SEASON 2 - EPISODE 171 - OCTOBER 29, 2025

Fizzy Q's and A's

With the launch of Fizzy getting closer, 37signals co-founders Jason Fried and David Heinemeier Hansson use this episode to answer listener questions about the upcoming product. They talk about how Fizzy and Basecamp will coexist, why aesthetic design choices matter, and which AI features are actually worth using.

Watch the full video episode on YouTube

Key Takeaways

00:00 - Introduction
00:11 - Fizzy and Basecamp, competitors or complementary?
02:00 - Finding the right tools that fit your workflow
11:39 - Why aesthetics matter in software
18:45 - Not every AI feature adds real value

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 Kimberly Rhodes from the 37signals team, joined by our co-founders, Jason Fried and David Heinemeier Hansson. We’ve been talking a lot about a future product we have coming out called Fizzy, and we’ve gotten a couple of questions either from our YouTube videos, Jason’s been doing some posting on X. Thought I’d pull some of those out and let you guys answer those directly. So the first one is from YouTube. Someone wrote, “Something that just passed through my mind, isn’t Fizzy a potential Basecamp competitor?” What are you guys thoughts about that?

Jason (00:36): Sure, maybe not. I don’t really care because they’re both ours, in a sense. But really they’re quite different in scope and scale. I mean, Basecamp is something you can run your entire business on. It has all the different tools you might need. Fizzy is not that kind of thing. You can run a lot of stuff with it, but hey, choose whichever one you want and we’re happy to have you as a customer is the way I look at it. So I think that something in someone and somewhere is always competing with you and it might as well be yourself if that’s kind of one of the outcomes. So yeah, that’s how I kind of see that sort of thing.

Kimberly (01:06): And I guess we could say Basecamp, project management software, Fizzy, you’re not considering project management software. It’s more of a bug…

Jason (01:13): You could, I mean there’s different scales of projects and people, for example, manage entire projects on Trello, which is a Kanban board or with to-do lists or with paper and pencil or no software whatsoever, you know, in their heads. So you can do all sorts of things with all sorts of tools and this is just another take on a way to organize information, to track issues, to track ideas, to track bugs, to track a project in fact. And I think people will use it for that. And again, if they want to use that, that’s a better fit for them, I’m glad that we can provide that option. So, that’s my take on it.

David (01:42): I think the greatest illustration of this is that we’ve been competing against email with Basecamp for literally 20 plus years. People still using email. They’re also still using Basecamp. And the two things coexist even within a single organization. So almost anything can do almost everything. If you look at these information tools, you can do everything in notes. As Jason says, you can do everything on pen and paper, you can do everything noodling in your little head, but what has the right grip for the particular project and for the particular group of people, that’s quite different. When you find something that really feels like, yes, this is it. This is what I’ve been looking for, that shape fits perfectly with this problem, you’ll know. And the only way to really know is to pick it up, to try it out. Does this feel right? It’s funny I mentioned the knives because we’ve been actually shopping for some new knives and I mean just like cutlery, and I am always amazed by how quickly I can look at a piece of cutlery.

(02:47): I go like, eh, no, I don’t like that. No, I don’t like that. I didn’t know I had all these opinions about cutlery and their design, but I clearly do because in about 2.7 seconds I can look at a picture and go like, no, no, no, absolutely not. Or ooh, I don’t know, maybe. And I think the same thing is even more true when it comes to software. You look at something and you think like, oh, I have this thing we want to work on. This would be perfect for it. Or you don’t know. You simply just like, well, I’m going to spelunk about, I’m going to have a look. I’m going to kick the tires. I’m going to see if this is something that I actually like to do. This is something I think Jason and I used to do both more. When I was living in Chicago, I loved going to the dealership lots literally to kick the tires.

