Wade Foster

This week, I chat with Wade Foster, co-founder and CEO of Zapier. In this episode we talk about the business impact of features, how to make use of partnerships, the benefit of Y Combinator, managing a huge fully remote team, content marketing, and more!

Wade: Hey, what’s up, Josh?

Josh: Wade, how’s it going man?

Wade: Pretty good, how are you?

Josh: Cannot complain at all.

Wade: Awesome.

Josh: How are things?

Wade: It’s going well.

Josh: All right. Are you in the Bay area? Did I read that?

Wade: I am, I live in Sunnyvale.

Josh: Oh, okay. Cool. What’s the weather like there right now? I assume fantastic.

Wade: Yeah, it’s pretty much awesome all the time.

Josh: You were in Missouri.

Wade: Yeah, I was born and raised in Missouri. I moved out here to the Bay area about 4 years ago, so I’m familiar with seasons and things like that.

Josh: Then, you promptly abandoned them and just … perfect weather year around.

Wade: Yeah, exactly, right? If I want to go see snow, I can go do that on my own time and pick it.

Josh: You dictate what the weather is.

Wade: Exactly, right?

Josh: Well, cool man. Business is good?

Wade: Yeah, it’s going really well. We’ve had a pretty good year. We launched multi-steps apps back in February and it kind of changed our trajectory a little bit, so I’ve been trying to just keep up with that.

Josh: Yeah, that’s actually one of the first things I wanted to talk to you about because I know, for us, when that specific feature came out, it changed how we used Zapier, right? We had kind of used it for random like one off things, and then within a week became this thing that we were using … I mean, we’ve probably tripled or quadrupled the number of thingsj that we used it for.

Wade: Yeah, that’s great. The same happened to us internally, too.

Josh: From a business standpoint, you mentioned it sort of changed your trajectory, is that just the number of businesses that use you guys or how much people use you?

Wade: Yeah. We put together a really marketing launch for it and we got the top app in a category, so we found the best CRM company, the best event management company, the best marketing company, the best note taking app, like each one of them to help promote multi-step launch as part of the launch. That ended up being companies like Evernote, and Todoist, and Trello, and Pipedrive, and MailChimp and Slack, all these people were writing about it on their sites, sending e-mails to their user base, all that sort of stuff. That brought in just a huge number of new people and basically, overnight, our daily sign-up increased by 70% or 80% and it stayed high. It wasn’t just a spike that then went back down, which was kind of what I was expecting. It just stayed up there, and it’s kind of been growing. Plus, it was a little in the summer months, but now we’re back up to seeing more record sign-up days, and so, we weren’t really prepared for it. We knew it would be a good launch, but we weren’t sure how good, right? I’m always skeptical that any individual feature would really have a huge impact, just because that’s what I’ve seen historically.

Josh: Right. Was the –

Wade: But, this did.

Josh: Was the multi-step stuff one of those things where it’s been on the roadmap forever and ever and it was such a big undertaking it took a while, or was it sort of like, “Hey! We should do this. Okay, let’s do it,” and then you knock it out really quick?

Wade: It was definitely the former. It was something we had heard from customers from pretty much the beginning. We hadn’t really architected Zapier in such a way to allow multi-steps apps, so it was a really big undertaking to … we had to change architecturally how we run Zapps from an infrastructure standpoint. That took a while. Then, we had to change the UI as well to accommodate for it. We used to have this whole left to right thing going on which didn’t really make sense in a universe where you could have infinite … not really infinite, but you could have many, many steps. In this app you don’t … people don’t scroll left to right, so we had to accommodate for that. Then there were just a whole bunch of other random things that you don’t think about like support tooling, little marketing sites, and help docs, and things like that that all have to get updated for this new version of what the product is.

Josh: Right, right, right. Did it require any of the integrations to do any updates on their end to make it work, or …

Wade: No, no updates to make it work. They all worked just as is, but we did add a new action type, search actions which is something that a lot of apps can take advantage of so they could improve their integrations, so a lot of what we’ve been doing, pre-launch, was getting like a handful of folks to do that. Post-launch has been kind of circling back to a lot of our partners and trying and encouraging them to add search actions to their app because it’s a good thing.

