SEASON 2 - EPISODE 192 - May 13, 2026

Clearing the decks for launch

With the launch of Basecamp 5 approaching, 37signals’ co-founders Jason Fried and David Heinemeier Hansson discuss what happens inside their company when it enters “launch mode.” They talk about temporarily shifting priorities, the difference between routine operations and launch periods, and why putting lower-priority tasks on hold is often necessary to make meaningful progress. Plus, Jason and David share their favorite aspects of the newest version of Basecamp.

Watch the full video episode on YouTube

Key Takeaways

00:10 - Pausing processes and the power of “not now”
03:41 - Moving quickly and deciding to just build it
08:38 - Pressing pause on routines to stay focused on the goal
11:20 - Staying flexible while working through the chaos
13:54 - What life after launch will look like
20:54 - Jason and David share their favorite Basecamp 5 updates

Links & Resources

Transcript

Kimberly (00:00): Welcome to Rework, a podcast by 37signals about the better way to work and run your business. I’m your host, Kimberly Rhodes, joined by the co-founders of 37signals, Jason Fried and David Heinemeier Hansson. We’ve taken a little bit of a podcast hiatus and that is because we are in launch mode. I thought we’d talk a little bit about what that means and what we’ve cleared out of the way in these final few … Oh, I don’t even want to put a timeframe on it, this final time before we launch a product. So maybe we just start with the podcast as a first thing. Jason, tell me along with that, what are some of the other things you’ve just put on pause until we get ready to go?

Jason (00:39): Yeah. I mean, the podcast was something, we do these on Tuesday mornings and we’re a remote company and an international company. So we only have a few hours of overlap. For example, in the morning with Europe, I’m on the West Coast of the US. So talk to people in Europe. I’ve got between 8:00 AM and 10:00 AM-ish kind of, you know. And we used to record the podcast around that time. At that point, it made sense not to do those for a little bit so we could squeeze in some more meetings with people and conversations and design reviews and stuff in the morning. And so we just kind of pushed that off for a minute. We’re back to doing one before the launch of Basecamp 5, but sometimes you just kind of move some things aside for a minute. We do one, move some things aside. We don’t have to make these rigid decisions about is this in or out.

(01:20): But we are, for example, right now, not going to fix a non-critical issue in HEY. Let’s say a customer reports a bug. It’s non-critical. Yeah, we know something’s busted but it’s fine. Here’s a workaround. Otherwise, you can find yourself constantly doing all sorts of other things and having no time for the thing that you really need to do, which is right now we need to get Basecamp 5 in launch shape. We are literally a few weeks away and at this point it just wouldn’t make any sense to divert ourselves into doing other things. And this is also true with Fizzy. We’re not currently working on Fizzy. There’s some issues over there. There’s some things we like to do, all sorts of things, but we’re like, no, not now. In fact, we have a column in Basecamp called Not Now, which is exactly what this is all about. Maybe later, but not now. And that’s, I think, the way to think about this. And so there’s a whole variety of little things like that, big things and small things that are currently in the Not Now column internally.

David (02:12): And one of the other things we put aside is process. I mean, we have run Shape Up with these cycles that run in a fixed amount of time with the shaping and all that stuff. In a launch mode like this, it’s more like new product development where it’s a lot more free rolling. Jason and Brian on the product side are in there making decisions about exactly what goes in and if the quality is right and if the screens are where they need to be on a daily basis instead of sort of on a cycle sequence. And I think that energy is also just really important to distribute in the entire company because we don’t do this that often. Even with the launch of Fizzy, it was actually not the entire company that was fully dedicated because it was much smaller product that we were pushing out.

(03:02): Basecamp though is huge. So it really is the entire company that is dedicated to it. And then some of it too is simply making yourself available. We have a Security, Infrastructure and Performance team that does a lot of work across a bunch of different things and the projects that they had don’t always overlap with product projects that we work on. But as we’re coming to the final stages towards launch, kind of the plate is clear. If we need anything, everyone is available to jump on, whatever it is, as soon as it is, there’s nothing that needs to be scheduled or lined up, just full availability across the board.