(03:32): Now, I don’t know if I actually kicked the tires, but to take a test drive in a car that I thought, do you know what? Maybe there’s a 7% chance I’m actually going to buy this. I was always trying to rationalize in my head that I was not going to waste someone’s time. If there was literally no chance that I was ever going to buy that car, I wouldn’t go. I wouldn’t ask for a test drive, but if there was a 7% chance that I might like it, I’d go test drive it. And you know what? Lo and behold, you realize all sorts of things. I would build up this idea in my head that like, oh, this car, this is exactly what I want. I have the checkbook in my fucking pocket and I show up. I just want to do the confirmation drive.

(04:08): I give it a drive and like absolutely not. How did I get into my little head that this was going to be a good fit for me? I hate it. I hate everything about it, and then I’ll try something else. And it’s like, ooh, I had no idea. I had no idea that it, I like just the way this feels. I like the buttons, I like the clicks, I like the rattle, I like the ooh Mercedes. The whole game there where you realize that just those tactile elements, that’s actually what sells you on it. And software has so many of those, so many of the micro transactions that you can’t even capture in a screenshot. Like the flows. What does it feel like to put in seven issues to track that, assign it to a few people, move it around, keep up to date with it. Do you know what you’re not going to know until you take it for a test drive?

(04:59): Now the great thing about almost all internet software is you can literally just take all of it for a test drive. This is why whenever folks ask me is like, so what should I pick? Should I pick Basecamp or Clickup or whatever, one of the competitors? I’m like, you tell me. Go for test drive. They’re both free. You can literally figure it out in 10 minutes. 10 minutes is actually enough to get a 60% indication on whether this is something you’re going to enjoy or not. Because as I am with my cutlery, as I was with the test drives, humans are really good at honing in on what do they like, what do they don’t like? And you will realize quite quickly you sign up for one of these competitors and you give it a go and maybe it’s just right. Wonderful, I’m so happy for you.

(05:44): Or maybe you realize this is a little clunky. This is hard to figure out. I had to look up the manual three times, and then you try Basecamp and it just goes like click. That’s exactly how I think about problems actually, projects and problems, and now it all flows in. So Fizzy is essentially another way of giving you another car in the lot and you want a new model. It doesn’t do all of it. It’s not an SUV, right? It can’t fit everything. It can’t go to fucking Lake Tahoe in the middle of winter. That’s small tires. So you got to stuff things in a little more neatly maybe, but also maybe that’s just the feel you like. Maybe that’s all you need. Maybe you just drive to the fucking grocery stores a few times and one time around a little hill and that’s your transportation needs. Great! Fizzy, man.

Jason (06:31): That’s the slogan. Fizzy man. So I think the knives metaphor is actually quite interesting because you could think of Basecamp almost as a set of knives. You can buy a set of knives together and then you can buy specialty knives. And a good chef would tell you, all you need is one chef’s knife. There’s all sorts of different ways to use these things, and I’m thinking about we have a set of knives. I actually have a couple really, really nice individual knives. They’ve got some separate pairing knives from a different brand just because we wanted to try them and I happen to the handle, the blades are fucking fine on all of them. It’s the handle I like on one of them, and that’s the one I reach for. And you just don’t necessarily know what it’s going to be until you try these things, which is why, like David said, people should try all sorts of things.

(07:13): Now, one thing you don’t want to get into, I don’t think, is this habitual dissatisfaction with everything and feel like you need to keep jumping around forever until you find something. You should find something that works, use that, and then have curiosity about other things, but the feel of something, it really does click in. This just feels, like the handle of a knife, I could describe it, but when I feel it, I’m not describing it, I’m feeling it. I just know why I like this somehow, right? It’s a preference and it’s very hard to describe your preferences, ultimately, it’s just something about, maybe it’s the weight or the feel or the material or the texture or the thickness or the thinness or whatever. However it hits your particular fingers. That’s what products are like too. And I think when people try Fizzy, they might go, oh, I’ve used Kanbans-esque things before, but this feels right to me.

