The no-code movement is picking up steam, with more people than ever building apps and businesses without knowing how to code themselves. Ben Tossell (@bentossell), the creator of Makerpad, is betting his business that no-code is the future of work. However, Sahil Lavingia (@shl), the founder of Gumroad, isn't so sure that code. In this episode, I hosted a lively discussion between these two thoughtful bootstrappers about code vs no-code. Which approach should a new indie hacker should take? What gaps in the market are opening up due to the changes in tooling landscape? And what does the future hold?
What’s up everybody? This is Courtland from IndieHackers.com, and you are listening to the Indie Hackers podcast. On this show, I talk to the founders of profitable internet businesses and I try to get a sense of what it’s like to be in their shoes. How did they get to where they are today? How do they make decisions both at their companies and in their personal lives, and what, exactly, makes their businesses tick?
The goal here, as always, is so that the rest of us can learn from their examples and go on to build our own profitable internet businesses. Today is a special episode. I have two guests who’ve been on the show before. Ben Tossell is back from Makerpad, and Sahil Lavingia, the CEO of Gumroad, is also back. Sahil, Ben, welcome back to the podcast.
Thanks for having us.
Thanks for having me.
So the three of us are gathered here today to discuss the no-code movement. Sahil, you are an early engineer at Pinterest. You built Gumroad by coding it and hiring engineers to help you code it, so obviously you’re going to represent the programmer’s voice in this discussion.
Ben, you are not a developer, and your company, Makerpad, helps others like you learn to build and operate businesses without needing to code. You are riding the wave of the no-code movement, and I think that makes you the perfect person to explain. Ben, what is no-code, and why does it matter?
For me, no-code is just being able to build something, a solution to your problem where there’s a small tool to help you do your podcast now. One of the guys just did a tutorial on automating the episode link to guests, for them, after their podcast, or whether it’s a big thing, like how to build a Gumroad Patreon clone without code, which we did last week. That was not intentional for this podcast.
Sounds like somebody’s coming for you, Sahil.
It could be a range of things. It’s not needing to be technical and not using the “I can’t code” excuse as a barrier to build anything. But yes, it will be interesting to see how this conversation goes in terms of, it seems that there’s a big one versus other type discussion on the internet. It feels like you’re either in one camp or the other, where I don't think that’s quite how it is. But we’ll see.
Well this being a debate that’s probably exactly what we’re going to do is draw a line down the middle and duke it out. It’s funny to look at the history of no-code. I’ve been making websites since I was a kid in the 90’s, and even back then there were a lot of no-code tools around to help people put together websites without writing any code.
Sahil, you learned the code well over a decade ago. Do you think there’s any real difference in the no-code tools of the past versus the ones that we’re seeing today?
Yes. The first websites I built were using, I guess, no-code tools, like Dreamweaver. I don't know if the audience is familiar with Dreamweaver, but back in the day when everyone used tables to design stuff on the internet and a billion images because browsers couldn’t really render anything natively, rounded corners and dropped shadows and all these things. WYSIWYG, I guess, is what it was called back then.
Yes. What you see is what you get and drag and drop website builders and that kind of stuff.
What you see is what you get. Yeah, I guess Dreamweaver back in the day. Now I think it’s gone. I moved to WordPress and started building all my websites on WordPress, which you can argue is halfway a no-code tool. It takes care of a lot of stuff for you, but if you want to do anything super custom on your theme or something like that you have to finagle around in PHP.
It’s like a low-code tool.
It’s a low-code tool, and then transition completely to just building my own stuff from scratch or using a framework like Ruby on Rails or Django or something to build stuff but taking care of all the active elements in-house.
So you took a no-code approach to learning how to code. First you started with Dreamweaver. Then you moved on to WordPress and eventually writing your own code from scratch. I think the fact that this all happened so long ago highlights the fact that these no-code tools have been out forever.
They’ve been out for decades, and yet there’s never been any sort of groundswell, there’s never been any transition where suddenly, developers don’t have jobs, or suddenly most companies or the best companies and startups are all built without code. Ben, your company, Makerpad, is really a bet on the current no-code movement and the idea that it might turn into something big. Why is now the time? Why is this going to be bigger today than it ever has been in the past?
Well I just think the tools are obviously getting hugely better. It’s also how everyone seems to be wanting to work. Everyone wants to build their own thing, which, Li Jin from Adreessen posted something about the passion economy recently, that was like, if you want to start a niche newsletter that you get people to pay you for, so that’s a little side project or whatever it is, you can do that on Substack, or you could build that with other tools and manage it yourself.
I think these weren’t necessarily as easy maybe back in the day with Dreamweaver and all that sort of stuff. I don't know. I just missed all that stuff. I wasn’t technical, basically, even enough to pay attention to any sort of tech stuff until it was a bit late, which is why I was probably looking at tools to help me skip the curve a bit and almost cheat my way to getting the end result that I was thinking of.
But I think there’s a big change in what people think of as startups, especially with the whole Indie Hackers movement, which obviously you’re familiar with is, there’s tons of people building small businesses whatever way they want and making a living from it. If you can make $300 grand a year from having a website up, using no-code, to me that sounds like phenomenal business if it’s just one person running it.
I think that’s what more and more people will start looking into and starting to do. I think that even Sahil’s Gumroad empowers those sorts of people too. It’s definitely a huge inflection point in that sense. I think the no-code tools help those people along the way.
Yeah. I think it depends on how you measure growth. It’s unlikely, in my view, that you’ll see a billion-dollar startup that doesn’t have a code base on GitHub or GitLab or something in the next five to ten years. But I think you’ll see a huge amount of newsletters, journalists, educators, teachers, writers, film makers, going more and more direct-to-consumer.
These people were never technical. Some of the least technical people that I know are creative people, which is surprising to me, or was. But those people make six figures a year, and there might be hundreds of thousands of those people. Will they cross the threshold into making $15 million a year? I don't think so. Not with no-code.
But it depends on what you want. It depends on, do you want to build the next Stripe? Which frankly there will be one more of, maybe. There’s just not that many opportunities to do that. Then yeah, you don’t need to.