Jason (03:41): By the way, I think it’s a great point that you made up of a process I totally forgot, which is the most obvious thing that’s really changed around here is it’s more like we’re playing Tetris in a sense. We’ve got two days, let’s fit something in there. We’ve got six hours, let’s fit something in there, versus waiting to kind of shape everything up fully, put it on a more formal process timeline. So it’s very much every day things are shifting a little bit, shifting a little bit less as we get really close, but there’s all sorts of things that are just changing that normally would’ve been like put on the back burner or like maybe we’ll get to that later or let’s have someone write that up. It’s like real time, just in time, on the fly with everything. So it’s a little bit chaotic and there’s regressions every once in a while.

(04:22): It’s like, “Hey, this doesn’t work anymore. It just worked yesterday. What happened?” This is kind of what happens when you’re running really fast near the end. There’s some regressions, but for the most part, I find it to be incredibly exciting. Actually, I was talking to Jeff, one of our employees who’s been here for what, 19 years or something and he’s like, “This is how it used to be. This is how we used to build things, way, way, way back in the early days. We would just build stuff. We would have ideas and we’d make something. Oh, here’s a good idea. Let’s do that. All right, let’s do that.” And we would just make stuff. It was a very small team. There wasn’t time to even sit around and draw things up. When you have more people and less connection with the day-to-day work, it does help to draw things up and think things through a bit more because you’re going to hand stuff off and not be in it as much. But right now we’re very much in it and it’s kind of a really nice return back to the way we used to make things. And I’d like to see if we can keep this pace up in a lot of places actually. I think we’re going to make better products this way.

David (05:16): Now the funny thing about quality is actually a bit of a throwback to perhaps one of the most maligned expressions in software development. Facebook used to have this slogan, run fast and break things. And for all sorts of reasons that got politicized and got ridiculed and it’s also completely true in a moment like this. When you are running fast, you are going to break some things, but you know what? It doesn’t matter that much if we’re all here to quickly fix it because the flip side of whatever moving slowly and not jeopardizing anything is that you move slowly and you don’t get to the finish line, you don’t get to the great product. And I think there’s time for a little bit of both. There’s that run fast and break things mode we are in right now where we have the full attention of everyone such that if we do mess something up a little bit, it doesn’t take that long.

(06:11): And we got a great test of that just last week really or last two weeks where we made a bunch of fundamental changes to how the UI in Basecamp 5 works. We use a lot more permanent UI elements with a permanent sidebar and how notifications are going and permanent trays. And that basically just changed the performance profile of all of Basecamp, and it didn’t actually feel that great. It looked great, but the performance regressions were severe enough that the product just felt sluggish. And here we are just a few weeks out from the end and we’ve run quite fast and a little bit of something was a little bit of vibe experimentation making it into the code base. Some of it was just not paying attention to that because it wasn’t so important until you realized what shape you wanted things in. We weren’t going to optimize it until we knew exactly what we finally wanted, but now that we did, we fixed it.

(07:09): About a week’s worth of really intensive focus on making sure that Basecamp felt as fast or faster than the existing version just paid huge dividends. And you could have looked at that and freaked out. You could have looked at that and gone like, “Wow, there’s only a few weeks left. We’ve been working on this stuff for months and months and it’s in this sorry state of a performance predicament?” Yeah, it is. And we can just fix it. We can just move fast and fix things. Just as you can move fast and break things, you can also move fast and fix things. And I think that mode of having some tolerance for a little bit of risk, a little bit of mess, a little bit of incomplete thoughts and a little bit of even by sloppy code making it in, as long as you clean it up and as long as you’re cognizant of the debt you’re taking on, is great.

(07:59): And the freedom you get from moving in that way without a lot of planning, without a lot of structural overhead where most of your direction is simply guided by what’s broken next, what’s not fast enough? Here’s a card, here’s a card, here’s a card. I went through a tear last week just filing cards like a robot, like an AI could fill in a whole card table in Basecamp with all sorts of little niggles that weren’t quite right and would you look at that? Boom, boom, boom, boom, boom. By the end of the day, they’re all gone and the product just got better and now we’re almost ready to go.

Kimberly (08:38): Okay. I want to go back to the process changes because there’s two things that I have never seen happen before with this launch that we change as far as policy. One, Heartbeats, the product team got a little break on writing their heartbeat. I want to talk about that. And then we also shifted Summer Fridays. So kind of talk me through those changes and those decisions in light of this whole launch mode situation.

