← All Episodes ← Prev Next →
SEASON 2 - EPISODE 0003 - SEP 21, 2021 REWORK Revisited

Learning from Mistakes is Overrated

with Jason Fried and David Heinemeier Hansson
Listen now

“Fail early and often.” You hear this all the time in the tech start-up world. Failure has long been held up as a badge of honor for new start-ups. This week Jason and David take on this idea and make a pitch for learning from your successes instead of your mistakes.

Show Notes

Transcript

Shaun: (00:00:00) You said you’ve been working out Monday through Wednesday?

Jason: (00:00:02) Monday, Wednesday, Friday mornings. Yeah.

Shaun: (00:00:04) Uh-huh.

Jason: (00:00:04) Do you want me to flex and prove it?

(00:00:07) Broken By Design by Clipart plays.

Shaun: (00:00:09) Welcome to REWORK, a podcast by Basecamp about the better way to work and run your business. I’m your host, Shaun Hildner. This week we’re discussing the essay in REWORK, the book titled, “Learning From Mistakes is Overrated”, which pushes back against this idea that failure is somehow a necessary part of Silicon Valley entrepreneurship, and even considered a badge of honor.

(00:00:29) We’ll also discuss what you can really learn from your mistakes versus what you can learn from your successes. And here as always, to talk about these things are Basecamp co-founders and the authors of REWORK.

(00:00:38) Jason Fried, welcome.

Jason: (00:00:41) Hey, Shaun.

Shaun: (00:00:42) And David Heinemeier Hansson, welcome to you.

David: (00:00:45) Hello.

Shaun: (00:00:46) So in a previous life, I used to have to cover a lot of those tech startup accelerator pitch days. You know what I’m talking about? Teams would have something like two months to build a company, and then they’d all get to pitch to potential investors. Well, during that time, I don’t think there was a more prevalent phrase tossed around than, “Fail early and fail often.” How did this become such a thing, that failure is just a built-in part of entrepreneurship,

Jason: (00:01:12) I think it’s more of a Silicon Valley thing. Because the guy who runs the pizza shop on the corner, he doesn’t want to fail early and often. The dry cleaner doesn’t want to fail early and often. Most people who start businesses don’t want to fail, they gotta make it work. It’s not even an option, they got to make it work, they got to figure out how to make it work. So this is more of a Silicon Valley thing, where, if you kind of fail, it’s like, well, it doesn’t matter, because someone’s willing to give you more money again to try another time. And in fact, it seems like in some cases, the more failures or whatever you have, the better your resume, that you’re willing to try new things, and you’ve got a lot of ideas. And I think that it’s celebrated to an unhealthy degree.

(00:01:54) Not that people aren’t going to make mistakes, and things aren’t going to work, because most things don’t. But in most cases, people need to make these things work. And they’re putting their life savings at risk. And they’re doing it for themselves and their family versus spending other people’s money. I think it’s a very different mindset.

David: (00:02:12) I think a lot of that comes from the fact that the Silicon Valley economics is so different. The odds are so different, that they literally need one in a 1,000. One in 10,000, to pay for the entire lot. That you need one breakout success that can pay for everything else. And those breakout successes? They are often outrageous. I mean, if you’d pitched Facebook before Facebook’s… hey, what if you had this service where you just give all your personal information to us. And then we slice it and dice it and we sell it to everyone who wants to target just exactly your demographic, what about that? It would have sounded like a bad idea. I think it’s still a bad idea. But it’s clearly a profitable idea that wasn’t obvious in the same way that perhaps a pizza shop or a project management software company would be.

(00:03:04) And those kinds of odds necessitate that most people will be wasting their time. Because there’s just so few Facebooks and unicorns in this world, compared to how many people have to try to get there. So I think it’s a very virtuous relationship with the celebration of failure in that community. Because they know that the odds are so long that most people will be wasting their time, most of the time. So if we can make that seem normal and good and everything, then maybe people won’t look like, what have I been doing with the last three, four, five years of my life. Why did I bet that slice of my existence on this tiny blue dot, on these ridiculous odds? Well, it’s because failure is good. Okay, great. Let me try again.

Shaun: (00:03:56) And I’m assuming this has not changed in the last 10 years since you wrote this essay?

