Today we're thrilled to welcome Wade Foster, co-founder and CEO of Zapier, the leading AI-powered automation platform helping over 2 million businesses, including Suzie, work smarter, not harder. Under Wade's leadership, Zapier has grown from a scrappy startup to a $5 billion company, pioneering no-code automation, and now AI agents that empower teams to move faster and stay ahead of consumer needs. Wade, so great to see you, my friend.
Yeah, thanks for having me, Matt. So, as you know, I am just such a huge fan of your product and, you know, I have so much gratitude for you and the product you've built because it's really allowed me to do so much as somebody who is not a coder by trade, but has, you know, kind of creative instincts and really wants to build quickly in digital. I know Zapier began as a side project. Like, what was the moment when you realized it would become a scalable business and kind of how did this whole thing get started?
Yeah, I think to answer that, I actually have to go to a company that I was at before, Zapier. So I was at this other startup and I was ostensibly hired in to do marketing. I'd never done marketing before. But I was like, 'Hey, how hard could it be? Like, we'll just try and figure it out.' And I worked at trying to get people into this product for I don't know, a year. I learned search, I learned email, I learned social, I learned direct sales, I learned biz dev. Like, I'm trying everything. Nothing's working. Like, it's just not going well. I'm like, surely they're gonna fire me. Like, surely, like, they're gonna fire me. There's no way I'm surviving this. The product I thought was great, like well-designed, well-engineered, super slick experience, etc. But yeah, just nothing happening. And eventually came across the Lean Startup from Eric Ries. And I came to believe I was like, 'Oh, maybe we're actually building a product that nobody wants.' Like, maybe that's the problem. So filed that in the back pocket, you know, just lived experience. You know, fast forward a year or two. We're starting Zapier. We're working on it. Nights and weekends, we're hacking on this thing. You know, we're like, okay, we need like, the product is not very good yet, but we're like, let's see if we can just get somebody to use it and see if it will actually work for them at all. And came across this forum online where Andrew Warner, the podcaster behind Mixergy, was asking for a PayPal Highrise integration. So, you know, shot him a cold email and said, 'Hey, did you find somebody or a product that could do this PayPal Highrise thing? If no, like I might be able to help.' And he responds back and he's like, 'Ah, actually, I don't need that anymore, but like did you actually build something that does this?' And so I'm like, 'Okay, door's cracked. Let's see if I can get him interested.' So, I'm like, 'Well, sort of, like we can hook up PayPal and Highrise.' Then I go look at his site real quick and I'm like, what other tools does he use? And I notice he uses Wufoo, AWeber, and I was like, hm, maybe there's something there. So, I'm like, 'Yeah, and it does PayPal, Highrise, it does Wufoo, AWeber,' and then I list off like two or three other things. And he emails back and he goes, 'Actually, I really need a Wufoo AWeber integration. Do you think I could use it for that?' And so, then I'm messaging my co-founder Brian and Mike. I'm like, 'We got to get Wufoo AWeber. Like, can we do that tonight?' Yeah. And so we're building that that night and shoot him a note and say hey here's access to this thing. It has AWeber, like check it out and see what you see, see if you like it. About a day goes by he emails me back and he's like hey wait can you jump on Skype like show me how this works and I'm like sure of course. So you know jump on Skype and walking him through the setup and as we're going through setting up this app he's having a bad time and it's not his fault. Like the product is just not ready for prime time at all every step of the way. This is a painful experience and I'm taking copious notes, but I'm mostly like trying not to be too embarrassed because this is going rough. But we managed to get the zap set up and we go to test it. So we say, 'Hey, go to your site, fill out the Wufoo form.' Submits in. Okay, let's go to AWeber. Refresh your list. Let's see if it's in there. Boom. Email address is in AWeber. His face lights up. He's like, 'Oh my god, Wade, this is going to save me so much time. How much money do I owe you?' A light bulb goes off on me like, 'Holy cow, like this is a horrible experience.' And yet, because the promise was fulfilled, he's excited about that. And so that then he gets to kind of enjoy the benefits of that automation moving forward.
Totally. It's one less thing you have to worry about every day.
Totally. It's the exact opposite experience I had at the other company. And that moment I felt like if we can just make this easy to set up, if we can make this so that regular folks can do it, sky's the limit.
So you then obviously set out on a mission to do that. What was that process like going from that original aha moment to where we are today where you integrate with over 6,000 different third-party applications? Obviously, I'm sure you had a lot of twists and turns in between. You know, what were some of the milestones that stick out to you?