(07:59): This feels like a good handle on a good knife. I want to use this instead. So they reach for that and you can reach for both. We use Basecamp. Basecamp has card tables, which is like a Kanban-esque tool, which Fizzy in some ways is based on conceptually, but visually is very, very different and we use both. Some of our teams use card tables in Basecamp. Some of our teams use Fizzy. We’re having more people use Fizzy for certain kinds of things. You can use both. It’s okay too. You don’t have to wonder which one do I have to use? The one that feels right for the job. Companies have different jobs, so I think it’s kind of a nice thing to offer two possibilities.

David (08:32): And this is why the feature comparison chart is such a bane on this industry that it reduces all products down to check boxes. Does it have this thing? Now put aside for a moment that they’re all bullshit. All of them are like, let’s just come up with some feature and we’ll frame it in such a way that there’s an X next to the competitor and next to us there’s a nice little checkbox. But you know what? Even if we’re comparing our own products, if you are just going to compare Fizzy to Basecamp, Basecamp is going to destroy Fizzy on the checkbox comparison chart because Basecamp just does a whole lot more. It can run your entire company. Fizzy is less and what a moronic way of comparing those two products and figuring out which one you’re going to go with. The feature chart comparison is the worst possible way to pick a piece of software. This is how you end up with the enterprisey bullshit that no one wants to use, right? Oh, I checked all these boxes. Yeah, but did you like it? Did you want to use it? Could you figure out how it was? Then you onboard someone new, could they figure it out? Or you’re just pulling your hair out? It’s actually remarkable to me just how much software in our industry alone people truly hate with an absolute passion of a thousand suns. Now, I haven’t actually ever used Jira, so I feel

(09:59): Okay just reciting the amount of hate that Jira gets online. It’s unbelievable. Maybe it’s one of those classic things where the people who talk shit about Jira outnumber the people who would say nice things about it simply because that’s what the internet does. It detracts all the shit that someone wants to say about something, and it’s not the full picture, but holy shit man. The amount of dislike that exists in this world for some of these products should give you some indication that the feature chart is not the be all end all. If you look at the feature chart of a Slack or a Jira, you’re going to find about 2 billion check boxes, right? It does this, this, this, this and about a trillion other things, and none of it matters because for a ton of people, the grip is not just wrong, it’s backwards. It’s like, holy shit, this is a knife, blah.

(10:51): And then I’ve zapped myself in the goddamn chest. Yeah, okay, no, don’t like that knife. Don’t like the one that tries to murder me. Don’t like the one that wants me to jump out of a window Jira style when I’ve been frustrated for umphteenth time. Now, some of this is of course also people just projecting their miserable lives onto inanimate pieces of software, and that’s also a thing that happens, but you know what? It also falls the other way, and that could be quite positive. Sometimes an inanimate piece of software can give you all sorts of nice fuzzy vibes. You like the people who make it, you like how it looks, the frivolous color choices, you like some of the little flows, you like some Easter egg stuff and it gives your little smile in your otherwise dreary day, and isn’t that a service?

Kimberly (11:39): So let’s talk a little bit about that. Since David, you brought in aesthetics, Jason, I’ve seen you posting things on X about Fizzy and the visual aspect of it and how that is what makes it different. Talk to me about that. What have you guys put into this product that people are going to see soon that has made it different from other kanban tools?

Jason (11:59): I think it is interesting because in our industry, most software actually looks the same. Really almost everything looks the same. In most industries though, things stand out because of the way they look. Clothing, cars, art, obviously there’s a lot of things that have kind of come a little bit closer together in the car world. BMW and Audi kinda look similar now, and there’s things like that, and of course each brand kind of goes back and forth and there’s sort of some consolidation around things that are forced because of safety regulations and whatever, but generally each brand has its own distinct look and feel and vibe and taste and flavor. In the software world, it’s not typically that way. For a long time it was like everything looked like Slack.