David: (00:04:00) I don’t think so. I think, if anything, it’s become even more so that the outliers are even more, to some extents profitable, when investors can see that well, you know what, this isn’t just going to be a billion dollar company. It might be a $10 billion company, it might be $100 billion company, it might be a trillion dollar company. How many failures can a trillion dollar return pay for?

Shaun: (00:04:25) Right?

David: (00:04:25) What? Tens of thousands? Hundreds of thousands? Lots and lots, right? So I think, if anything, the dynamics have been pushed even further into direction that, try outrageous things. More outrageous things. Anything that could possibly turn into this $10, $100 trillion business. That’s what’s worth it. Because that’s the thing about investing is it doesn’t really matter how much money you lose if one of them pays for everything else.

(00:04:55) But I do think it matters, obviously, for the people who are volunteering their lives to be one of those shots, because for them, it doesn’t really matter whether their investor gets a Facebook-like return. That doesn’t help them. If you’re one of the 1,000 or 10,000, who is not on that train, what good is it for you that, for the investor, everything got paid back? Well, I suppose you’ve got a salary, while it happened. But I don’t think you’re going to look back upon that time in your life and think that, you know what I just, that was great. Well, maybe some do, but as a broad whole, it seems like an utter waste of human potential to have 999 teams chase these impossible things such that we can have one outlier give us the next Facebook.

Jason: (00:05:45) And I think some people might hear that and think that David’s suggesting that you meet need to be extremely conservative with what you’re doing. I mean, I don’t think that’s the point here. You can take plenty of risks. It’s more about there’s a difference between taking a risk and putting yourself at risk. And if you put yourself at risk in a way where if your idea doesn’t work, the whole thing falls apart, that’s probably too much risk. But you can certainly swing for things. We’ve done a bunch of things that haven’t worked. Not hundreds of things, but we’ve done a few products that haven’t done as well as Basecamp. We’ve taken risks on a variety of different things. But we didn’t set it up in a way where we actually consider it a failure. Backpack is not a failure, it just didn’t do as well as Basecamp. Highrise is not a failure at all, it just didn’t do as well as Basecamp.

(00:06:29) So it’s also a little bit of a matter of how you phrase it. But, of course, going out of business or having something completely crash where you can’t afford to pay the rent, like that’s a different story. But there’s different degrees of risk here. I think that’s it’s important.

(00:06:43) Also, just not framing something that didn’t work as a failure. It’s not a failure, it just didn’t work as well, something else. It’s fine. That’s the thing I also don’t like, is this quick jump to failure. It’s like, if it didn’t work, it’s a failure. No, it’s not, it’s not really a failure at all. It’s just a thing that you tried.

David: (00:06:59) And I think that that’s the insidious nature of this venture capital definition of success being this outsize thing, that they actually want to declare failure on a bunch of things that are not failure in normal land. Hey, here’s a nice $5 million a year business, here’s a nice $10 million a year business, that’s a complete failure to someone doing a Series A round of investment. They don’t get anything back on that. So it’s far better to declare that a failure, to recycle the raw materials, the people, into another attempt at a unicorn. At a billion dollar or trillion dollar business.

(00:07:36) The worst thing in that regard is actually to tie up attention and the raw materials of people into things that just do well, because the game is not about doing well. It’s about the slam dunk, hitting it out at the moon. And again, what a waste. We could have had a bunch of businesses that would have been wonderful, $5, $10, $50 million businesses. And yet they have to be crunched and recycled into the chase for another unicorn.

Shaun: (00:08:07) In this essay, you lay out some of the points of failures, like hiring the right people, pricing correctly, spending less than you’re taking in. In the last decade or so are there any other points of failure we’ve seen in startups.

Jason: (00:08:21) I think there’s been a, I don’t want to use the word failure, because I’m not a big fan of the word, but I think there’s been a lack of imagination in terms of business models. And also to David’s point, sort of a lack of recognition that there’s different scales of businesses that can be wonderful businesses. That you don’t need to keep hitting it out of the park to make it work. You can build a $8 million a year business or a $15 million a year business. And if you can take care of a dozen employees really well and you can take home a good amount of money yourself and support your family. That’s a success story. It’s not a failure, because it didn’t get to $100 million.