Jason (09:01): Right. I’ll take the second one first. We normally do summer Fridays, I think May through the end of September, I believe is the range.

Kimberly (09:08): May 1st through September 1st.

Jason (09:09): Okay, that’s where it is now? Yeah. Okay. It’s changed over the years. So we generally take four day weeks during the summer months essentially, which we cannot do if we’re about to launch a product for a number of different reasons. So we postponed that, pushed that off I think till June or something. So we’re going to kind of shift that whole window one month out. And it’s just like an obvious decision. There’s no big group meeting about it. I think Andrea maybe brought it up to David and I and we’re like, “Yeah, that makes sense. Done.” Literally five seconds. This is how we try to make almost all decisions in five seconds. Yeah, that makes sense. Let’s do that. And then if we made a mistake, maybe we make another five seconds and fix the mistake. But this is also kind of like breaking things, whatever. It’s not a hard thing. We don’t need to sit around and discuss it very much, but it made sense we did that.

Kimberly (09:53): I will also say just from an employee standpoint, because we just got a notification about it, I think everyone was like, “Yeah, that makes sense.” I’m like, “Of course, that makes perfect sense.”

Jason (10:01): I mean, I could see some people being annoyed, maybe they had a vacation. I get it, but it also, this is what we’re here for.

Kimberly (10:06): Sure.

Jason (10:07): This has to happen. Everyone knows that. So we’re doing that. The heartbeats and kickoffs and stuff, those are typically cycle based. Historically every six weeks the team will write a Kickoff or Heartbeat. Right now we kind of have suspended this six-week cycle thing. There’s some notion of it, but not really. And we’re just running fast and finishing. We will probably fall after launch, we’re going to still be running fast because there’s going to be things that are broken and things we didn’t see and new ideas we have that we didn’t get in and all this stuff. But at some point we’ll probably settle back a little bit and get on a more regular cadence. But right now it doesn’t really make sense.

(10:42): Also, this is the only thing we’re really doing primarily. So we’re all seeing the work that’s happening also. A big part of Heartbeats is to sort of share work across the company that people may not be aware of necessarily and celebrate that work. That’s kind of happening almost every day in a sense right now. So yeah, it’s not quite as formalized, but it’s just kind of a suspension of formality. Formality typically is seen as process that’s unnecessary. I think the Heartbeats and Kickoffs are very, very useful, incredibly useful. But right now we can just hold on them and just we’re all just going full speed ahead basically right now and reporting can happen later if it needs to.

David (11:19): And I think what’s key about that is to realize that all these processes we’ve set up, all these Heartbeats and Kickoffs and whatever, they’re there for great reason, but they’re not written in stone. We can change how we do things for a period that asks for something else. And one of those periods is right now as we’re finishing a product up, we’re sprinting. Now I would say it’s a very measured sprint compared to a lot of other organizations who do sort of death march style crunch times where everyone is sleeping under the desks and in the office. This is our version of like, do you know what? Normally we have a very calm, regularized cadence to how we do things. And then occasionally, very rarely actually, especially on a product that’s large as Basecamp, do we launch something massive and new? And when that happens, once in a blue moon, once every few years, it’s completely fine to suspend all sorts of things, all sorts of processes, all sorts of expectations, moving the Summer Fridays back, all of these things that we have that feel like on a day-to-day basis as though they’re just part of the water, part of the machinery, this is just what we do around here.

(12:38): Yeah. Okay. What we do around here can change when it needs to and it can change in a way that’s temporary. I mean, we also know that this mode would not be pleasant I think to be in if it was going to last two years or even if it was going to last six months. The whole acceleration of the sprint is because we’re nearing the finish line. We can see the line, we can see the flags, there’s going to be champagne and it’s going to pop some fireworks maybe. And that’s what’s giving that energy to get that last mile and actually feel good about it, that there’s an expenditure of extra energy towards a specific goal. I’d actually say that when we’ve gone too long without something like that, feels like something I’m just missing. It feels like a certain acceleration is not present. I wouldn’t want to do it all the time, but occasionally if you don’t have it, I think you’re going to go sloppy. I think you’re going to go a little precious. I think you’re going to become a little rigid if you are always in the same lanes, on the same schedule, at the same time. These things are great as a moving average of how we usually do things and then leave the little excitement.