(12:37): Actually, a lot of things still do look this way. You have a thin sidebar on the left, a darker sidebar, the rest is white and it’s like that’s how it looks. Or things are always in perpetual dark mode and it’s just text and boxes and dark mode, right? And in some cases that’s cool, but it’s rare to actually find software in the business world specifically that has its own distinct flavor, and we try to bring that to our products. Nothing else looks like Basecamp. A lot of things look like Monday. A lot of things look like Clickup. A lot of things look like Asana. Nothing else looks like Basecamp unless someone’s distinctly copying Basecamp and nothing else looks like Fizzy. Nothing else looks like HEY, there’s no other email tool that looks like HEY. There’s a lot of email tools that look like Gmail and it just is like people tend to follow the leader in that respect and they forget that they’re not making a name for themselves.

(13:25): And so we’ve always tried to do stuff that doesn’t look the same as everyone else, not only for the sake of being different, but sometimes simply for the sake of being different. There’s value in that too. It’s okay to be different for the sake of being different. We happen to think though that there’s actually just a feel, these things have a feel and a mood to them, and I feel like it’s incumbent upon us since a lot of other companies don’t seem to do this is to think about the mood of the product that you’re making. How does it make people feel when they use it? Do they feel like they want to smile? Does it make them feel kind of good about themselves or is it like this dour, sad, depressing, everything looks the same situation? And so with Fizzy, we’re trying to bring some color partially inspired by what David’s been doing on the Linux side of things, themes and colorful things and colorful backgrounds.

(14:06): It’s nice to see some color again. It’s nice to see some saturation again in the colors as well, like pastel colors have been something that some people have been using in software for a while. We’re going full on saturated colors and some gradients and bringing some texture back and having some texture behind things. I just think it’s nice to see texture and color and shape and vibrance, and so we’re bringing that to Fizzy and some people are going to go, I don’t like this is too much. I get it fine. That’s cool. I think a lot of other people are going to go, this actually feels pretty nice. I feel good when I use this. I want to actually turn to this. I like when I hold this knife. I like when I hold this product. And so that’s sort of what we’re aiming to do here.

David (14:44): I think what’s so interesting with the car metaphor as Jason puts it, is that very few people feel any need at all to excuse that the reason they bought a certain car was because they liked the way it looked. In some cases, it’s literally just, I liked the color, I liked that it was red. I wanted a red car, so I bought this car, and they don’t mind that like, that’s a very prime objective for how they pick things. Somehow when it comes to software that’s almost unprofessional. I just need to be functional. This is what I meant, especially in the Linux world when Omarchy was first bursting on, I put a lot of emphasis on the aesthetics, on the saturated color, on color coordinating everything, so it looks really nice and people were like, why are you wasting your time? As though somehow aesthetics were beneath us, as somehow aesthetics weren’t important.

(15:35): This is extremely important. Surrounding yourself with beautiful things that give you joy and pleasure in life is right up there. Alright, bread, water, aesthetics. That’s about the Maslow’s pyramid of needs here. You need beautiful things to sustain yourself, and we feel perfectly fine doing it and all these other facets and aspects of life. When it comes to software, suddenly we have to be Mr. and Mrs. Rational all the time. Can I get that in cornflower blue? That’s about the degree of flourish that you can have, right? And I do actually think, I want to blame Apple for some of this, and not because I don’t like Apple. I think Apple does exceptionally well of what they do and their aesthetic. It’s so uniquely Apple. But it’s also, as Jason says, is incredibly muted in a lot of cases. It’s all about this piece of aluminum that’s just been just right.

(16:34): Okay, it’s gray dude, and it’s been gray for 20 years. Could we get that iMac back that had all the funny flourishing colors? That’d be great. I’d love to see a goddamn MacBook that didn’t just look like the same gray it has for 20 years. So, when a company like Apple has had such a dominant position in our industry and has influenced so many designers, I think that’s sort of just seeps into the water and they feel like they all have to be monochrome and monotone in the same way and all this gray and you know, no, no, they don’t. Let Apple be Apple. Continue with the great bullshit for another 40 years. Great. You do it really well. Everyone else? Find some fucking color. Reintroduce it. The world is going gray already. We’re talking about cars for a second. I remember this chart I saw about the color palette of cars.