(00:08:59) So, it’s, again, I think it’s in the definition. And then, the business model thing, it’s mostly… it’s kind of advertising or subscriptions are the two business models for the most part. And I don’t know, there’s got to be more than just two, there’s just got to be. So I feel like there’s just a lack of imagination there, or also a lack of appreciation for the fact that some people, or more people, I think than are expected, are willing to pay for things. So a lot of advertising based businesses could probably be subscription based businesses, or other kinds of payment setups versus just assuming that advertising is the default mechanism if you’re going for a big huge business or whatever.

(00:09:37) So that’s not the case everywhere. But I do think there’s got to be some more gradient in between the poles here.

David: (00:09:42) And I think one of those examples is, I mean, it’s sort of ironic since we run a subscription-based business, but it used to be that you could buy software, not on a subscription. I mean, that’s almost dying out as an idea, even. That you could pay once and you can have a piece of software and that was good enough and it did what it needed to do, and it ran on your computer, not someone else’s servers such that that model worked. And I think it’s dying out, in part because you know what? Subscription-based businesses are more profitable. So there’s more gold there. But does that mean there couldn’t be an entire other slice of it? Why can’t we just buy one thing and pay for it once and that can work at a certain scale.

(00:10:22) I remember on the Mac, in the early days, it was very common that you had these small software shops. Two people, five people, eight people, they weren’t hundreds and hundreds of people, although some of them were right? Adobe, for example. I don’t, can you still buy an Adobe product for?

Shaun: (00:10:38) I don’t believe, so. I think it’s Creative Cloud only.

David: (00:10:40) Right.

Shaun: (00:10:41) So it’s only subscription.

David: (00:10:41) Which, to me is such a good example, again. I mean, maybe this sounds hypocritical, since we’re selling subscription-based businesses. But I really like Lightroom, right? And I use it about once a month, twice a month, maybe? I never really want any of the updates, per se, right? I’m not looking for more features. I would like to just buy that piece of software and use it right? Why does that have to be a piece of subscription software?

(00:11:05) And I think some of it is because these definitions of success have become so narrow, that these small successes no longer count as such in the realm we’re in right now. I think the other thing that’s interesting is, as we say in the essay, the problem with learning from mistakes is that you might learn what not to do again, which, if everyone to the same lessons away from all these software companies, the 999 that didn’t work to one unicorn. Wow, that would be tragedy. I mean, hopefully some people are going to try those ideas again, because those ideas are actually just fine. They’re not fine for becoming a unicorn. Maybe they were not fine this year, but they’ll be fine next year. I think this is one of the things we’ve seen with HEY.

(00:11:51) So, HEY, I think if we had launched HEY 10 years ago, the same year, we wrote REWORK, HEY would have been a failure. We wouldn’t have been ready. The market wouldn’t have been ready. And if we had tried that, and then taken that lesson away and gone well, you can’t sell email. That’s just not possible. Gmail has it. It’s over. What a tragedy that would have been. Because all we had to do was wait 10 years, and we were ready, the market was ready, boom.

(00:12:19) And I think there’s a lot of things like that. In fact, you could argue that that was sort of the lesson we wrongly learned with Campfire. So Campfire was Slack before Slack about a decade in advance. I think Slack came out in what ‘15, or something, ‘14 or whatever. Campfire came out in 2005. And didn’t really take off. It was fine. I mean, we made our money back. But it didn’t become a slam dunk success even though it was essentially the same thing. The world was just not ready to move to chat-based collaboration. And then it was 10 years later, and we weren’t on that train because, perhaps in part, we weren’t good enough to make Slack. I don’t know. Or perhaps we had seen, this thing didn’t take off for 10 years. Why would it take off now? Well, boom, it just did.

Shaun: (00:13:08) Yeah.

David: (00:13:08) And there’s so many things about business where you think you’ve learned something about the world. And it’s kind of like that book The Halflife of Facts, which, things age quite quickly. And in some ways you are at a disadvantage when you’ve been in business for 20 years, like we have, because we’ve learned so many things. And I guarantee you that half of the things we’ve learned aren’t so.

Jason: (00:13:33) Yeah, and because they’ve worked for us. You just, then you take them for truth, also. You don’t, like one of the one of the hard things to do is to reconsider something that worked. That’s a really hard thing. It’s easy to reconsider something that didn’t work, but to reconsider something that worked is challenging, really hard to do. And that’s, I think what happens when you’re around for a long time. You have those opportunities, but they’re really hard to change your mind about something that actually went well.