Josh: Sure. You mentioned, as part of rolling out the multi-step stuff, that you basically partnered with integrations that already existed, basically, so do you guys do that much when you’re making use of these built-in partnerships to launch stuff, or was this the first time you sort of wide-scale …

Wade: That was the first really big one, kind of where we connected the dots between a whole bunch of our partners. Typically it was like this app plus Zapier and that was the focus of it, which kind of limits … you’re naturally constrained to the size of our user base and the size of the partner’s user base, in terms of who you can talk to. But when you can … Something we realized was hey when we can start to pull in multiple parties we get more unique things that we can talk about and it engages a much broader audience.

Josh: For somebody who’s … what’s the biggest barrier that somebody has to using Zapier? Is it understanding it, or just being convinced of the value?

Wade: Well, there are kind of two types of users that we have. The first is someone who knows they’re looking for some sort of integration or automation, right? They’re like, “I need a sales force and MailChimp integration.” Those type of users find Zapier and have no problems getting set up because that’s what we do, right? It’s like, I have this known problem and there’s a known solution, and I set it up, and I’m gravy.

The ones we struggle with are … as the Zapier brand has grown, there’s a lot of word of mouth now people saying, “Oh, Zapier’s great. It automates all sorts of stuff for me. You should check it out.” Well, that person goes to Zapier.com, they see it, and are like, “Okay, this looks kind of cool. I guess I could do something with this,” and then they sign up and there’s like 700 apps they can hook things together with. They’re like, “Where do I even start with this thing? My buddy told me it was cool and would help me out, but I don’t really know where to start.” That’s the type of user that we struggle with and we’re trying to improve the onboarding experience and all that stuff for people who are just new to a lot of this.

Josh: Yep, yep. When somebody signs up, that whole barrier to entry of figuring out what they can do with it, what’s … other than just educating them is there anything else that you guys can do to basically, convince them of the value or speed up that stuff? What stuff have you guys tried?

Wade: Well, what we’ve got running right now, which is kind of the best variant we’ve stumbled across is like an in app picker. When you sign up there’s a big window and it says like, click on the apps that you use, and kind of has them sorted by popularity. The most popular apps are at the top and hopefully they’ll see a few apps right out of the gates that they recognize the logo for, and they’re like, “I use that. I use that. I use that,” and then from there we narrow the set of potential Zaps they can set up from there. There’s like a pre-baked simple one that they can jump into. That’s worked the best. I do think there are still a lot of things that we could improve upon with their, but the challenges, they become increasingly more sophisticated from an engineering standpoint, and we’re not sure if they’ll work. That’s the challenge.

Josh: Sure. Sure. Early on, presumably, you guys built a bunch of the app integrations yourself? Correct?

Wade: Yeah.

Josh: Do you guys still build any of the integrations or is it …

Wade: Very rarely. That was kind of one of the best things, I think, we’ve done was about … we built 60, 70 integrations ourselves, and then we launched our developer platform which allows other people to build apps on Zapier now. In doing so, we kind of realized there is a lot of demand from saas partners to solve the integration problem. They don’t have the engineering power or resources to build out dozens of integrations. There was a lot of motivation from them to plug into one place and be done with it, right? Zapier, with the developer platform, we offered a way for them to do that. A lot of folks just jumped right on it. We had HubSpot, Podeo, and Gravity Forms, a whole bunch of folks right out of the gate. Heck, Slack when they launched their beta version in 2014, that was one of the first integrations they built was a Zapier integration. Since then, it’s mostly just been partners building apps on Zapier. About the only time, we do it is something huge, like, we built the Excel integration recently, or if we hire a person as a kind of get to know Zapier project, they’ll build an integration themselves just to kind of learn how it’s done.

Josh: You say 60 to 70 integrations like, meh, but that’s insane. We’ve got essentially 3 integrations Stripe, Recurly and Braintree, and it’s a nightmare, right? Just having things continue to work. I cannot imagine.

Wade: But, that’s what we do, right?

Josh: I know. It’s a lot to manage.