I would say the other point I have is, I like the idea of being able to choose no-code instead of being forced to do it. I like the idea that, if I want to, I can understand what’s going on under the hood if I need to, versus be constrained potentially by basically having to take it to a mechanic. If we use a no-code or low-code tool that doesn’t do what we need, or just increases our pricing by a ton, it’s just risky, potentially, to build a business like that.
Having the flexibility, like for example, we use Circle for our integration testing. We could do all that stuff in house. We could step up a little CI server but we choose not to. But they just increase their pricing. So that’s a decision that we might make. It used to be $300, now it’s $1200 a month. Is it worth it? But at least we can have that conversation, whereas if we were like, “We can’t do that,” that’s a $900 a month cost to not being able to do that.
I’ve interviewed dozens of founders on the Indie Hackers podcast who are non-technical. They don’t know how to code, but they’ve still built wildly profitable businesses. I think one of the most common patterns I’ve seen from these non-technical founders is that they still work with developers.
A lot of them found a way to partner with a developer as a co-founder or to hire developers for cheap, or fun and creative ways to work with coders. Very few of them built their businesses entirely on no-code tools. Ben, are you seeing any companies that are built entirely on no-code tools today that have been able to reach a substantial size?
This is one I guess sort of goes to the scaling question and tons of people ask this. Lambda School was built on no-code tools. It was built with Airtable, Slack, Zoom and Notion. At Makerpad, we’ve got one of the guys from Lambda School who’s been able to tell us how they’ve been able to build certain things without code.
At this point now it is just starting to break, I think. I think that they’ve really pushed where these tools can go, which is fair enough. I think that there were, what, 150 million in Series B, was it? And then they’ve got thousands of students all working together every single day on these tools. That really is pushing the limit.
They’re pretty big.
Yes. I think that’s fair enough to think of, at this current point, you can literally take it that far. It can get to that point if you made a good enough business, though I don't know why so many people worry about that stage when you maybe get to $10 first. Get to a hundred dollars, get to a thousand. But I think everyone worries about,
“What’s my Airbnb version of this look like?” when you’re probably not going to get there.
That’s what happened, I think, with me gravitating towards no-code, was initially I thought I had to build something that was, it’s got to be Airbnb style. It was going to be, you have a co-founder who’s technical or you learn to code and you build this massive thing, and that’s just how startups are. That’s the just world of startups.
But for the last few years, with the Indie Hacker movement and everyone’s seeing different things, you can just do your own path. And like we were talking about before we started recording this, there is no right answer. The ones you read about are the outliers and the exceptions, so don’t worry about the hugely successful version of your own product when you haven’t even started building anything yet, cause that’s probably a long, long way off.
I think that’s reasonable. For me, I built Gumroad in a weekend with code. It probably would have potentially taken me longer.
That was 2011 so who knows if it was even possible to build that with no-code back in the day, nine years ago, eight years ago. But even if it was possible, I think it was faster for me to build Gumroad with code then read Makerpad for hours and hours, probably more than – I know, because I get into the rabbit hole. “I don’t need to learn. I don’t need to build anything. I can just watch other people build stuff. This is really cool.”
So that’s the meta-point. It’s easy to be distracted, I think, by no-code potentially, where you might solve the problem faster. I guess that’s a fear that I have, is that people that could learn how to code, it’s like a one marshmallow now, two marshmallows later sort of thing. Delayed gratification, where maybe no-code is pushing people to get the one marshmallow now whereas if they just invested a week, two weeks, three weeks, they could get two marshmallows later.
I agree that it is unwise to spend a ton of time thinking about problems that don’t matter and statistically will not be relevant to you. I think the idea of potentially building, let’s say Gumroad, with no code, launching it, seeing how well it does and then deciding, “Okay, we have certain aspirations that require us to rewrite the whole thing.” We did rewrite the whole thing from Python at Ruby. So feasibly we built the app again. We thought it would be easier to hire and our first engineer was a Ruby developer.
That’s actually pretty common. I’ve talked to so many founders who’ve built a V1 of their app, especially founders who outsource that development to, I don't know, some sort of Dev shop in Eastern Europe or something, and it turns out that that gets them to the starting line and they’re able to get their first set of users, but then their app crashes or it’s crappy and they need to rewrite the entire thing from scratch, which I think speak to Ben’s point, which is that maybe that first version doesn’t need to be in code.
Maybe you could have done it faster if you used no-code tools, got something out there to test the market, validate that your idea’s a good one, that people want to sign up. Then since people who write apps and code often rewrite them anyway, it’s not that difficult for you to then think about using a code basically and sticking with your no-code solution.
I feel like I just scored an own goal.
But it’s still, a little point you made was for you, the quickest way to do something, the most effective way to do it was to code something. For me, that’s definitely not the quickest or most efficient way to do it. If that’s your tool, that’s the way your brain works or you like doing it that way, then that’s absolutely fine.
I think there’s a way to build something Gumroad-esque or Patreon-esque without code, but how far is that going to get you compared to what the Gumroad product looks like today? Nothing like it, I imagine.
Yes. Oh, yes, for your sake, I guess.
One of the common themes here is that no matter what kind of company you want to build, if you’re coming in as a founder and a creator, you’re coming in with a lot of background knowledge.
In Sahil’s case, you’re able to launch Gumroad in a weekend, sure, but really you learned to code for years before that. It wasn’t really just a weekend that it took you to get Gumroad off the ground. It was based on all of this other knowledge you’d accumulated throughout your life.
And Ben, the same is true for you as well. You’re building without code. You were able to build Makerpad without code. You’re able to put out all these tutorials teaching others to builds apps and businesses without code.
But you couldn’t have done that if you didn’t spend all the time learning the ins and outs of these different no-code tools and their landscape, and which tools are better or worse for different jobs.
So we already talked about Sahil’s path to becoming a developer. Ben, what does your path look like to becoming a no-code master? How long can someone expect it to take for them to learn how to use these no-code tools effectively?
I think mine was completely unintentional. Like I said, I felt like I was bit behind with how I didn’t jump on the code wagon early enough. I was lucky enough to get a job at Product Hunt, which meant I was surrounded by people making things every day, which sort of lights a fire under your belly to think, “I’ve got to do something. I want to do something here. I need to figure out how I can do this.”
Then I was around at Product Hunt when things like Webflow and Bubble were launching. I got to see that and thought, “Wait, I could use this to build something.” And then it would just ignite that maker, whatever you call it, I wanted to tinker around and play with things. From there, I just never stopped.