Shaun: (00:14:03) Yeah, that’s sort of the meat of this essay, right? It’s not that you can’t learn anything from your mistakes. I believe what you say is that you can learn more from your successes than you can from your failures, right?

Jason: (00:14:14) Yeah, the idea kind of is that, okay, so something didn’t go right. And you analyze it, and you don’t necessarily know what it was, you think you maybe know what it was, and maybe you’re right, maybe you’re wrong, but it’s easy just to kind of, you got to write a story, you got to come up with something. So you kind of do it. And then you think that you took something from that.

(00:14:36) Now, what you don’t know is what to do next. You know not to do that thing, which is, one of a billion things you could do.

(00:14:43) What I think is a little bit nicer about focusing on the things that have worked, although they also don’t necessarily provide perfect lessons, but you might have a better chance at really understanding what to do again, or what to try again versus what not to try again. What not to try again doesn’t really give you a step forward. But what to try again, at least you have something you could try again and see if it works.

(00:15:04) They’re both imperfect. Because like David said, it’s a it’s a timing thing. It’s a million things. Its luck, its place, it’s a million different things, the market. But at least you have somewhere to go, some guide on which direction to try to head again, versus the alternative, where you just know not to take a step forward. But you’ve got 360 degrees, you can walk, which direction do you turn? You really just don’t know.

Shaun: (00:15:30) In the two decades that you’ve been running Basecamp or 37signals before that. Can you point out any mistakes that you have learned from?

Jason: (00:15:39) Um…

Shaun: (00:15:39) David just rolled his eyes at me.

Jason: (00:15:42) Here’s an interesting one, I think. This isn’t necessarily a mistake that we learned from but it was a reconsideration of something. When we launched Backpack, Backpack was our first consumer-facing product, which was… Basecamp was our business product. Backpack eventually became a business product, but initially was like five bucks a month for just direct consumer use. And it’s hard to make those economics work, it’s hard to make that business work. It’s a very different kind of business. And I think for a long time, we shied away from doing anything that was consumer-based because business software was something we knew really well and worked really well and paid really well.

(00:16:16) And then with HEY, we decided to try the consumer side of it again, and the consumer side is doing a lot better than the business side right now. Now, that’s not to say the business side can’t do better and there’s a couple of reasons why we think it could do better and will do better once we build some more stuff into it and talk about it the right way and whatever. But that was sort of revisiting something that we thought we wouldn’t do again. It’s not really a direct answer to your question, but it’s a change of mind.

(00:16:42) And to me, that’s a little bit more interesting than what failures were right or wrong? It’s more like when were you willing to change your mind? Or how often are you willing to change your mind? I think that’s actually the more interesting version of that question.

David: (00:16:54) I think the lessons from failures also ties to the point that Jason often makes about not risking the company on a big call, on a big roll of the dice, where you could really look back and think, oh, I got that one wrong, and it’s gone. The consequences of being wrong on that decision would be so high as to be catastrophic, as to be company ending. Because those are the things I think that really scar themselves into your mind as, “Ooh, ouch.”

(00:17:26) When you think of it in a much smaller way, we make mistakes all the time in terms of designing a feature or planning something in a certain way. And then we make small corrections. I think we have another essay about making lots of small decisions, accepting that a fair number of them are going to be wrong, and then correcting later.

(00:17:45) I think a parallel to this is hiring. We have a very rigorous hiring process and we put a ton of work and diligence and heart into it, sometimes to the point of almost too much. Because it becomes very precious to the point of if someone does not work out, it feels like a catastrophe, when perhaps more healthy approach to it is if you’re going to hire 10 people, do you know what? If six, seven, maybe eight stay? Wow, that’s great.

Shaun: (00:18:17) Yeah.

David: (00:18:18) Which means that four, three, two of them don’t. You price in some ratio of failure from the beginning and you accept that that’s the price for making decisions. And you move forward. And you can do this in all sorts of domains. This is why we’re often quick to make these small decisions because we know we’re gonna be wrong some percentage of the time and it’s not going to matter that much. And we can change our mind and move on from it.