Kimberly (13:53): Okay. So let me ask you about post-launch, kind of asking you to predict the future, but with your launching product experience, when do you imagine things to be out of this launch mode? Because I don’t imagine it’s like the product launches on day one and day two, it’s like no more launch mode. Everyone’s just back to their normal life. Do you imagine that there’s like, is it a whole other cycle of this expedited work? What do you think?

Jason (14:19): I think it’ll be more chaotic post launch. So we have the advantage right now, I don’t know if people realize it who have not been part of a launch, but we have the advantage of not having to put any of this stuff in front of customers right now. We are our own tough customers, but it’s not the same as dropping a massive change on everybody. We’ve tried to make this as familiar as possible, but also as fresh as possible. So there’s a fine line, but certainly once we launch this to hundreds of thousands of people, there’s going to be incoming, a lot of incoming praise, complaints, misunderstandings, total understandings, like all the things, all the spectrum of feedback, great, good, bad, ugly, you guys suck, you guys are amazing, all the things. So we’re going to have this huge barrage of particles being shot at us basically.

(15:06): And that’s actually the hardest part of it all is I think the first couple weeks post a major launch like this because now you have to take tens of thousands of opinions into account, anecdotal situations where people, someone is used to one thing, this just changed on them, but they’re really good at persuasion and telling a story and you’re like, “Wow, this is a really big deal. Maybe it’s really actually not.” You’ve got to kind of sit with this stuff for a little bit, and that’s also hard, like hearing all this stuff, even if it’s great, but still hearing it all and being like, “We’re not going to do anything right now” or “We really need to fix this because we did break something badly.” It’s hard. So I think this is going to last for, I’d probably say at least another six weeks or so of sort of accelerated work, but actually maybe even faster because now we’re reacting, but we also need to stay proactive and that we still need to continue to improve the product in our own vision.

(16:02): There’s things we want to get in that we didn’t get in. There’s things that we know we’re going to ship with that are not quite right and the challenge is not to all of a sudden go 100% on what other people want. It’s to maintain forward motion and then also deal with the reactions. So it’s going to be chaotic for a bit as well. I think it’s again, equally exciting if not more so, but that’s reality. When it dies down, I don’t know, but it usually takes some weeks for sure.

David (16:26): Now I think on the one hand, we have a blessing here by the fact that Basecamp 5 is built on what we call the chassis, the technical underpinnings of Basecamp 4. It is the largest change we’ve done to an existing chassis, the most elements that have moved around, the most introduction of new features. But fundamentally, the chassis is very proven. It’s very well tested. The database schema is broadly consistent and the same. We’ve changed some critical areas and we have to keep an eye on that, but we do not have to keep an eye on all 400 screens because there’s a lot of it that didn’t see fundamental change to how the backend works, for example. And I think that gives us an enormous amount of reassurance that this isn’t completely unproven. We’re not going to flip a switch and then we’re just going to have an enormous amount of untested code out in the wild.

(17:23): But on the other hand, this is also the highest criticality we have. This company exists because of Basecamp. It exists in this size and this shape because of Basecamp and because of all the customers that we have. So when we flip the switch, the full brunt of that is going to come to bear. We have all this very nice tested chassis set up, but we’re introducing a massive change to an enormous amount of people all at once. Because the luxury we have when we launch a brand new product, at least sometimes as we did with HEY, or the plan was with Hay was we’re going to do a ramp. We put the new product out there and then we sort of let in a few thousand here, a few hundred there and we can take it measured. We don’t get that with Basecamp. When Basecamp 5 goes live, it’s live for everyone. We’ve done a little bit of early adopters or early view stuff now, but it’s not the same thing at all. So it is this mixed blessing where we have some solid ground to stand on, but then we also know that this is the biggest stage there is.