I assume it would be the exact same thing if you got to a point that those hours are always just spent on Product Hunt learning things, trying stuff out. If you’re doing that just as, “Okay, I’m going through some code basics. I can now launch this thing and build this thing,” you then start tinkering more and pushing it more with the code thing.
I think those avenues are very similar. I definitely tried to code a few times, but I just kept on getting blocked and thinking, “Okay. This text video doesn’t work because the last thing I did just broke. So now I’m stuck and I don’t know what to do so I’m just going to give up,” whereas with the no-code thing, I always somehow managed to troubleshoot my own problems.
I just figured it out. It was just an easier visual thing for me to be able to think, “Okay. I know how I can figure this thing out,” whereas with code, I’m literally just copying where the text was on the screen. Everything came out right. I thought, “Fuck it. I’m not coding. I guess we’re heading back to the drawing board.
I just kept on going down that path and it happened over a number of years, then, and I didn’t really realize I was building a foundation for building a business without code, teaching others how to do it. I kept on thinking, “Well, I want to build this type of thing,” and then figuring out I can build something that looks like this type of thing. Then people just kept on asking me, “How are you doing that?”
Do you think there needs to be a Lambda School for no-code, a curriculum dedicated to teaching people how to learn these no-code tools? And if so, how much of an investment do you think it will take? Cause it can easily take a year to become a competent developer who can build your own web app. How much time do I really need to tinker with no-code tools before I’m competent enough to start building real businesses off the back of my skills?
I do think it just completely depends, cause there’s so many different types of things. Again, IntegraMap works differently to Zapier. But I like Zapier. We work with them and I just happened to fall down the Zapier route.
That means I use Zapier for basically everything, whereas IntegraMap, I could just as easily swap that out on many of these occasions. But for whatever reason, I’m in the Zapier boat. We love working them. We use them on basically everything.
The curriculum part is interesting though. I think we at Makerpad are looking at this, because it just seems like an obvious thing that people always saying, “What about Lambda School for this?” It’s something we’re looking at but it’s nothing we’ve figured out yet, to be honest.
I think there’s so many different options for, what would that curriculum look like? Is it a case of, we do a lot of different courses, like how do you create a podcast? How do you be a podcaster without needing to be technical? I think there’s opportunity there, which is what we’re currently tinkering with inside.
Sahil, what’s your take on all of this learning? You mentioned earlier that it might be a one or two marshmallow situation and that maybe there’s more benefits to be reaped in the long term if you spend your time learning to code instead of learning these no-code tools. Why do you say that?
I think I historically treated code as a rung above no-code on this linear ladder. I think this idea, I could build Gumroad using code in a weekend, I don't think today I could build Gumroad in the same period of time using no-code tools. My argument that you should be able to choose code as an argument for code should also, I assume, apply to no-code then.
I’m not competent enough in no-code, don’t understand it enough, don’t understand the glue. So much of coding is not coding. It’s gluing things from stack flow together, fixing bugs and hoping that you Google for a comment in Github that has 60,000 uploads and you’re like, “Yes. This is the thing that’s going to solve my problem.” Then that did not go away.
I think no-code, I guess to me it almost feels like, how you describe code is how I feel about no-code, where I don’t really understand it. I don't know even the limits of it, potentially. I don't know how the things work together and how this ecosystem is, whereas is someone was like, “Hey, I’m throwing a podcast.”
I would very quickly be able to build a website, RSS feed, from backend tools to allow for publishing of content, automated stuff to go to Twitter, etc. I could do that with probably a simple Ruby app or something like that on Heroku, is probably what I’d use. I could not do that with no-code, and maybe that’s a sign that I should.
I would say one argument for code is that there is this system that has been around for a long time, so you can ballpark yourself in terms of, “Can I do certain things? Am I getting better at the things that I need to get better at in order to make a fulltime living doing this?” whereas I think with no-code, it seems like you could build something that looks really great and really powerful way faster, but then what’s next?
You can get to the end really fast, but then there’s not something beyond that. Will you just hit a limit where Bubble or Glide or Webflow can’t do what you want, and then you’re screwed. You just sit around emailing their support every month, being like, “Hey it would be really great.”
I know with Gumroad we get people, creators, that have asked for the same feature for years and we’re like, “Sorry, it’s just not a priority for us.” It would be cool, I assume, if they could fork Gumroad and then build their own, the feature they need, for themselves. That’s where we want to get to eventually.
I’d say that’s an argument for code. It’s just older so you have more context and more resources and more search results in Google if you run into a problem that you need solved. I also wonder how you hire people under no-code.
If you're building a startup with no-code, you have a bunch of, oh, the scale of this is large now with no-code that if one or two people want to be an organization that would have 15 engineers in it 10 years ago, or do you end up with a place where you have 15, 20 no-code people?
How do those people work together? How do you have separation of concerns? With code it’s easy. You have the back end, you have front end people. With low-code or no-code, what does that look like? How do you have different departments in a no-code environment?
Well Ben, you’re actually doing this at Makerpad.
Yes. My current contractors are no-code people. One is operations. One is tutorials and we work in the same tools, Notion and Airtable and Zapier. You are right. I think it was interesting what you said about, it’s as much about learning how to put these things together, figuring out what went wrong, how to fix the bugs. It’s the exact same type of process in no-code.
That’s why I keep on saying that we’re on the same path, or a parallel path rather, to those things and still a similar way to figure things out. But like you said, code has been around for a much longer time. These tools, Glide and Webflow, etc., they’ve come a long way in that they are building new things quickly and seeing, luckily, how people in Makerpad or the no-code community are trying to build things which then helps them move their roadmaps or prioritize things over other things.
I know that Webflow, especially this year, has clung to no-code movement and talked about no-code. They’re at the no-code conference next week. It seems like that was a valuable piece that they’re trying to build toward. I think with hiring people, I think it’s more like, what are the things that you’re trying to get done?
If it’s, we need marketing systems and we need things for marketing, social and content and all that sort of stuff, you could easily have people who are no-coders go in and create these systems. But it depends what tool you were working on as a business. If you’ve got a ton of code related stuff within app events and all this stuff happening with coding you can only get so far with some no-coders.