Well, yeah, I mean, it's mostly just a lot of hard work is what it boils down to. There's obviously like some pivotal decisions that come in the mix. You know, we were working nights and weekends for like six, nine months. We applied to YC and then got accepted, moved to California. So, that was a big pivot point. You know, the first 50 or 100 integrations on the platform were built by us. Then we built our developer platform and now we started having all these other people building integrations. So that was a big pivot point. You know, we start growing the team. We decide we're going to do distributed only back at a time when that was like super weird and not normal. So there was like a lot of learnings around that. You know, we went from being an integrations tool. Zapier used to only have single trigger single action to now you could do these full-on workflows. You know, we went from being something that you could only use by yourself to being something you use in teams and organizations. You know, now we've gone from workflow to AI and agents and orchestrating like all these workflows across the company. And so, you know, mostly every step of the way has just been trying to layer in something new, expand the capabilities, deepen the capabilities, but at the heart of it, it's all just like ease of use. Like you know, you mentioned like how much value you get out of this as a non-technical person who has some like creative horsepower. I'm the same way. And so that was the thing we were always fixated on is like how do we take these technical tools that are so powerful but give access to most people in the workforce. And yeah, I think we've made a lot of headway but I still think there's a lot more that can be done to like just make these tools much more broadly accessible. And I do think AI is going to be like a big unlock for a lot of new use cases.
I definitely want to spend time getting into that. But before we do, I mean two million businesses is a massive amount of companies. So obviously, you know, we were just talking about product market fit and you evolving the product through hard work, but sometimes hard work is not enough to get 2 million companies to adopt it. This is largely a podcast that has listeners in the marketing and advertising world that obviously clamor every day to capture attention and market share of the end consumer and business are targeting. What was some of the secrets to your success that allowed you to capture some scale in the B2B world?
Well, I think one of the things we got right very early on was we had a, at the time, somewhat novel approach to distribution. Now it's been copied over and over again. But we borrowed this tactic from a guy who made bingo card sites of all things. And he'd realized that he makes these bingo card sites, teachers will buy this stuff. And if teachers search for like history card bingo, he could rank number one if he had a page for history card bingo or physics bingo or you know insert like whatever bingo. Now, not a lot of people search for those things, but it didn't matter. He was number one in the rankings. And if he was able to spin up a landing page and got the cost to make a landing page for that low enough, all he needed to do was generate one sale and the cost of that page was profitable.
And then anytime anyone would search for that unique combination or search again, you would be the one there.
Exactly. And so we looked at that and said, 'Huh, that kind of works for us, too.' Like most people are trying to do these integrations at the time. If you went and did a search for like integrating X and Y, the search results were abysmal. Like there was just nothing there for, right? Like it was like, you know, scripts and GitHub repos and pastebins and API keys and just a bunch of stuff that like was not what people were actually looking for. And we're like, if we can give them just like an actual page that describes what they're looking for, a nice sign up button, make an app, like that is going to be the ideal thing. And so, you know, we built out this app directory that, you know, allowed you to set up landing pages for every integration we supported. And so, again, not that many people searched for the individual integration, but enough would. And then in aggregate, it actually worked pretty good. And so, as we added more apps to the platform, we had more keywords we could rank for, more people we could pull in. And then over time, we layered in new channels and new ways, but that set us up on like a really good and efficient foundation.
So, interesting. So basically the use of your product essentially created a roadmap for content that you would optimize via SEO and it was basically micro content for these micro use cases but over time in aggregate you know you would own all these micro use cases and aggregate would own the notion of integration.
Totally. It was so integrated to the core product experience like I think this is often what as an industry we get wrong when we talk about product market fit. I think a lot of folks treat that as like an ordered list where it's like, okay, do the product, okay, then do the market and then make the fit, but it's actually one thing where it's like the best products, you can't distinguish between what's the product and what's the marketing. It's just so integrally tied together. It's one and the same.
I think a lot of companies obviously try to overcompensate for the lack of product market fit with marketing and that never tends to work.
Well, you just, it's really expensive, right? Is what it boils down to. And at a time when you know there is a zero interest rate environment, you saw a lot of companies grab growth because capital was everywhere, but then when capital became harder to come by, well then you really saw where the great products lied.
Totally. So, and so you mentioned you know you made decision to go remote which obviously you know a decision you made before it was trendy. Another decision that you made before it was trendy was a notion of bootstrapping. You know, we went through a world where venture capital funding was everywhere and I know that your company went a very long time without raising outside capital. What was behind that decision and how do you think it helped shape your culture?