Wade: It is. Ours are probably simpler than yours, though. You have to do all this … you’re reconciling the difference between all this stuff, and accounting for all this stuff. You’re like a pure analytics platform.

Josh: Right. Right.

Wade: Zapier, we’re shuffling data from one side to another.

Josh: Do you guys have to deal with that? I mean from a support standpoint, do you spend much time dealing with … even if it’s not your fault, just some obscure app who like changes, is that a decent amount of support load? Or …

Wade: Not changes. Changes are usually not too big of a deal. Downtime is usually what we end up dealing with. I know the Wunderlist team is really struggling this week with some tough downtime, and, so that creates some support tickets for us as well, because users are like “My zaps aren’t working. What’s going on? Why am I getting this error alert e-mail?” It’s like, well, Wunderlist is having some problems, so …

Josh: How do you approach that scenario where it’s not necessarily your problem, but not just throwing the integration partner under the bus like, “Meh, their problem. Too bad,”?

Wade: Yeah, it’s tough, right? Because the approach we have is we’re all in this together. It’s to everyone’s benefit if Zapier stays up, and they stay up. I mean, nobody wants downtime, right? I think that’s just generally accepted, so I think if you come in there with that approach, that like this wasn’t something … Wunderlist isn’t over there being like, “Oh yeah, we’re just trying to screw Zapier over by having down time,” right? Not at all what’s going on. They’re over there going, “Oh, crap. Oh, crap. How can we get this up right away?” If you just come at customer service with that approach and be like, “Hey, we’re all working on this together. We’re just trying to get this back up and running,” I think that helps.

The other thing we try to do is try and provide some transparency, a little bit, in the messaging. We haven’t baked it all in yet, but if you go to status.zapier.com you can see a whole list of basically logs of outages that have happened with other apps and it just tries to reduce the burden on our side. Like when something like this happens, they can check the status board and see, “Okay, it’s not just me. This is a system-wide issue with this app.” We’ve got a message there that says, “Hey, should you worry or should you not? When Wunderlist comes back up everything’s going to be gravy. Zaps will just be delayed. There’s nothing you need to worry about. Just hang tight with us.”

Josh: Sure, sure. Cool. What were you doing before Zapier?

Wade: I was a … I did marketing automation, technical marketing, some SEO work at a mortgage company in Columbia, Missouri, and then a bunch of freelancing and stuff, too.

Josh: What kind of freelancing? Like engineering or marketing stuff?

Wade: Anyone that paid in dollar bills.

Josh: Do whatever.

Wade: Yeah, right. WordPress stuff. I did some e-mail like … quite a bit of e-mail design copy writing, stuff like that.

Josh: Got it.

Wade: Things like that. Just whatever. I was maybe a year or two out of college at that point. I’ll take any job that involves the internet because I want to get experience with anything right? That was kind of my approach, and got a pretty broad understanding of a lot of different things, in doing that.

Josh: You mentioned going to college. What’d you go to school for?

Wade: I have a degree in industrial engineering, kind of like an efficiency degree, but it’s mostly applied in manufacturing facilities and things like that. I think the most well-known is Toyota, if you’ve ever heard of the Toyota Way, and all that stuff was kind of popularized by industrial engineers that worked at Toyota.

Josh: Yep. Did you in any way, form or fashion put your schooling to use?

Wade: There is some transferable stuff, but not really. The core technical skills not too much. The learning how to think about stuff, yeah.

Josh: Sure. Sure, do you think from … What’s your opinion on higher education?

Wade: I think it’s pretty important, but I think the thing I wish I would have done earlier, in higher education, was have a job alongside of it. I wish I would have done more of that. I think I started … like an actual professional job, not just like a flipping burgers sort of thing. I wish I would have jumped into professional jobs quicker. I started in my junior year of college, and that’s with a professional internship sort of thing. Having that really put the education piece into perspective for me. It really helped me just in deciding which courses I wanted to talk to, which professors I wanted to talk to, what was BS, what’s not BS, and that really helped guide it in terms of what I wanted to do. I think that was something I wish I would have done differently, I guess, if I could do it all over again.