But I think there is going to be more and more people being able to build their own mini tools, as no-coders, to say, “Oh, I could monitor our reddit comments and any time our company or podcasting is mentioned on reddit, we can have that come into a Google sheet. Once that goes into a Google sheet, that gets posted in my Slack and then I can one-click this one thing and send this response,” or whatever it is.
Those types of things would have previously been done with code, and those are now becoming way more accessible to people who don’t know how to code or don’t even need to, because maybe that is the quicker way to build that thing.
Yes. I think a lot of those services, especially internal tools, things around marketing and growth and BD and sales, all of those things should get out of the code base. That’s something at Gumroad that we’ve been working on, is taking stock of all the things in our code base that aren’t fundamental to Gumroad as an app, as a product.
For example, someone signs up with an @nytimes.com email address, we might want to email them and say, “Hey, what’s your number of Twitter followers” or something. We might want that to show up in Slack. And that’s currently tons of different Ruby for worker processes that run on every user, between 50 and 100 lines of code each.
I think it’s great to delete all that stuff. Delete all of it and move it to an external app, a no-code tool. One thing, as I’m thinking about, we just hired a content person and she’s working on creating a Facebook group for all Gumroad creators that have made over a thousand dollars. For example, she can’t code. She doesn’t know SQL, and so she has to ask someone on the engineering team, in this case myself, to give her a list of emails of everybody that’s made over $1000.00.
On Gumroad it takes me five minutes because I know SQL. I can do a simple join on users to purchases. I have all these things. What’s the answer to that in the no-code ecosystem, if we want to get - I know there are certain assorted tools to get general data, but if I want something very specific, like how much volume has comics done on Gumroad in the last 30 days, was another query that I wanted to do recently. Knowing SQL is, to my knowledge, the only real way to do that.
Yes, but I think there’s tools like Clay or Retool which have some ways that you can, as a developer and non-developer, work together in the same tool, so if there is something you need to link up to say, “Okay, this where all your data will come from,” then that goes to the no-coder in this scenario where they can just run those queries and do those things.
There’s also tools like Standard Library which, again, is similar. There’s lots of these that are bridging the gap between the two, which means that it doesn’t have to be one or the other. Then almost it becomes, the no-coder does these queries or figures these things out and then recognizes the pattern for, “Okay. I know how to do that on Designers” or whatever it is, the next thing to evaluate.
Then they pick up smaller bits of, how do I manipulate this query or this code stuff by tiny bits at a time, which also then may help them want to figure out that they want to learn to code, which is an interesting pattern of no-coders, is this might be the best and strongest way for someone to figure out, they do want to learn to code.
They haven’t just done the first three modules on Code Academy and now are still getting stuck that what we’re telling them or thinking, well, “I built this thing and then I want it to do these things, and yeah, maybe if I learned to code I can do these. I know I want to manipulate data and work with data in these sorts of ways, and that’s where I want to go with my career.” That definitely could be some help with both of those paths.
This is interesting, because I grapple with this at Indie Hackers. The entire website is built using code that I wrote myself, but everybody that I work with is not a developer. They don’t know how to write code. They’re not familiar with our code base so they have other responsibilities, and so what I find myself doing is building tools for them.
Ben, you mentioned Retool. I’ve got a ton of retools set up to let you do things without using code. For example, you can deploy an episode of the Indie Hackers podcast and take it live on the website. By filling out a form you can change the title, you can change the release date, etc., all without writing code, whereas before that was a process I did myself using code.
If I look at Gumroad, I think Gumroad is similar to Indie Hackers. It’s all built using code. Sahil, if you want to hire some no-coders to bring them in, you really need to build some tools just like I have, whereas Makerpad was built without using any code from the get-go. It’s all built on no-code tools., so Ben wouldn’t have to do any extra work to bring on some no-coders.
Yes. I think one of the powerful things about building on code, too, which I miss when I leave the code sphere of influence, is get inversion control and this very clear log of what’s happening in terms of security, in terms of feedback, in terms of making sure that only a certain quality of code gets through. What I would love to see more broadly is that apply to other things.
For example, in Notion I would love the idea of pull requests and merges and merge conflicts and comment within just the document setting, because I think that’s something I really struggle with. Someone can go in Notion and edit a bunch of stuff and there’s no real great way to say, this stuff you can’t edit or it has to be approved before it’s merged into the master.
The idea of forks and branches, I think, that come from code typically are just phenomenal ways to think about really building anything. When I write, when I do other things, I think of that forking and merging metaphor all the time.
I think that’s another thing about code that’s really great, is it gives you access to some of these paradigms, conditional data structures, algorithms that you might be able to use in daily life outside of code that are just interesting concepts that you might not get in the no-code setting.
It’s funny. We’re unearthing some good startup ideas here. That is, pretty much all the things you do using code, people who using these no-code tools are going to want to do.
The fact that we have tools like Git that are super powerful, that are a staple of every single code base, the fact that that’s missing from the no-code ecosystem makes it not only hard to adopt no-code tools, but it also means that there’s an opportunity to build something there that could be useful for lots of people in the future. But it’s also ironic to know that these tools will almost certainly be built by people using code.
Yeah. It’s ironic in a way. I was thinking about this today, this morning, some of the most important companies in the no-code movement are one, built using code and then two, typically, because they’re building something that needs to be quite complex, they raise a ton of venture capital.
So you have this little ethos drive a little bit where no-code seems to be more about bootstrapping, building a business site that can sustain you and your lifestyle and a community they can care about, not going for a billion-dollar exit. But then to do that, we basically have to start using all these tools that are unprofitable, won’t be profitable for a long time, will likely get folded up into Facebook, Google, Apple or die.
That’s another interesting dilemma I think that the no-code, similar to crypto I think, where you have all this great enthusiasm on the app layer level, but then there are very few companies that are really smart enough, large enough, ambitious enough and have the money to spend on people that can do the deep level thinking on a lot of these things, the architecture. I don't know, it’s just an interesting, sort of weird dichotomy.
The infrastructure level is usually pretty complex and requires code. Nobody’s building website builders without code, for example.
One thing I would say about, you said about version control, we have a version of that in Makerpad. In our Airtable base, when we get any story submitted, if I go and edit that it will go on Slack and say, “Ben edited this and changed it from that to this.” So there’s something there, so maybe we should do some more tutorials around how you do version control and things like that without code.