Jason (18:31): The other thing is, there’s stuff I would say is like there’s some surprising things that happen. I’ll share one small example. So we have, think about 20 customers now on Basecamp 5. They’ve raised their hand. They go, “I want to be an early adopter.” So we’ve added them to Basecamp 5 on their account. So one of the things we’ve done in Basecamp 5 is we’ve brought activity to more places. When you hit a certain screen, you can see what’s been happening in a given project or in the account. And he goes, “I love this, but I often give presentations where I put our Basecamp account up on the screen and just yesterday I did something in this other project, which was like this HR project and this one thing is showing in the activity stream that I don’t want other people to see.” Now this is not something that enters our mind when we’re making a product, which is like, out of the five things that show up in activity, maybe one of the things shouldn’t be shown to some people in a presentation because you kind of can’t think that way, otherwise everything becomes like, well, we have to hide this and we have to hide this and we have to hide this and I have to hide this and think about every scenario you could possibly imagine.

(19:32): On balance, we think it’s better to show activity than to not have activity. It’s better to show action than static in some of these cases. So anyway, his point was totally completely valid. I can imagine the scenario, I’ve done this myself, I’ve wanted to give screencast a Basecamp and I’m showing something in our account. I’m like, oh shit, it has a launch date on it. Now everybody who saw this presentation, if they notice that, now they know when we’re launching this thing. I mean, the only way to go through that is to actually go and edit a message and delete the text and whatever. And it’s just part of accepting the fact that every screen is not for every situation and there are going to be some moments when things don’t work exactly as you expect or want and to solve all those edge cases is not worth it.

(20:18): Now, we could go into presentation mode where you could redact. I could imagine a feature where we could do all of that, but we’re not going to do that, especially not right now. So anyway, my point is that when we ship this to not 20 people but 20,000, 100,000, 200,000, whatever it is, there’s going to be all sorts of things like that that are going to pop up that are not really scenarios that you can really fully, truly account for or even that you would have an answer for. Some things are just going to exist, but to field all that feedback and have that come at you is sort of the hard thing, especially when people have anecdotes that are really important like that.

Kimberly (20:54): Okay. Before we wrap up, one open-ended question for both of you, what is your favorite thing? What are you most excited about with Basecamp 5?

Jason (21:02): I mean, I just think the whole product is so much better. I was going back and forth between four and five recently and I just don’t want to use four and it’s that feeling. It’s not like, oh, we’re leaving something behind that I still want. I don’t have that feeling at all. There’s like no nostalgia for Basecamp 4. I’ve actually had nostalgia for Basecamp 2 in the past. I think we did some really nice stuff in two and some of those ideas actually been brought over to five. Two is a very interesting version of Basecamp. But so for me, it’s the whole experience. I like opening it up. I like having the sidebar open sometimes what has my notifications. I like that I can do multiple things without having to leave where I was, but there’s a lot of small, little subtle things, hundreds of little improvements that overall just feel a lot better to me. It also just feels a lot more modern and current and I just like the UI a lot better and there’s a lot of little things like that. So it’s not like one thing for me, it’s just collectively I want to be here and not there.

Kimberly (21:55): David, what about you?

David (21:57): I was going to say, we’re going to give you one thing, but I’m going to give you two things.

Kimberly (22:00): Okay.

David (22:00): The first thing is how quickly I can get caught up on everything that I have that’s unread in Basecamp 5. We’ve just made it so much easier to bam, bam, bam, go through five, seven, 22 things that you need to get caught up on, especially because in an account that’s as busy as ours, we are in Basecamp all day long and this is a company of 60 people constantly updating everything. And I happen to be on a great many number of projects. There’s just so much stuff going on all the time that a lot of it I need to just look at for three seconds to know whether this requires my attention or someone else got it they can run with, it’s great. And be able to just accelerate that flow of here’s a tall stack of things I need to look for if I need to be involved in this or I don’t need to be involved in.

(22:53): And just being able to process that so much quicker. It actually in retrospect, how did we even do this before? Which is funny because I mean, we haven’t been using Basecamp 5 that long. It hasn’t existed in this usable shape this long, but I’m already like, Jesus, if I had to catch up on these 27 things in Basecamp 4, it’d be kind of painful. The other one is keyboard accessibility. Now, obviously after moving to Linux, after building Omarchy, after going all in on the terminal, I’ve really become in tune with how fluid you can command an application just from the keyboard alone and not touching your mouse. And we’ve gone all out. This is such an enormous upgrade to Basecamp’s keyboard accessibility, bigger upgrade than we’ve done any other product. And I mean, we’ve had some good keyboard accessibility in HEY for example, but it feels so much more novel in Basecamp because historically that wasn’t something we focused on and now we just have a beautiful system.