Josh: Yep. Yeah, I went to school for graphic design, ultimately, but it took a couple of years of college before I even landed on that. My internship was at this print shop, and it was miserable. It was one of those things where I quickly within a week of being there, it was likely, no way am I ever going to do this. I do think that a lot of kids miss out on … there’s this big disconnect between education and practical application of that and they go … Your head’s down in the books for like four years or however long it takes you, and then you kind of get spit out and you’re like, does this have any actual use in real life? Yeah, so you went to school for industrial design, then you kind of did some random odd jobs, and stuff, and then, how did that lead into Zapier?

Wade: I had, right out of school, I had worked at a real small like a web shop. I guess you might call it a start-up. They had some tech products. They did some tech consulting, a lot of random things. That was where I was like … I kind of talked my way into that role, and I really loved a lot about it. It provided me a lot of room to just mess around with stuff because they didn’t really know what they were doing, so I could just experiment with all sorts of different things. I kind of realized this software as a service thing is pretty cool. This is back in 2009, 2010. I was using tools like MailChimp, Wufoo, and stuff like that. I was like, “Man, this stuff is really fun. I want to be able to make stuff like this.” I started teaching myself how to code and stuff. I started to learn some python, some HTML, CSS, and some JavaScript and things like that.

There was a guy I’d known from school, the only person I knew who was doing this was Brian Helming. He was making all these really cool apps, and sites, and stuff. I pinged him because I knew him a little bit from some music classes we’d had together, and I was, “Hey, you want to help me learn some of this stuff? Maybe help me build some of this stuff?” He was like, “Yeah, I’d be happy to. Do you want to come play in our band?” Because they had a jazz trio that they needed a saxophone player, and I’m a saxophone player. So, we kind of traded. I was like, “I’ll play in your band and you help me learn this stuff, right?” Brian and I got to be pretty good friends. We

We worked on freelance projects together, and we often got asked, as a part of these freelance projects, to build these kind of one off integrations for folks. Get PayPal sales into QuickBooks for us, or get this list of leads into sales force.

One day Brian messaged me and was like, “I think we could probably build something that lets people do this out of the box where they don’t have to hire us to do these one off things. They can just set it up, plug, and play.” I was like, “That makes a ton of sense. Let’s do it.” That was basically the start of Zapier.

Josh: I guess at what point did you think, did you make the transition from, hey this is a cool problem to solve to hey we can make this a business?

Wade: Yeah, I think we kind of wanted it to be a business all along. I don’t know when it was obvious that we were going to … It took a while before it was like, yeah, this is going to be a thing because we’d worked with Mike, our other co-founder, teamed up and we were working nights and weekends on this thing for maybe like 6 or 9 months or so, and I would say, we started in October 2011 and by December 2011, we had a customer, so we got Andrew Warner at Mixergy paid us to do a thing. We’re like, okay, this might be a business. By January or February of that year, we had a working beta that we could let people into. We got a few more customers until we had a pretty growing list of people who wanted to sign up, that we weren’t quite ready for yet, so we’re like by about February of that year, we’re like we think, “This could be a real thing.” We applied to YC that year as well, and by the time we got accepted and got in, we were like, “All right, we’re not making a ton of money, but yeah, this could definitely be a real business.”

Josh: Got it. You mentioned Y Combinator, what was the main motivator to even apply to something like that?

Wade: Well, you know Zapier was still nights and weekends thing. We were really grinding pretty hard. We had … we were building all these integrations ourselves, but we didn’t have a ton of relationships with the vendors themselves yet. YC was kind of a shortcut for us to building those relationships. It was like a badge of approval, a little bit to get in the door with some companies that were tougher to get in the door with. I remember, we were having a tough time building a relationship with Sales Force. We get into YC, and one e-mail from PG later and, you know we’re talking to some VP and problem solved.

Josh: Done. Right, sure. Sure. At that point, it was just you and your two co-founders?

Wade: Yep.

Josh: Then, you just stayed in the Bay area, is that …

Wade: Yeah, so that first year, Brian and I were in the Bay area and Mike moved back to Missouri after YC to be with his then girlfriend, now wife, was finishing up law school, so he was in Missouri for a year. That’s kind of when the remote aspect of Zapier kind of solidified, I guess.