There was like a couple things I think. One, we were probably naturally disposed to bootstrapping. You know, we were at this company, Veterans United, that was owned 50/50 by two brothers. You know, they had never raised a dime. When I joined, I was employee 500 and when I left like 10 months later starting Zapier, they were already at 1,000 employees. So, we'd seen what was possible building companies this way. Then, you know, we're from the Midwest. We're not around venture capital. You hear the stories of, you know, oh man, this venture capitalist came in and replaced the founders, fired the CEO, like took over everything. And, you know, we're just like, we don't want that. Like we want to own our company and we want to make it great and you know we want to control our own destiny. So we had that sort of probably like natural disposition. I think the second thing was we also felt like, you know, when because we had the product well suited, we had that distribution curve, money was never the constraint to growth for us. There was always something else that was harder for us to figure out. So that was a thing on our favor. And then the third thing is when it like in software companies the biggest expense you have is people. It's all, it's the number one thing on any go look at any software company's P&L you know and you'll just see like salaries is like a huge huge expense for them. And we had a very different approach to hiring than I think most companies did which was we just didn't hire till it hurt. And part of why we did that is we wanted to make sure that every person that joined the company, there was a very real and visceral need for them. And we knew exactly what task to assign them cuz you'd hear these stories of these companies that would raise a bunch of money and then they'd hire 10, 15 people and then it was like a mess. Like you know that no one knew what each other was doing. They're all trying to figure all this stuff out and you know it's not this thing where you can just take a small company and toss bodies at the problem and expect the machinery to all keep working. And so we were just a lot more, you know, meticulous about like adding new hands to the deck where it was like very clear, hey, if we set two more hands on this problem, we're going to get the output from it at the other side. And so that just naturally kept our expenses pretty low and our output pretty high.
Yeah, allows you to be much more efficient. I think a lot of companies, they raise money, you have to spend it, especially in a zero interest rate environment and you just bring on more and more people thinking that throwing bodies at problems is going to solve them. But often throwing bodies at problems just creates more problems.
Yeah. Well, it's been interesting to see on the other side of zero now with AI, this is kind of becoming trendy. Like I heard this word seedstrapping for the first time like about a month ago. Which when we were doing it like that didn't even exist but now it's a thing where it's like raise one round and then just grow from there. And a lot of more founders are paying attention to it and they're like hm I think like with the AI with agents with automation, with low burn rates fewer headcount like it feels like an actual viable path for a lot of companies in a way that probably hasn't really felt that way for the last you know 10 years in tech.
Yeah. I mean one reason why I believe is that used to be the gating factor of creating a software company was hiring a ton of engineers and now many believe that you can just do it on your own without hiring any engineers. I don't think we're quite there yet but you look at tools out there like Lovable and Cursor where you can you know build at least MVPs for software products just by entering prompts. What is your take on that whole new movement of kind of do-it-yourself software engineering?
Oh, I think it's incredible. I think, you know, we're going to see, I think we're already seeing two really big use cases. One is, you know, just like the age of personal software, where you as a person who has a need, you can just make something for yourself and you can use it for yourself. You don't have to try and sell the thing. You don't have to do anything else with it other than it's just like it's software for me. I needed this thing and so I built it and I'm happy. And I think that's really neat that, you know, you can kind of have that phenomena. I think the second thing that you're seeing is the way software gets built, professional software gets built now is changing. You know, it used to be you would have a product manager that would go out and, you know, talk to customers, talk to prospects, collect all this customer feedback in, form a hypothesis on the market. Then they would sit down with the designer and they'd start to mock up all these pages and figure out what that could look like. And then they would go get the developers around and be like, 'Okay, let's go build a prototype.' And then you know that whole process could take two weeks, four weeks, etc. And then they're finally able to show something working to the customer. Well, now you can do that first part where you go talk to customers, etc., but you can then feed that into an LLM and say, 'Hey, summarize my notes.' Then you can say, 'Hey, create a PRD for me.' And then you can feed the PRD into these tools like Lovable, Bolt, whatever, and say, 'Can you actually go generate a prototype for me?' And the cycle time on that whole thing is one person who just does all the work. And it takes a day, 2 days, etc. And so now you just have short-circuited this loop that every software company has had to go through and just meaningfully reduced the time to value. And so I think you're just going to see problems are just being solved so much faster than they were even a year ago to be honest.
Yeah. I mean let's talk about that. I mean, we've talked, you know, offline just about how my experience as a customer of Zapier has just been elevated dramatically since AI is flowing through the systems. It's almost like the best analogy I can give for those of you who haven't used it is I think pre-AI Zapier was like a really well functioning factory and but there were robots on the floor and now it's like there's humans on the floor like you know it's not as mechanical anymore because now you have the power of AI that can kind of fill in the gaps with real intelligence and because of that the power of what you create is you know 10,000x.