Jason: (00:18:47) I think it’s a language thing to like, the way you just framed it, I wouldn’t consider those two or three people who didn’t quite fit in for their reason, or for our reason to be failures. It didn’t work.

Shaun: (00:18:57) Yeah.

Jason: (00:18:57) And I think language really matters here. Because failure is such a, it’s like you said, it’s such a common thing in our industry, people love to talk about it, but it’s such a harsh word. It’s so polarizing. It’s the worst thing, failure. Versus, like, it didn’t work. Or that person wasn’t the right fit. Or we weren’t the right fit for them. That’s something you can understand that happens in the world and you’re not afraid of it. It’s just something, of course, not everyone’s gonna work out. Everyone knows that. But to say you put in all this effort and you failed at hiring the right person. I don’t know, that kind of makes you nervous about doing anything. Like I don’t want to fail at that. I’d rather just give it a good shot and if it doesn’t work, it doesn’t work.

(00:19:41) So, the words you use and the mindset that you come at this stuff with has a big effect on your on your mood and sort of the self esteem of the organization as well. If the word failure is being thrown around a lot, and if someone doesn’t work, it was a failure, that damages the self esteem of the organization. At some point people become afraid of doing anything because who wants to be labeled a failure.

David: (00:19:59) I think it also ties back to the whole framing of this essay, though, is what should you learn from? So, say you hire someone, it doesn’t work out. You could try to look into that and find some reason and then try to see if you can find some pattern and learn something and then, oh, next time, we have to be sure we don’t do that. And I think that’s a lot more faulty of a process than, do you know what? Here’s six, seven, eight people who worked out really well. What are some of the commonalities that these people share? What would work well, oh, for us, for example, we learned that hiring great writers is a mark of success at Basecamp. We’ve learned that managers of one, people who can set their own direction and follow up on it. Wow, mark of success. Those are the things to focus on.

(00:20:46) If you read anything we have about hiring, it’s usually not like ooh, avoid these people. Right? Because that’s the other fact here is that if we talk about hiring, everyone is flawed. We’re not trying to minimize your flaws here. We’re just trying to find, hey, do you know what are some marks of success such that you could do well here. That’s putting it to a point. Of course, there’s some flaws that are just terminal for the relationship. But I don’t think that those are sort of diamonds in the rough, you really have to unearth versus with the what works well at Basecamp. And that’s also the important point, as Jason says, it’s not just about like, is this person the right fit? No, no, no. Is Basecamp the right fit for this person? Does Basecamp work for this person is just as important as the other way around, if not more important.

(00:21:35) So accepting what kind of organization you are. And then, who are the kind of people, what do they look like, the people who have success here. And are there some things we can extract and try to find more people like that, which is, in many ways is a similar thing to common ways of marketing? Like, Facebook and others, they have all this look-alike matching. You try to find customers who are roughly like the customers you already have, because whatever it is about the customers you already have, well, that was a success. They like your product, they’re paying for your product, if you can find someone who’s liked them, they’re also likely to like your product and pay for your product. So that’s a good virtuous cycle to investigate, hopefully in more productive ways than surveillance marketing. But just the general idea that, figure out who are the customers this is really right for? Who are the employees, this is really right for? What are those right things? And then double down on that.

Shaun: (00:22:29) Yeah. I think to wrap this up, I want to kind of continue that thought experiment that David started earlier, imagining that HEY was your first product? Would it have worked, which you said, probably not. But I kind of want to dive in what about the successes of Basecamp informed the success of HEY?

Jason: (00:22:47) I don’t know if, looking at Basecamp, the product had much to do with that. I think, though, that this fundamental interest we have in novelty and bringing new ideas and building the kinds of things that other people aren’t building. And Basecamp was that kind of thing. Basecamp was not a Microsoft Project look-alike, which of course, we’re throwing way back here because Microsoft Project, I don’t even know, actually, if it exists anymore. But it was the thing at the time.

Shaun: (00:23:17) That was project management.

Jason: (00:23:18) Yeah. And so we built something that had nothing to do with that. We pushed hard against some of the fundamental ways in which people manage projects, which were Gantt charts and reports and whatnot back then, and went in a very, very different direction and created a very strong contrast between what Basecamp was and what Microsoft Project was.