That’s pretty cool. Let’s talk more about this, because I think for somebody like Sahil or me, it’s easy for us to misjudge the limits of no-code, but Ben, you have a better perspective than we do. What are some things that we might not guess you could build without code that you actually can?
It must a ton. We’ve got over a hundred tutorial now on Makerpad, which is crazy. I’m trying to think of all the things we’ve built. We’ve built, like I said, the Patreon clone which was recent. There’s things Instagram-type clones, Airbnb clones and things like that.
I think there’s two avenues. One is what’s interesting? What’s going to grab people’s attention, which are these bigger Airbnb-style things that are like, “Oh, wow, I never thought I could build that without code.” I built an Airbnb app on Glide and it just uses one Google sheet and a scrap, a host app and a user app, and that can work fully, even sign up as a host, add your properties, add ratings, all that sort of stuff. We built a Cameo clone on Glide as well, again with sort of a celebrity side app and the user app, where you can pay for a video, send a video, and all that sort of stuff.
So there’s lots of these which I think we’ve done quite a few of recently to gauge people’s interest and say these are the possibilities right now and these are cool things we’re pushing these tools to do. But I think there’s a huge opportunity especially for us to build the more useful things, that people can say, “Oh, that would help my day-to-day workflow. I can use that.”
The amount of people who are going to go and build an Instagram clone, a Cameo clone or whatever is not super high. Like you said, these sort of people may not have an app in amount of time. It might just be a, had this idea for Cameo for celebrity animals and it just wouldn’t go anywhere and that sort of thing.
But whereas if you can sort of get into, we came to either of you guys or said, “Okay. What are the things that you’re having to build tools for yourself to connect developers and non-developers? Where are the gaps there and what are these automations that you think should be done and automated but maybe shouldn’t be done without code? Can we build them for you? Can we do tutorials around those?” and that sort of thing.
What does that look like? I think the next part of what we’re doing, is finding real business use cases of, how do we ingrain this a bit more into professional life rather than, “Oh cool, you can build this Airbnb clone,” and everyone’s like, “Oh wow, that’s really cool.” But is it useful? I don't know.
I almost take the opposite stance where I think there is - because people know that no-code can do a lot of that useful day-to-day, automate the boring stuff sort of stuff. A lot of the questions that we’ve been talking about are can no-code do that? Can it get all the way through an Airbnb or an Uber or Cameo? I think, even though there is such a focus on, no one is going to do, there’s very few opportunities to do that, I think people can’t help but want to think like that. “If this works all the way through, will I get” - even though that’s a thing that one percent of the time will actually happen, being able to answer that question.
For example, if there is a Cameo built with no-code, that is way more affordable to celebrities, they take less of a fee because they don’t need hundreds of engineers. They haven’t raised any VC because they have this super lean, mean team that’s all built on no-code, and then Cameo goes out of business because you’re just - to me, that sells the story of no-code so effectively, because the cynic in me says, well, if that’s possible, if you can build Airbnb, Uber, Cameo - Uber is a great example. WeWork maybe is another.
You have these companies that are raising hundreds of millions of dollars building teams of thousands of engineers in San Francisco, paying each person two, three hundred thousand dollars a year. If there is a case for no-code to be able to basically wipe the floor with some of these companies because you build Uber, except instead of taking 20% you take 1%, the economic incentive for that is so large. It’s like, well why hasn’t that happened yet?
If it is possible to build Cameo, you still need to do the growth, the sales, the marketing, the content, all that stuff. But the actual, the 10, 20 million dollars a year that they’re burning, Patreon is a great example. Patreon, burning between 30 and 40 million dollars a year on their staff and office and everything like that, on operational expenses because they have 170 people, to do that they raise $170 million dollars. Creators are rightly concerned about that, because that’s a huge hole to climb out of.
If you’re Patreon and if you’re a creator using Patreon, your business might rely on it, which why we’re building a competitor, but I think you could stretch it even further. Gumroad will be far more affordable because we’re profitable now. We’re not looking to be a rocket ship anymore. We don’t $170 million in preferences to get out of.
But is there an even more lean version of Gumroad that is one person in the middle of nowhere that built a Patreon clone and it, charging - maybe it’s free. Maybe it’s a dollar a month or something insane as someone’s side project. A creator, I think, would find a huge amount of appeal in that. They’re incredibly price sensitive.
It’s like, most basketball players will never be Steph Curry, but the idea that you might be is so motivating to people. The idea that I could have been Mark Zuckerberg was so motivating to me as a child to learn all these things, even though it’s almost delusional, really, to believe that you will be, but sometimes you need that. You need that crazy carrot on a stick to get you even past the first three or four steps.
You might realize down the line, like I did, that’s not appealing to me at all. But it was important to make the investment. I see this in prejudice all the time, where people are like, “What’s the point in trying because even if I get all the way better than most people ever will, I can’t get to the top because there’s this artificial ceiling there?” and then that motivates people from even getting to step two and three or four. So I think showing people you can build this thing that you might not even want to build, but just show you that it’s possible. Here’s a billion-dollar startup that’s run by one dude in Croatia that is Uber but decentralized, and it charges drivers $1.00 a ride flat.
And I think when Meetup changed their model recently - so we did a Meetup clone based from that. We just put pants on, just started and just did that without code. Yeah, I agree that that is always going to be a problem, that you do need to have, this is what you can build.
Because so many people’s ideas that they come to me and they say, “How do I build this type of thing without code?” and I’ll say, “How does it work?” Most of the time, it’s a marketplace. It looks like Uber. It looks like Airbnb or Cameo. It looks like the same sort of thing, and I just say, “Go look at these tutorials.” I think it does use a lot of that, but I think there is things to look into, so it would be interesting to see how it grows.
I can attest to this, too, cause a huge part of Indie Hackers is inspiration. What inspires people to start a company? A lot of it is this aim for the stars. You see your heroes who have these outlandish stories. They’re just crushing it and you want to be them.
And even if you aim that high and you land on the moon, as the cliché goes, you’re still on the moon. It’s still a pretty good place where you’ve gotten, but you wouldn’t have gotten there if you didn’t think you could make it to the stars. Ben, you know more about building apps and websites without code than probably anybody.