Yeah, spot on. Like I think the real unlock that AI brought to tools like Zapier is the ability to work with unstructured data and unstructured text. You know, before to your point, everything had to be like very mechanical. And so, you know, there was use cases that that's great for. It's like when this happens, do this, do this.
Right. You get a lead on LinkedIn or Google Ads, you want to put it to a Google sheet, it'll do that for you because it's very structured.
Exactly. But the moment you had to do, you know, take this PDF or take this big blob of text and do something interesting with it, it was like, I mean, we'll try for you, but you're probably not going to have a very good time with this type of use case. But nowadays, you can say, you know what, actually, here's what I want to do. I want to take the lead that I got from my site. I want to go do a bunch of lead enrichment and maybe I hit a lead enrichment survey service or maybe I go do a web search, you know, powered by ChatGPT or Claude or something like that and I want to pull in a whole bunch of contacts there. Then I want to take all that contact and I'm going to run it through a bunch of different prompts and generate a bunch of different outputs from that and at the finish line maybe I'm going to send off an email to them or maybe I'm going to input all this stuff into a CRM. And those types of use cases are so common inside of the workplace, but had been a lot harder to do with automation tools prior to AI availability.
Yeah, for sure. And now, of course, here in 2025, we're entering a whole new realm, the role of agents. What does AI agents mean to you? And what are some of the more interesting use cases, especially in the realm of what your company can power?
Yeah. So agents I think are really impactful because you're for the first time we're having like real work problems that you can truly delegate the whole thing to software where the software can make decisions for you. So a good example would be you know if you're using Zapier agents a real common use case is hey I want you to go reply to my emails. Which that's kind of like a crazy like that's not a thing you would probably do before. And now you can set it up and you can say, you know, you can give it rules. You can say, 'Hey, I want you to look at, you know, say a common email that I get is for folks that are looking for a job.' And I can say, 'Hey, anytime someone comes in and asks me about a job, I want you to go look at our job page. I want you to match it up to the job requirement. Then I want you to reply to this person with a link to the job. Can you do that for me?' Great. So now anytime someone emails in with something like that, agent takes over and does everything for you.
I could spend 15 hours a week in Zapier and I haven't even thought to do that. That's just an example. It's the first thing that comes to your mind. There's just so many use cases to unlock. To me, this is where I get so excited because for the last two years, most of the attention has been on, you know, model sizes and, you know, the funding rounds and all the drama, the personalities. The personalities of AI, but I think most companies and even the folks that are like actually kind of on the cutting edge of this stuff, we are only tapping the surface around use cases. And so to me, more of this thing, this needs to happen where it's like, tell me how you're using it. What did you figure out how to do? And I think we're going to see just a lot more knowledge sharing around how this happens, you know, agent marketplaces and you know, whatever you name it. But like that to me is where if the technology stopped advancing at all today, if we just, you know, the servers got shut down, like nothing could happen. I still think that there is enormous amount of value that society has just not yet figured out how to unlock because we just haven't thought like oh I guess I could make an agent for that.
So obviously it's going to have a massive impact on the workforce. It already is. Automation is sometimes another words for replacing people. It doesn't have to be. You know I think the good CEOs will tell you automation takes away the mundane tasks from their employees so they can be more strategic. But not every employee can be more strategic, right? So what is your advice to people that are listening to this podcast in terms of what they should do to futureproof themselves so they can use tools like yours so they can be on the right side of this tidal wave of change that we're all experiencing?
Yeah, I think the, you know, I liken it to any sort of tech transformation. And I remember sitting down to talk to my mom where she works in a pharmacy at a hospital and they used to do everything in the hospital by hand. You know, pen, paper, etc. Computers come along and, you know, hey, everything's going digital. That's how things are done. And so AI to me is just like another important technology shift where it's like, okay, you can be the person that still is like, hey, I'm going to do it pen and paper, but the reality is the whole industry, the whole world's moving to these tools. And so I think the best thing that I would advise to anyone is like just build some fluency with these tools. Play with them, mess with them, experiment with them, figure out what these things will be. You know, right now is like one of the best times to do it because with relatively low investment, you can stand out compared to everyone else in the market, compared to everybody else in your company. Because you can kind of do simple things and people's jaw will drop. They'll be like, 'How did you do that?' I got to say that there's an almost like an arbitrage opportunity right now because the knowledge gap is so wide. There are the 0.1% of people that really get it and understand where it's going and the other 99.9% are just clueless. And if that 0.1% that understand this can leverage it and say well look at I just created this, to the rest of society it's like did you spend months building this and the reality is like no I spent an hour doing this people don't really understand it so it really gives you a disproportionate advantage if you can dive in right now.