(17:25): You go back to the eighties and nineties, great variety, and then you see the whole thing just compress. And now the majority of cars are white, gray or black. You’re like, no, get a fucking green car. Get a red car, get a blue car. Get any color that isn’t white gray or black. And I also like white, gray or black. I’m talking as much to myself as anyone else, but I did just buy a red car recently and it’s sitting in the garage riding on Copenhagen next to two gray cars and I was actually thinking like, do you know what? It has to be a color. It has to be a screaming color. I cannot be a contributor to this monochrome world. It’s too depressing.

Jason (18:06): Look at us black and white. You’re wearing black and white. I’m wearing gray and gray. Anyway, when there are two choices I’m going to pick the more colorful choice is what I’ve decided to do most recently, and it’s enhanced my life a little bit, actually, in the smallest little way, but in a meaningful one. It’s just fun. There’s different degrees of fun, but I think it’s a little bit more fun to use a product that’s a little bit more colorful and has a little bit of texture and the transitions are kind of delightful but not so ridiculous and not in your face. And speed is fun, and feeling like you’re able to enjoy things when you look at them is fun. It’s fun. Fun should be part of it. Now of course, it’s not as fun as a lot of other things, but it’s a little bit of fun is a good, it goes a long way I think, especially in software.

Kimberly (18:46): Okay, I’m going to go to this one last comment for you guys to noodle on. Let me know what you think about this. I know we’ve talked a little bit about introducing some AI into Fizzy, maybe not right at first, but eventually. Someone wrote, “For what it’s worth, anytime a company adds AI to their product, I cringe. As a customer, AI no longer feels valuable as a feature to me. I use AI, I like it, but I want to choose when and how I use it.” I want to get your thoughts on that.

Jason (19:14): I generally agree, which is why we actually had some AI baked into Fizzy early and we pulled it out and there’s one more thing sitting around. I’m not sure we’re going to ship it. It was an experiment and I think it’s good to experiment and I also think AI can be very valuable in a variety of different applications and different kinds of things. I’m sure David can speak to some stuff at Shopify, he’s talked to Toby about and whatnot. AI can be incredibly valuable. I think when it’s actually just pulled for the lowest hanging fruit and like, well, it can just summarize this, so let’s have it summarize that, or it can help you write this, so let’s have it write that. That’s kind of sloppy actually, and I don’t think it’s really very handy at all. In fact, it’s probably worse. I think that there’s a cringe factor, but I don’t think it’s a smart thing to cringe anytime you see something like that used anywhere. It can be used well in some places. It can add a lot of utility in some cases. But a healthy skepticism I think is healthy, but you also don’t want to close yourself off because certainly there are places where it’s valuable. I don’t think a lot of products have actually figured that out yet, though, to be honest is sort of my take.

Kimberly (20:11): I feel like this is such a polarizing issue. We have people ask anytime I do a live office hour session, someone asks about AI in Basecamp in particular, and people feel very strongly one way or another. Like, don’t touch my product, I like it how it is. Or, everyone else is doing it. Why aren’t you guys?

Jason (20:28): Well, I think one thing I’ll add, I know David’s chomping the bit to get in here, but the beauty of it is natural language. The fact that I can tell the computer what I want it to do for me is an incredible leve. And typically, historically, we’ve all had to use interfaces, which are translation layers essentially, and I hope this thing has a button that lets me do the thing I want to do versus just being able to tell it what you want to do. So I think there’s a ton of wonderful things that are going to come from just natural language interfaces and the AI part there is primarily translation. You tell it what you want it to do and it’ll figure out what it can do. I like that a lot, but the general low hanging fruit summarization stuff, I’m not finding very useful at all.