I wonder what the limits are. If wanted to build a Cameo or an Airbnb without code tools, at what point are things going to break down? Is it going to be the lack of user authentication in signup? Is it going to be the fact that there aren’t great no-code tools for mobile? Is it going to be that I can’t scale the database? What is it that breaks down eventually, if anything?
Well I’ve never pushed an app to that point, and I don't think many people have. I think that’s what I said with Lambda School. They are struggling now but I think they’ve got tens if not hundreds of thousands of lines in Airtable that they must have people whose sole job it is to actively manage the Airtable base and all these in-house tools.
My guess would be that there’s going to be that level of, that’s what it’s going to be. It’s going to be having ten or a thousand, ten thousand users on your app doing all these things every single day. If one little thing breaks, it may not be as easy to fix or reverse as with code.
But I want someone to use a tutorial, like a Cameo clone, build it and go on something that, let’s see how far these things can go. Because there’s very little people out there who would be able to tell you their answer because not very many people have that successful app regardless of code or no-code.
So it’s a tricky one, and I know that’s one that everyone’s talking about. Like I said, I would always default back to my, don’t worry about that until you get there. If it comes to you’ve got an Airbnb clone for animal houses or whatever it is and then you’ve got a thousand people and it starts breaking, you’re in a very good position to do the next stage of your startup, whether that’s raise money, hire developers. You’ve probably made money on that already.
What about in the short term? Do you find yourself ever frustrated or stymied that there’s things that you have trouble with, things you wish existed in the no-code ecosystem that stop you from getting certain app ides even to phase one?
I think a lot of things look very similar. The building blocks are quite similar, of like, this is the landing page. This is the data base. This is the glue. And it’s just ways of, “How do I wrangle all of these things together in a way that makes it feel and look like a freelance marketplace,” for example.
That can be done. Just if one things breaks on Zapier and something doesn’t quite go through, then the whole thing sort of stops and it’s like, “Okay, now I’ve got to go through all these things and check my Zap history and see what went wrong and debug, all that, too.” So yeah, just figuring these things out, and we need to build the tutorials there to help with those things and how, on-demand, how can people in a community who are just waiting to help on, okay, I’m an Airtable expert so I can help you figure out some of these things.
We’re getting there, cause I think the no-code space is an interesting one that you find us because maybe you didn’t have another path. It was either learn to code or find a technical cofounder and none of those fit you. So you find no-code. You go and find the tutorial. You build something in an hour, two hours. And you think, this is further than I’ve ever got with anything, so this is great.
And then all of a sudden, you fall down that rabbit hole hard. You either build multiple things or you start using a certain Stack. Maybe it’s Webflow, Airtable, Zapier, and then you become a seasoned expert on a few of these things. Then you know, because you’ve come to all the errors, all the ways it works and ways to connect things, you've gone through and figured out how to really get these things working well together.
Then you become that person, the go-to person of, if you want to know how to make a marketplace with Webflow, Airtable and Zapier, then I know the guy for you. It’s this person. There’s a guy doing no-code conference so that’s why this is the example. He’s built one and it works really well for him.
That’s his business and it’s like a full-on, go and find a freelancer. Filter by their skills and hire them and stuff like that. I think if you want to do the no-code, it’s that piece of becoming one of those buckets by using this. It’s interesting to see who fits in where.
Sahil, what are your thoughts on the limit of no-code, not necessarily in building some sort of gigantic company like Google or Airbnb or Stripe, but in just trying to get a new business off the ground? Are there any ideas, any areas where no-code is dead on arrival and you should have started by building on code?
I think those limits exist. There are certainly things that I would not consider building on no-code. For example, if I was building an app to help no-coders do integration tests on their website so that if Zapier broke you’d get a pingdom or something that says, “Hey, this flow on your app that we run every three minutes is broken now,” or whatever the integration test, reinforced QA looks like for no-code.
I think that’s a great thing that should exist for no-code. I think that’s the meta-answer. Anything that’s considered infrastructure, that is considered a service to no-code or other startups, I would probably not use no-code to do.
I think maybe there’s a one-layer-deep stack where you have a product. That product is no-code or uses no-code, and then everything, or the things under that, need to be code or something like that. Maybe you don’t want too many layers of no-code or something. I do think, in general what we’ll see is just like 10 years ago. I think people were super skeptical or did not think Slack and Zoom and Stripe and all these companies were going to be as large as they were. Startups for startups they call them.
No one really, I feel like, knew how large the industry for startups building startups, or startups for other startups was. Now you see it with Brex and with all these crazy companies raising all sorts of cash to go do that, Pilot and et cetera. I think we’ll see a similar thing with no-code, hopefully, where you’ll have thousands and thousands of no-code companies that now exist that will need tools that they’ll pay for.
They’ll be financially successful enough to be able to pay for these tools, which will create all of these great tools for no-code businesses and more no-code businesses will lead to more tools which will lead to more, and hopefully that builds, which I think is happening. You see that with, Stripe is probably the best example of a company that I feel has created opportunities for new companies to get started, which has created new opportunities for other startups to help startups to help startups, etc.
I do think, though, I definitely would default to code for almost anything that I build. For example, I had this idea yesterday called 20XX, which is an idea that is like a WordPress or Substack or some combination of tools that a potential politician would use to run for office at some point in this century. So it’s like, “Hey, I might run in 50 years or in 30 years or tomorrow, but I want to have an issues platform, get feedback, have version control on it, build an email list.”
I just think that would be really great and really important and it would be free, run by donations or something like that. My default would be, I’m going to build that with code, even though all of those things, email newsletter, clearly there’s a tool for that. Donations, there’s I’m sure a no-code tool for that. There’s all these things.
My default is still, I guess maybe comes from fear, but this idea that I need to control everything. I think as it’s coming from a design background especially, I feel I have this stigma, I would say, against sometimes things like bootstrap and things that, in the context let’s say with Gumroad, I wonder if I was able to build Gumroad in no-code, but because it was built like that, people felt that and then didn’t support it in the early days enough or weren’t as excited about it, and therefore Gumroad wasn’t successful and never became a thing.
