is it because the sales force has no idea what this product is really doing or how it does? Is it because they don't have the sales tools that they need? Is it because there's no reference customers? The point is, this is what we are supposed to do on a product team. And explicitly, this is on the product manager.
47:40 Harry Stebbings? But you mentioned Surya Astori earlier. He says about the three types of product leaders, the visionary, the craftsman and the operator. I see a huge amount of operator product leaders who honestly just install red tape. They lack creativity. And they're just real process people that hold things up and they come in at series C and they just churn two years at a stellar company, two years at another stellar and they're just not very good. Do you agree with my assessment there of this rise of this kind of operative product?
48:08 Speaker 00 I've also heard Trias describe this, which I absolutely love, but there's a whole dimension to this. In fact, I'm gonna write an article and I'm gonna ask him to review it that he doesn't talk about, but it gets to what you're describing and I think this is huge because in general, those three buckets are ripe. There's a whole dimension though, which is there are two kinds of operational people. There are those that believe in operation or scale, really what this is usually about with process and those that believe in scale with coaching. The problematic behaviors that you're describing, which by the way, you are spot on. And I hate to say this because I absolutely love Europe, but it's a bigger problem in Europe than it is in the US.
48:51 Speaker 00 The reason is because so many of those people are process people and they don't understand. And I've been sharing all these quotes from people they respect, like Elon Musk says in big companies, process becomes a substitute for thinking or Bezos describing, be very careful because process ends up controlling everything. Steve Blank talking about process people taking over. Steve Jobs talking about process people taking over and no good product people want to work there anymore. This is such a serious disease and it has killed so many companies, but not just in Europe, all over. But this is for some reason, process is especially attractive to Europeans.
49:35 Speaker 00 You might have a better theory for why than I have.
49:38 Harry Stebbings? I think it's because we lack self-confidence more actually. And I think process gives you a structure with which you can point to and it gives you something to rely on other than pure intuition and gut.
49:49 Speaker 00 I see some companies like Spotify has done fabulous and they are much more set the dial like we would in Silicon Valley for sure.
49:57 Harry Stebbings? I just had Gustav, the CPO and CTO on the show and he actually said the most fascinating thing. He said, talk Is cheap. And I said, yeah, I agree. And he goes, so we should do more of it. We should have a lot of debate all the time. And I thought that was so interesting in product. Cause I was like, no, like it's about speed of execution. It's about iterate, iterate, iterate, like go. And he's like, no, talk is cheap. Let's talk lots.
50:19 Speaker 00 You've probably also heard the think deeply to move quickly. Same idea. So for example, the written narrative is a way to force you to really think things through. And then you're going to move a lot faster.
50:32 Harry Stebbings? Marty, I could do what you all day. I do want to do a quick fire round. So I say a short statement and then we rock and roll. Sound good?
50:37 Speaker 00 Sure.
50:37 Harry Stebbings? What should you focus on if you want to get promoted as a PM?
50:41 Speaker 00 The right answer, the short answer is outcomes. If you deliver results, you are gonna do really well. That's an outcome. Like I said, literally, it's an outcome, but the best way to get there is to do your homework. So the biggest thing I emphasize to new product managers is do your homework. That means go deep with your customers, go deep with the data, go deep on the industry and go deep on the enabling technology.
51:05 Harry Stebbings? How do you create PMs with founder mindsets?
51:09 Speaker 00 This would have been a great topic in the bulk of our discussion because this is really the difference between an empowered product team and a feature team. In an empowered team, you are literally giving ownership responsibility to the team. And that they build those skills, they build that confidence, they build those muscles. In a feature team, they're literally just mercenaries doing what they're told. The way you do it is you empower them. You say, look, you own this. This is why on a product team, on an empowered product team, product manager is responsible for the value and the viability of the solution. In a startup, the founder is responsible for the value and viability of the solution. That is not an accident.
51:49 Harry Stebbings? Which product leader do you most admire?
51:52 Speaker 00 Honestly, I get to beat a lot of product leaders. I'll just mention that I came out with a book recently called Empowered and I featured eight product leaders that I really admire. And I highlighted them because very few people know them and that really bothers me. They should be well known. But you know, they all have worked in companies where the founder got glory. But they were awesome.
52:14 Harry Stebbings? Tell me, what one piece of advice would you give to a product leader starting a new role today?
52:18 Speaker 00 I would say the first and foremost thing is you have to establish that foundation of trust with the rest of the leaders of the company. And the best way I know to do that is to immerse with customers and really become first name basis with major customers. Make it so that nobody in the company doubts that you know your customers, what the issues are, what you need to do going forward.
52:43 Harry Stebbings? What would you most like to change about the world of product?