David (21:06): I think the term healthy skepticism is one of those encapsulations that we throw around all the time. I’d like to see some healthy optimism. That’s what I try to embrace because I totally have the same reaction to a lot of the AI sludge and slob that’s just filling out like, no, I don’t want this. Where do I turn it off? But I constantly try to pair that with some of that healthy optimism. Do you know what? I’ve seen AI do the most incredible things where I just had to lean back in my chair and go, hot diggity damn, what a time to be alive? I didn’t actually say hot diggity damn, but that’s what I was thinking in my head, that AI truly is incredible, and when you see the glimmers of its majestic abilities, it’s hard not to be taken aback. What we have right now though is mostly glimmers, mostly this shimmering of, do you know what?

(22:06): At some point it’s going to do most of everything really well. We’re not there right now. We have these glimmers. I use it all day long, just as a developer it’s incredible. Just the amount of information it’s able to distill for you, APIs, and the rest of it, even when it keeps hallucinating and it makes up stuff that doesn’t exist it’s still still hugely valuable. But you take that something that has a relatively high error factor and then you put it into real product that someone actually has to use, and suddenly your tolerance for error really drops. The great example we’ve had for quite a long time is self-driving cars. Elon famously went out in 2017 and said that at the end of the year, all the cars are going to drive by themselves. We don’t even need to manufacture more steering wheels. Now, that didn’t pan out. Firstly because, well, it wasn’t even really AI at the point, but then even after it became AI, I drove my Tesla for a while with the self-driving version.

(23:04): I was like, you’re insane if you’re trusting your life to this. Absolutely goddamn insane. And then suddenly, seemingly overnight, you were no longer insane. You were almost like, why aren’t you using this? It was so incredible. It’s been working so well. The journeys I’ve had just taking me from where we are in Malibu to LAX, the airport in Los Angeles, truly magic. So you have this that somewhere, some people are figuring it out. You should be trying to figure it out, but you don’t have to ship all the in-betweens. I think that’s what people don’t like right now. Is there are all these experiments and then there’s all this pressure that you’ve got to stamp AI on freaking everything because that basically means new. It means modern, it means with it, but a lot of times it just means failed experiment. I mean, we have a long list at this point of experiments that we’ve done internally, features we’ve created, features we’ve injected AI into and then benchmarked and run with and ended up deciding, you know what?

(24:11): Either this isn’t better or it’s not better often enough. It fails too often. It’s wrong too often. Fizzy had several really interesting interface ideas built around this natural language recognition that were really nice until they weren’t. It’s actually the age old clippy problem. Microsoft was one of the first companies trying to go with this, hey, it looks like you’re writing a letter to your lawyer. Do you want me to help it format for you? And you’re like, what? No, I’m writing the PTA. What are you talking about? I don’t want this reformatted in some formal way. Comic sans will do just fine, thank you very much. That when you are wrong, when AI is wrong in a context where you don’t feel like it’s appropriate to be wrong, the context I have for self development, I know it’s going to be wrong all the time.

(25:02): I don’t need a hundred percent hit ratio. I just need like 80%. I’m good. I can filter out manually the 20% that’s nonsense and still take the 80% that’s valuable. That does not work in an end user application. If the interface idea we have for Fizzy only worked 80% of the time, you forget those 80% real quick and just focus on, can I throw this fucking thing out the window because it’s wrong one in five times, a little bit like the self-driving cars. Even being correct like 95% of the time, that sounds really high. Wait, does that mean it’s going to murder about 200 people a day? Yeah, no, that math doesn’t work. For self-driving you actually got to be right 99.99996% of the time maybe, and the bar’s not quite as high for the software that we do, but it’s probably 99. Even 1% error rating on something you use on a regular basis is going to feel quite annoying, and way too much software is being shipped right now with AI features that at best work 87% of the time.

Kimberly (26:08): Okay, well, with that, we’re going to wrap it up. Fizzy is coming soon from the 37signals team. This has an episode of REWORK. Rework is a production of 37signals. You can find show notes and transcripts on our website, 37signals.com/podcast. Full video episodes are on YouTube. If you have a question for Jason or David about a better way to work and run your business, leave us a video question. You can do that at 37signals.com/podcast question or send us an email to rework@37signals.com.