So I wonder if no-code things potentially defeat ideas that may have been successful. But that idea is one that I think is perfect for no-code. There’s three or four tools you would need as a would-be politician. You could use Notion and Substack or Gumroad or Mailchimp or whatever and Zapier and hook them all together, and donations. Maybe I’m wrong. Maybe that’s a great example of a thing that should exist.
But the idea, I just have so much comfort in like, “Okay. Well I know the first ten steps if I do it with code. I’m going to create a Rails app. I’m going to create a Github repository. I’m going to open source it from day zero. And then I can design basically anything. I don’t have to think about it, because I know I’m going to be able to use html, CSS the way I want to make it look the way it should.
I just perfectly solved that problem instead using a template and add some CSS to modify the visual aesthetic of it. Just kind of a tangent. But I would say that you mention mobile as a place that maybe no-code doesn’t shine yet, and that is essential. Because the thing about code is that I can build from one platform, but then the minute I have to mobile format it. I have to make an android app. I have to make an iOS app.
I think that’s a huge opportunity for no-code to just blow it out of the water. I have to do all my own metatags for Google search results and Twitter, to show up nicely on Twitter and to show up nicely on Facebook and to show up nicely on YouTube. I think the idea that you can build something once and it’s universal is one of the most broadly appealing things to me about no-code. I think we’ve mentioned it on the podcast.
But the point is not only code and only no-code. Almost everything, I think, will be built in a combination of the two. I think Heroku was maybe the best example of that. We basically no-coded OPS, but you wouldn’t call using Heroku a no-code tool, necessarily, even though basically that’s what you're doing is you’re deleting thousands of lines of OPS code and potentially a role in your company for someone that is an operations engineer that for the most no-code you can get.
It’s interesting, Sahil, listening to you talk about how, if you were to start something new today, you would just default to using code, because that’s what you already know. That’s what you’re comfortable with. You already have the first ten steps of any new project more or less mapped out in your head. This is where all your knowledge and skills are.
That resonates with me because I’m the same way. I would probably default to using code too, and I might not even give these no-code tools the time of day. I might not even look into them, because why would I? I already know how to do exactly what I want to do using the tools that I’m used to.
But then another part of me wonders if this is what it feels like to be a dinosaur. Is this what it feels like to be running a newspaper in 1996, complaining about the internet and talking about how you don’t need it because paper has always been good enough. Ben, what are your thoughts here? Are we missing out on something huge? How do we make sure that we keep up to date with things that we don’t quite understand and aren’t just carried away blindly by the momentum of what we already do understand?
I think you guys are saying you already the first ten steps, I think I probably know the same. I know ten steps to do straight off the bat, too, with the tools that I already know. So I’m just as much a victim to those traps as well, I think.
But being currently in this, the finger on the pulse of the no-code movement, I get to see a lot of these tools. We’re trying to purposely work with these tools and say, when you’re releasing new things, release them with us as well so we can show people what is new, what’s possible, and then to go build the news things every day and do new things with no-code tools that we get to see.
I guess that happens with some people building things with code. You’re like, “Okay. That’s a new thing that I could have used in this scenario,” but I don't know. How does anyone solve these things? I think it’s a similar problem.
I think coders and no-coders are actually very similar in their challenges and their thinking, and how we’ve got to all use the things that are most comfortable to us, but also try and remember that things change every year and every month. We should be trying them out every now and then.
Sahil mentioned something that I think is really impactful if you think about the implications, and that’s that it’s way cheaper to build out a team of no-coders than it is to hire a team of software engineers with CS degrees who all expect to be paid $200,000.00 a year.
Ben, you’ve also mentioned the obvious advantage of no-code, being that you can quickly prototype things. In fact, I was talking to Tyler Tringas of Earnest Capital a few weeks back. He mentioned, Ben, that you’ll sometimes tell him that, “Oh, you’ve got a cool idea for a feature or a product,” and then the next day you’ll turn around it’s already done, because you built it with no-code tools and it was super fast.
Just thinking about these things makes me want to zoom out a bit and just think about the ecosystem as a whole. What are the implications if everybody can build applications for super cheap and super fast? If we do a thought experiment and assume that ten years from now, we’re looking back and today and we’re saying, “Oh, wow. No-code really was a huge thing and it really took off,” what’s that world going to look like? What has to change for us to get there?
Well I think like Sahil was talking about, it’s funny how the market of startups helping people build startups was huge, is huge. I think the no-code thing, for me at least, opens up the fact that more people can build even more things. That just blows that up.
I think we’ve all seen this emergence of hundreds of newsletters about very niche things which you could argue is a really good thing or a really bad thing. But I think that those are ways that people can own their own work stuff, where they’ve maybe got a newsletter and membership site that they just run. They get three, five thousand dollars a month, and that’s just how they live their life.
It’s like they’re not just doing some job that they dislike because they want to someday build something, pay an engineer to build or whatever. It’s just a lifestyle. I don’t want to say lifestyle business, technically, because people have a stigma around that. But I think it’s just enabling more people to be a creator and be creative with what they’re doing and the things that they like doing.
There’s always a community out there for that. So I think it’s for the good of everything, but I don't know, is it a terrible thing that everyone then has their own mini business? That could be easily a problem.
Well you’re describing a world where there’s a lot more Indie Hackers, so I like the sound of that.
Sahil, what’s your vision of a world in which no-code is taken over?
I think it would mean that people think about building software like they would think about, I assume, writing a book or making music, where it’s just a thing. You don’t need a degree. You don’t need a two- or three-year long investment course in introducing stuff. I assume music was that inaccessible or game development was that inaccessible, and now it feels like you can learn things. There’s Unity. There’s Unreal. There’s all these tools. It’s like, start making music.
I would love to see a world in which you wouldn’t really say, “Oh, I’m a software engineer.” That’s not a hobby. “I make software.” Most people would assume that’s your job. But if you said, “I’m a musician,” it’s much more common to have that as a side hustle. I would love for that to happen. I think no-code building software should be considered a creative pursuit. It’s like these other things.
I would say you’d probably see an order of magnitude at least in terms of new businesses being built. You should see 100X. If you can remove the bottleneck of having to hire someone with a CS degree at some level to build an internet business, you should see a hundred times more software businesses being created.