(23:54): I’ll just leave you one thing. You hold out the Shift key and all these little keys pop up all over the interface. So you can see that like, oh, if I click Shift S, I’m going to open the sidebar. If I click it again, I’m going to close it. And you can fly through these things in a way that feels very reminiscent for me of my text editor now, Neovim, which is also all about doing these combo moves. I had one conversation, I was on this one page and I was like, “Oh, I got to add a card from something.” I jumped quickly to the project I needed to go to. I went to a bookmark actually of a card table. I hit Shift M, instantly it popped up a new card. I started filling that card out. I hit control return to submit that card and at the end of that key combo, I was like, “Yeah, that was fucking cool.” The way you can move around the application with such speed and fluidity, when you learn just a handful of these key findings, which takes a little time to do, but very well worth it if you’re going to spend a lot of time in Basecamp.

(24:55): And in fact, it brings up one of my favorite mythical moments, which is the travel check-in person at the airport when they have one of those really old school systems that haven’t been converted to a web app yet and you can just see them go like, “Oh yeah, I’ll change your flight right now to Dallas going through Austin and back.” And you’re just like, “How many key combos are you posting?” You are a savant at this system, you’re clearly flying through it. It was probably designed in like 1972 and yet it’s amazing at that kind of usability, not just learnability, not just approachability. Those things are also important, but I feel like in modern software we’ve kind of tilted too far that everything needs to be infinitely discoverable as the only grail we’re chasing. Well, could we also spend a little time on the people who are going to take hours, days, weeks to become experts at how they use the system?

(25:52): If you do it great, you can have both. And I feel like Basecamp 5 has the closest approximation of both of those things. You don’t need to do any of the keyboard stuff if you don’t want to, but you hold down that little shift key and it’s like this magic portal opens and suddenly everything the matrix reveals itself and you just see all the keys and you go like, well, I could probably could do this. I can learn one, two, three, and then by the end of the week you’ll just be flying.

Jason (26:17): It’s a neat point, actually. If I could add something about that, it’s like this isn’t something we intended, but what sometimes you build a system and something reveals itself, which is that it feels like you can be like a Basecamp pro now. And in 3 or 4, you could get as good as anybody else, but you couldn’t kind of get beyond that. And it feels like now there’s with the keyboard command specifically and some of the UI, the way certain things work, it feels like you can graduate into being a real true pro with this product and move through it much faster and do more things than you could ever do before. And you actually want to spend more time in it because of that, because it’s actually kind of fun. It’s almost like playing a song, you’re playing with the product. I find myself wanting to be in it more for that reason.

(27:01): I also find myself just wanting to look at it more for other reasons, but I want to be in it in a way where I didn’t feel like I wanted to necessarily be in Basecamp 4 before, not to denigrate the product. You’re more in and out and you can still be in and out, but now it’s like there’s more things you can do and you can have fun doing them because there’s a skill that you pick up versus just moving the mouse around, which is, it’s a skill to know all the things a product can do, but it doesn’t feel like it’s a skill you can graduate beyond. So there’s something subtle about that that sort of was an emergent property of adding all these keyboard commands and also some of the other ways the product works. There’s a lot of things that are far more accessible.

(27:38): Part of this is the design. For example, you can be on something, look at your bookmarks and not have to leave where you were. You can be somewhere and look at your upcoming events. You can see upcoming events that are coming up in the next 15 minutes automatically. You can see tasks if you want to check on something without leaving. You can look at something, ping someone, you can be in a group chat, ping someone on the side, you can do all these things simultaneously, but it doesn’t feel overwhelming. You can put them away if you don’t want to see them. And that whole feeling I think really levels up the product in a significant way that people aren’t going to notice on day one, but that they’ll begin to notice, especially if we explain and show these things off that they’ll really, I think, feel like they’re growing into something and that’s a really nice feeling in general.

Kimberly (28:19): Okay. Well, with that little preview, we’re going to wrap it up. You can always sign up for a free Basecamp account at basecamp.com. REWORK is production 37 signals. You can find show notes and transcripts on our website at 37signals.com/podcast, full video episodes on a YouTube. And if you have a question for Jason or David about a better way to work and run your business, you can leave us a video recording. You can do that at 37signals.com/podcastquestion.