(00:23:36) And then with HEY, I think we did the same thing, fundamentally, the same spirit, which was that HEY does not look like Gmail. Gmail is the predominant, Gmail and Outlook are both the predominant—I mean, basically, that’s the whole market, essentially. And HEY looks nothing and acts nothing and smells nothing like those products. It’s very different.

(00:23:54) And, again, this isn’t about looking at Basecamp. But it’s about our general approach, that if we’re going to put something new into the world, it’s got to be new. It can’t sort of be like everything else. Initially. Of course, as you as you build it out, you begin to flesh it out and flesh it out. And some things start to, there’s some things you may have missed the first time around, whatever. And the same thing is true. We’re thinking right now about building a calendar into HEY. I don’t know when or if we’re gonna do it, but we’re thinking about it and the things we’re thinking about, we’re not just going to build a calendar into HEY. We’re going to bring new ideas and novel approaches to a calendar, because it’s about time and also the world doesn’t need just yet another one that’s sort of like the ones that are already here.

(00:24:37) So I think that’s really the fundamental lesson. The things that we like to do are things that other people wouldn’t do or haven’t done, or we think the world’s sort of missing a perspective and a point of view. I think that’s also true about to start business in general and that we have a different approach. We have an alternative approach. We’re not VC-backed. We’re… all the things we’ve talked about over the years. We want to make sure that there’s another way to do things, and that that way is shared and voiced and demonstrated that it can work.

(00:25:07) So I think those are probably the lessons that we take, though they’re not direct from product to product. But they’re sort of more, I guess, spiritual or something, based on how we like to do work.

David: (00:25:18) I think in the same vein of a spiritual approach to it is having some confidence in your own vision for the product. That we don’t design products by going out to ask a lot of people, when would you like? I mean, we currently have a lot of customers, you could totally go to those customers and say, what other kinds of software would you like us to build?

Shaun: (00:25:36) What do you want in a calendar? Yeah.

David: (00:25:37) Exactly. What do you want in calendar? That would be a very natural thing, that, there’s an entire industry of market research built up behind. And we don’t do that. We build the v1 of what we want. That’s it. After v1, you start listening more, but the v1 is very much what we want.

(00:25:59) And I think that there’s a much greater likelihood that you will produce that sort of weird, different take if you are a little bit wearing blinders. You’re not paying too much or too close attention to what everyone else is doing. You’re not doing a serious deep dive competitive analysis, analyzing all the features that Gmail has, and the percentage of users who use this, that and the other thing. No, the v1 is what we want and we look at what features do we absolutely need, and what features would absolutely delight us. And then, again, as this pattern matching, as we’ve found, there’s a lot of people who are just like us in terms of what they want out of products. They may not know it yet, but when we build something for ourselves, we find usually a ready market for it. That market is of varying sizes. Sometimes we’ve come up with things and they were just too early, like Campfire, 2005. We were like, hey, this really important product to us. We’re going to sell it and we sold it at a modest success. And then 10 years later, the market was really ready for it.

(00:27:01) Sometimes that’s thought products. We wrote Remote: Office Not Required, 2013. Thought, hey, market’s right. No, it wasn’t. If we had published that a year ago, we would have hit it bang on on the wave and we didn’t. So it’s not a foolproof method. But it is a method that is sort of batting to the successes that we’ve had and learn from that and trying to replicate those things.

Shaun: (00:27:24) Fantastic. Well, I think that’s a great place to end it. I do want to say before we leave. This episode, as well as all the other ones, are edited on Adobe Audition. So Adobe, we weren’t trying to drag you. I love your service. And next week, we’re going to be discussing the next chapter in REWORK, which is called “Planning is Guessing”. So we’ll have both you back for that.

(00:27:45) Jason Fried, thank you.

Jason: (00:27:46) Thank you, Shaun.

Shaun: (00:27:47) And David Heinemeier Hansson, thank you.

David: (00:27:49) Thanks.

(00:27:50) Broken By Design by Clipart plays.

Shaun: (00:27:55) REWORK is a production of Basecamp. Our theme music is by Clipart. You can find all of our past episodes at rework.fm. We are on Twitter at @reworkpodcast. If you would like to follow along with our little book club, crack open your copy of REWORK and flip to the chapter “Planning is Guessing.” We’ll be discussing that one next week.

This is an episode title

00:00 / 00:00