I would also, frankly, love to see the destruction of a huge amount of high market companies. When Gumroad, after the layoffs, and I was thinking about what I wanted to do next potentially, one idea I had was to build open source versions of things that I felt should be free or close to free and just take a $10 to $50 billion company, just build open source for them. Just kamikaze or something.
You just want to watch the world burn.
Yeah. I had a joke, if I can’t build a billion-dollar company I don’t want anybody to. There’s still a part of me that believes that. Especially with the narrative the way it is, I think it’s a huge opportunity to take the anti-VC, anti-billionaire sentiment, of which some is invalid and some is valid, and paint a huge target on Uber, for example, would be great or WeWork right now.
WeWork, people argue, is a real estate company not a software company. Then they don’t need engineers. They should be a perfect no-code start. They should take real estate and add a layer of great software that they don’t have to hire a bunch of engineers to manage. I think that would be a perfect candidate.
And then you could charge much less, or you charge what it costs to run the space times a tiny percentage or something like that. So yeah, I’m just excited in general about anything that creates more entrepreneurs, more people building businesses, more people solving problems, and more people just being financially independent and having choice, and not working in a place that they might not want to work just because that’s the only thing they feel like they can do.
Almost everyone I know that is building what they want to build or making what they want to make regardless of the financial viability of it, feels more fulfilled, especially in today’s world where people are dissatisfied with some of the macro trends. They like feeling personally fulfilled, like you’re working on something that matters.
I know so many people that are leaving traditional startup jobs and going to work on climate change or working on some of these things. I think the easier it is for them to comprehend how to build a business to solve some of these problems without being like, “Oh, I have to do Lambda School and a year from now I can build something.” I think that’s great. That would be a great shift.
All these tools are giving the average person much, much more leverage to be able to build something and do something impactful in the world. I think that’s super exciting. It’s pretty crazy how the world’s going to like ten years from now if the stuff really catches on.
To close out this discussion, I want to talk about the fact that most people who want to get started on something might be excited but they’re not really sure what a good idea is. They might not have any good ideas. They’re not even sure what a good idea looks like.
I think it’s interesting that since there’s two of you, we can get both of your perspectives. Sahil, you could talk about what a developer might want to work on and Ben, you could talk about what a no-coder might want to work on. Sahil, let’s start with you. This no-code movement might be taking off. How can a developer, how can a fledgling, aspiring Indie Hacker take advantage of this?
I think if you’re a coder and you want to contribute, you should talk to a lot of these people that are in the no-code ecosystem and figure out where their pain points are. I assume they probably have more pain points per day than almost any other person you can talk to, because if they’re building on infrastructure that’s so new there’s probably an insane amount of opportunity, just like mobile.
It was like, oh, we can build this, but now mobile. This, and now mobile. Uber and now mobile. Airbnb, Pinterest, etc. I think there’s a huge opportunity to take a lot of these startups that already exist but they’re proving now that there’s a need. I can give you a template for working on these problems and then apply it to no-code. I think Rainforest QA for no-code would be a great one.
Literally I would email a founder and be like, “Hey, I love what you’re doing.” They might be making $10K MRR or something like that on their no-code tool or newsletter or little business, ecommerce website. Just say, “What do you do every day that you wish you didn’t have to do?” or make a list of every single recurring task that you feel is redundant but you’re just doing it over and over again because there’s no API for doing something. Or you have to copy/paste something from this website to this website every day or you have to do all this math in Excel before you can do payouts.
Huge - yes, there’s just so much. There’s probably an insane amount of opportunity there. I feel like almost any engineer focused on this could get to $10,000.00 a month in passive income building really simple software for these new groups of people. Then once you’re there, then you have the freedom potentially to do whatever you want.
Ben, what’s your take on the kinds of ideas and websites and businesses that a no-coder should think about building?
I never try and come up with ideas. I think it’s more about the passion projects like I was talking before. If there’s a specific type of thing you're really interested in, there’s usually a community out there around it. I’d say most ideas that people come to me and say, “How do I build this thing without code,” if you break it down and look it down and look at what the moving pieces are, it often looks like a marketplace type app or there’s two sides to it.
The screens may look different from Airbnb or something, but a lot of the times the functionality and the way they are is often transferrable. I think that if you look at lots of apps and stuff that’s out there today, you can repurpose a lot of these. So if you wish there was a Gumroad but for music only or whatever it is, then you could build your own mini version of it.
Like Sahil was saying earlier, I think there are ways to build something, but for X that you are really passionate about. I think there’s loads of opportunities out there. I think you’ve just got to try them out and see which tools work for you and what tools you get on with and which tools you don’t like. It’s a learning curve if you decide no-code or you decide to code. So either way, I think it’s in the process.
Totally. I think the common theme that I’m hearing from both of your answers is the vast amount of opportunity. Sahil pointed out that the no-code space is just so nascent that it’s got a ton of room to grow, and it’s already pretty huge.
But also, the founders operating in the space have a ton of problems that they need solved by developers. If you just start talking to them, you could probably come up with an idea worth working on pretty quickly, and it’s probably going to be one of these infrastructural ideas that have the potential to something big in the future.
To your point, Ben, the internet is so massive that almost any particular niche or hobby or passion you have, there’s going to be a lot of other people you can reach. With the no-code tools that exist today, it’s easier than ever to build something impactful to reach them. All of this sounds super promising. Guys, thank you both for coming on the podcast. Can you let listeners know where they can go to learn more about Gumroad and about Makerpad?
Makerpad is Makerpad.go or @makerpad on Twitter.
Awesome. And Gumroad is Gumroad.com and @gumroad on Twitter.
All right. Thanks again.
Thanks for having us.
Listeners, if you enjoyed this episode, I’d really appreciate it if you’d let me Ben and Sahil know. Ben is @bentossel on Twitter and Sahil is @shl. I would also love to know your thoughts. I know this is a new episode format, but I might be doing more of these in the future.
If you go to Indie Hackers.com/podcast, you can sign up for the Indie Hackers podcast newsletter, where I share my thoughts on each episode. You can reply and tell me if you thought the episode was good, if you thought it was bad, and even share your ideas for topics I could cover or guests that I could bring on.
Did you know Indie Hackers has a newsletter?
Sign up to get insights, takeaways, and exclusive content from each new episode, directly from the host, Courtland Allen.