52:46 Speaker 00 There's a lot of things, but if I could, I am not anti agile. I've been a big advocate for agile, but agile has had this unintended consequence because agile is just a delivery process. But if I could remove one thing, it would be that there would be nobody calling themselves product owners. And all these people teaching product owners how to do their job would not be doing it. And they would instead be people that know product would be teaching them the job.
53:14 Harry Stebbings? Final one, what recent company product strategy have you been most impressed by?
53:20 Speaker 00 You mentioned TikTok. TikTok has been nothing short of amazing. I would say what Tesla has pulled off, nothing short of amazing on every level. And their product strategy has been a very visible product strategy. They had to because it tied in so closely with the financing strategy. Those are two very good examples. What Amazon has done with AWS has been brilliant. We keep going. Stripe is just hitting on all cylinders right now. And so.
53:47 Harry Stebbings? I think the amazing thing with Stripe is the new product creation and new product execution. It's not just maintenance of core, but the ancillaries that they have is insane. Marty, I feel like I could talk to you all day, but this was so much fun, Steve. Thank you so much for making the time and it was such a joy.
54:01 Speaker 00 I enjoyed it too. Thanks for inviting me, Harry.
54:06 Harry Stebbings? What an incredible guest. I absolutely love doing that show with Marty. If you'd like to see more from us, you can find us on YouTube by searching for 20VC or you can find us on the site at 20vc.com. But before we leave you today, this episode is brought to you by Mixpanel, offering powerful self-serve product analytics. You know, something we talk about a lot on this show is how startups can build the best possible product. Relying on gut feelings is an expensive way to find product-market fit, and that's a price many VCs aren't willing to pay. That's why Mixpanel will give you $50,000 in credits when you join their startup program. $50,000 in credits. With Mixpanel, startups find product-market fit faster, helping you take your company from MVP to most valuable player.
54:48 Harry Stebbings? Access your real-time insights with the help of pre-built templates and know that at every stage, Mixpanel is helping you build with confidence and curiosity for free. Apply for the startup program today to claim your $50,000 in credits at mixpanel.com/startups with an S. If you don't qualify for the program, Mixpanel has pricing plans for teams of every size, from startups to enterprise. Grow your startup like you've always imagined with Mixpanel. And next, we have Stitch. Stitch is your all-in-one platform for authentication. From a modern reboot of traditional passwords to passwordless options like email magic links, one-time passcodes, biometrics and web three logins, you can quickly and easily integrate Stitch to power your authentication use case.
55:33 Harry Stebbings? Our SDKs allow you to get up and running quickly with the option to use Stitch's pre-built UI components or fully customize the experience to fit your brand's look and feel. Or for maximum flexibility, their direct API gives you the tools you need to build out authentication flows to fit your exact needs. Make user and customer onboarding seamless to boost your conversion and retention rates by making authentication as frictionless as possible. And Stitch have raised over $100 million from the likes of Co2, Thrive, Benchmark and Index. And their customers range from individual developers to the fastest-growing companies in tech like bitcoin.com. And get started today with $1,000 in credits. Sign up at stitch.com
56:15 Harry Stebbings? ...com/20product, that's stitch, S-T-Y-T-C-H.com/20product. And finally, there is a common adage, data is more valuable than oil and gold. This got me thinking about my personal data and what I'm doing to protect myself online. And I thought, well, I wouldn't leave my front door whilst I'm on holiday. So why don't I protect my personal data and privacy whilst I'm online? This is why I decided to team up with the good folks at NordVPN. I downloaded the app, which is super easy to use. And within seconds, I'm completely protected whilst online on six of my devices. It gives me complete peace of mind, knowing that all my personal data, bank details and privacy is completely safe from falling into the wrong hands.
56:58 Harry Stebbings? I can even access streaming services, this is my favorite, in other countries by switching my virtual location. Hello, US Netflix! And it costs the price of a cup of coffee in London every month, a small price to pay for premium cybersecurity. And of course, I have a special discount for my listeners. To grab your exclusive NordVPN deal, head on over to nordvpn.com/20vc to get a huge discount off your NordVPN plan, plus four additional months for free. It's completely risk-free with Nord's 30-day money-back guarantee. As always, I so appreciate all your support and I can't wait to bring you a fantastic episode on leadership and decision-making on Friday.
Previous 20VC: a16z's Martin Casado on How the Venture Model is Broken, Why VCs Should Be Running Wall St, Who Wins and Who Loses in the Next Generation of Venture & Investing Lessons from Marc Andreesen, Ben Horowitz and Chris Dixon Next 20VC: The Story of Turning Around MTV, AOL and Time Warner | How To Be Effective When Making Hard Decisions | Tactics vs Strategy and Why Plans Are BS with Bob Pittman, CEO @ iHeartMedia Page updated September 29, 2026 · transcript and summary are machine-generated