Josh: Yeah, yeah, yeah. This kind of sort of fell into place, the whole remote thing or had you even really given that much thought?

Wade: We hadn’t really given much thought to it. Since Zapier was a side project originally, we operated like a remote team even though we were in Columbia, Missouri, all three of us, because when it’s a side project, you just work when you have time to work on it. You don’t wait to be in the same office to draw on a whiteboard together. You just do what you can get done when you can get it done. We kind of already had that ingrained in us a bit. The fact that Mike wanted to go back to Missouri, it’s like he’s a vital part of this company. We’re going to do it this way or we’re not going to do it at all.

Josh: You stay in the Bay area. Is there a business benefit to you being there or is it just, hey we love the area, so we’re going to …

Wade: I think, initially, there definitely was, like being close to partners and things like that certainly helped for a few things. Now, I think we could be probably anywhere. I think we’ve built up enough of a brand and a reputation that it wouldn’t require a Bay area presence, but we love the area for the most part. The weather’s great. My wife loves it out here, so we’re here for the time being.

Josh: Sure, sure. Do you guys have an office at all, or are you literally 100% remote?

Wade: We are literally 100% remote, no office, or anything. I work from the kitchen table most days.

Josh: Got it. Right on. Last topic here. Let’s talk about marketing a bit. My hunch is that content is the biggest source of new leads for you guys, is that correct?

Wade: Well, yes and no. We have two things. One is our partnerships drive a ton of traffic to us. The 700 apps all have like App Directories, and onboarding e-mails and things like that that send a ton of traffic for us and those are super targeted, so that drives a lot. We have the landing pages as well, the directory that has a lot there, and then, of course, our blog and learn resources drive a pretty substantial amount of traffic as well. I’d say those three things in tandem plus, word of mouth make up the bulk of our leads and sign-ups.

Josh: Is the integration, individual pages … was a big motivation to do that purely from a search engine marketing kind of standpoint?

Wade: Honestly, it was mostly to figure out … it’s kind of an art … it’s kind of an artifact. It does help us search now, but it was an artifact of the fact that we wanted to figure out what we should build, so we built this like directory. Originally, none of them were supported. It just said, vote to have this integration, right? When we were really, really young we would use the votes as a way to figure out, “Okay, we should add this app next.”

Josh: How would you send people to those pages to even get a vote in?

Wade: They would mostly come from … we would go to like forums and stuff, and drop in links and stuff say “We’re working on a project. If you want to learn more go here,” and then they would get indexed, and we’d get search traffic and stuff from that as well.

Josh: Got it. Is it … I’m poking around here. Does each zapp, like each what do you guys call them? Recipes?

Wade: They’re like zapps, zapp templates.

Josh: Okay, so, there’s got to be just thousands of these.

Wade: Oh, yeah, there’s a ton.

Josh: Man, crazy. Okay, let’s talk the content stuff a little bit. You guys pump out a ton of stuff just on the blog. Do you have any sort of system in place for getting content out there or has it just sort of been this slow and steady wins the race and you’ve built up this big following of people who read everything?

Wade: Yeah, there’s definitely a strategy to it, and it’s definitely slow and steady wins the race. A long time ago when we were starting to build out the blog, this was probably 2013, I was kind of practicing along, and I was like, “Okay, we need a cadence.” It was like once a week was kind of what I was aiming for. I didn’t always hit it. Then, when we brought on Danny, who now runs our marketing team, we were like, twice a week, let’s get something out. Tuesday and Thursday. I don’t know how long this streak’s been going for, but we have published every Tuesday and Thursday for years now. Yeah. We’ve cut it close sometimes. I know, at least, once where it was the night before, it got pretty late and there was nothing ready yet. I remember Danny was just like, “I’ve just got to buckle down. We don’t want to break the streak.” Right? It’s gotten to that point now we just do not want to break the streak so much that we are going to figure something out for the site because the streak has taken us so long to build that streak up.

Yeah, there’s definitely a strategy for it and over time it’s evolved, and we figured out what works and what doesn’t work. For us, it’s all about where apps meet productivity, so it’s like this combo of here’s the best apps for these kinds of tasks, and here’s how you can do them to do pretty awesome things. That’s kind of the bread and butter for us is just talking about the apps that are on Zapier and how you can get to do really cool stuff with them.

Josh: Then, how do you make sure that people keep coming back? I’m just poking around on the blog. There’s not this big e-mail sign-up presence. It almost seems like you’re pushing more people to try Zapier instead of sign-up for the newsletter.

Wade: Yeah, we get a lot of people signing up for the newsletter, but yeah, we try to push them to sign up for Zapier, too. The call to actions change, as well, based on if you’re logged in and you’ve already signed up, we’ll swap it to an e-mail sign-up form instead of a “Sign up” because you’re already signed up. So, we move the CTAs around and we experiment with those. The e-mail list is pretty sizable now, which brings folks back. Then, of course, we’re targeting various search phrases and things like that to draw people in. Each post kind of targets a specific keyword or two that we’re honing for. Then we get a fair amount of people just sharing stuff on their reddit, or Twitter, or Facebook, or whatever that drives people in, too.

Josh: Got it. For you guys, at this point, there’s always a balance between quantity and quality, but it sounds like if you can be really methodical about the quantity part especially with you targeting specific things and that kind of … there’s some inherent quality that comes along with that because you’re targeting things that people are interested in.

Wade: Yeah and we really … the quantity is, for us, every Tuesday and Thursday. We talked about trying to add another day, but we’re not going to add another day if the quality’s going to go down. The quality has to stay where it’s at because that’s why the blog works. That’ why the learn resources work so well is because it’s actually detailed information that’s well researched. It’s not like 500 word think puff pieces. This is real meaty stuff. If you read our blog you know. If anything it’s too meaty. It’s dense, and it’s like you have to set aside time to read something because that’s what we go for. That’s what works for us.

Josh: Yeah. What do you spend most … obviously, when you’re CEO, the job changes day in and day out, but what do you find, currently, you’re spending most of your time doing?

Wade: Hiring or recruiting has –

Josh: That’s a full-time job.

Wade: Yeah, has kind of been my number one for the last … I’d say 2016 that’s been what my main role has been.

Josh: Yep. How big’s the team right now?

Wade: A little over 50.

Josh: And are you … do you have “X” number of jobs that you’re trying to hire for right now?

Wade: Yeah, we have like a hiring roadmap. We have, I think, 3, 4, or 5 jobs listed on the site that we’re trying to bring in. We’ve got a roadmap that we’ll probably finish the year around 60 or 65 people, if we kind of get what we want to get in.

Josh: Sure. Sure.

Wade: That’s one thing that’s changed for us, too, is figuring out how to predict what our people needs will be based on the size of the user base is growing. Trying to get better with that.

Josh: Right, right. It takes a long time to hire people, right?

Wade: Yeah, right? Like, if you’re talking about support, right, it’s very methodical. It’s like you see the number of support staff is tied to how big the user base is because that is what dictates how many customer tickets we’re going to get and it takes us, at minimum, probably 6 weeks to publish a job listing, get people through interview process, get someone hired, and then to put in their two week’s notice. Then when they start, they’re not effective right away. It’s still going to take another month or two months for them to even be like effective at anything, not alone like good. You have to plan for and accommodate for all that stuff, and if you’re growing fast, you can’t just wait. Otherwise, you’re going to put a whole bunch of stress on your support staff. Then they’re going to quit and then you’re going to be in an even bigger problem.

Josh: The cycle just gets longer and harder.

Wade: Right, so you’ve got to get good at this sort of thing which is something we’ve been learning this year, for sure.

Josh: Yeah, yeah. Cool. All right. I think that’s all I’ve got. Before we wrap up, how can people get in touch?

Wade: Yeah, I think probably the best way is, I’m pretty active on Twitter, so @WadeFoster you can find me, or you can fire off an e-mail wade@zapier.com.

Josh: Right on. Good deal. Well, thanks a ton for hopping on the call, Wade.

Wade: Yeah, this was fun, Josh. Thanks for having me.

Josh: For sure. Take it easy.

Wade: Cheers.