Starting a business? Set your sights on what really matters

Our CEO and founder Ben Dowling was a guest on the SaaS-Story in the Making podcast, where he and host Matt Wolach discussed how he founded and grew IPinfo into a global company, and the importance of figuring out and focusing on the things that really matter when launching a business.

The transcription is below, edited for readability. Listen to the episode here.

Matt Wolach: Welcome to SaaS-Story in the Making. Are you interested in finding and servicing underserved and unserved markets? What can you do with IP information? We are going to talk about all of that and more. I am your host, Matt Wolach, and I am thrilled to be joined by my special guest today, Ben Dowling, Ben, how are you doing?

Ben Dowling: Good. Thanks, Matt. Really pleased to be here with you.

Matt Wolach : Likewise. Ben is the founder and CEO of IPinfo, the most reliable, accurate, and in-depth source of IP address data available. He was also formerly the CTO of Calm.com, and an engineer at Facebook. Ben, tell me what you’ve been doing with IPinfo and what's coming up?

Ben Dowling: There are two main things we are always doing at IPinfo. First, making sure that our data is great quality. That will be a project that will probably last forever. There's always ways we can improve it, additional data sets we can work on. The other piece is how we make it easier for people to access that data. We originally built our business around our APIs. Now we're offering data downloads through different platforms and APIs that make it easier to access the data. Those are the two main threads around all the work we do: how do we build out a data set to make it even better and how do make it easier for people to access that data.

Matt Wolach: Can you just tell me a little bit about what IP address info does?I know what IP addresses are, but that's basically where my knowledge stops. Can you tell me a little bit about what the product does and how it helps?

Ben Dowling: Fundamentally, we're a data provider that provides contextual information around IP addresses. Our main data sets are geolocation. So giving my IP address here: is this in Seattle? Is it in San Francisco? Is it a company? Is this a consumer ISP? Is this a business? We also do carrier detection. Is this a mobile carrier IP? Is it a landline? And then we do privacy detection. Does this look like a VPN or not? We try to provide as much context as we can around IP addresses. People use it for all different sorts of use cases.

If a website visitor is in Germany and you've got their IP address, there's not much context around it, right? You may see a new visitor at your site. But you haven't got any context; they're not logged in. And so we can help provide more context in this situation. You may want to personalize a website. Content could be something as simple as saying, hey, good morning in San Francisco or good evening in London -- that would be very simple personalization. A more complex example is, maybe you're a book publisher selling a book. You want to send visitors from the U.S. to Amazon.com, and visitors from the UK to Amazon.co.uk. That’s an easier experience for the user. Then you know, that instead of having them go and switch, they may not even know that the link is different for different countries.

You could even get down to the city level. Imagine a user experience on a travel booking website. If I went to a travel booking website, it might say, please enter your nearest airport, and you would tell me where you're flying from. Or, you could use IP geolocation to say, we know this guy's in Seattle. SeaTac is the nearest airport. Let's just show him the flights that are leaving from SeaTac and over the next 24 hours. That improves the user experience.

We can provide contextual information that helps with the user experience. Now there are lots of other places that have IP addresses and won't enrich them. For example, a cybersecurity use case would be where maybe somebody said, we've seen five failed login attempts from this IP address. They may want to investigate further. Is this an ISP in Russia that has tried this 10 times before that they want to block at the firewall level? It’s useful for preventing risk and fraud in e-commerce as well. If you know someone's trying to buy this thing and they're trying to ship it somewhere, does the IP address match the region that they want to ship it? If it doesn't, it may still be a totally legitimate transaction, but it's an interesting data point to say, okay, maybe we need to be more careful with the transaction. Maybe we need to do something differently about it. Or if everything lines up, we can say, Hey, you know, this is risk-free, we're just gonna ship this and give the user the best experience.

Matt Wolach: That's really cool. The amount of stuff that you can know about somebody is amazing. Is this something also that maybe a marketing team might observe and say, hey, we've got a lot of people from here, we've got a big focus in this industry? Are they able to aggregate this and see it all?

Ben Dowling: Absolutely. Marketing is another big use case. That could be in terms of collecting data upfront, to segment users and do different things. One use case would be safe sign-up. You may ask your website user, where are you based? Whereas you could also passively collect this from the sign-up IP. Then you could do a mailing campaign. You could send an email to all of your users in San Francisco telling them about your summer event or do a test with users in San Francisco and email and send different campaigns to different locations.

Matt Wolach: Yeah, absolutely. It looks different. Marketing-based use cases as well. That's just amazing. What gave you the idea to start this company? You started as a side project that had no ambitions for growing into a business, so what happened?

Ben Dowling: There were geolocation data sets that were publicly available, but you had to download them, set them up in your server, and jump through a bunch of hoops. I'd been working on a few different projects and had to jump through those hoops a bunch of times. I was in the process of setting up a new one. I think the data got outdated on one server. So I thought, well, I don't want to do this a fourth or fifth time just to set up a really simple API. If I can just pull that from all these different places it's going to save me some headaches. I could save some other developers some headaches if I launch it as a free service. I posted it in various online forums, on Stack Overflow, where there were a bunch of questions about, how do I get the location IP address? So I jumped in and said, just set up this free API. It's made my life a lot easier. Here's how you use it. It's free to use. It's really easy and it might help out those developers and hobbyists that were doing things at a small scale. And it grew very quickly.

It started getting a lot of traction and a lot of users. And so then I thought, well, why don't I try some paid plan and see what happens? And then very quickly we started getting some paid customers, which was great. And then we started to get some feedback about the data. At this point, our whole business was: this data is available elsewhere. We're not a data provider. But we wrap that up in a way that's really developer friendly. It's easy to get set up. It saves you a bunch of effort.

Very soon after that, we started getting more and more feedback. People were seeing, you you've got one location. It says San Francisco, but they’re in San Jose. Or sometimes even worse, it says I’m in the U.S. but I'm actually in the UK. We got more and more of this feedback. Initially we said, it's not our data. We make it really easy to use, but we're not in the data space. But then we noticed there was an opportunity there for us to improve it. We've had people coming to us saying, we love your service. We love your API, but your data needs improvement. And so then we said, okay, look, we've got users that love our product. How do we make the data even better to help them do what they're trying to do?

And then we thought, oh, this will be really easy. We can build our own geolocation dataset. We'll be done in a few months and it will be ready. But it's been many years and it's been a lot of hard work. We've come a long way and the same thing has happened with other adjacent data sets. Initially we're focused fully on the geolocation, and we did a lot of hard work to improve that. We have really world-class geolocation data. Then customers came to us for other things, like, hey, you know, this data's great. You're telling me that this guy's in San Francisco, but actually like we've got some signals that indicate he’s maybe on a VPN. It’d be great to get that context as well.

Back to the e-commerce example, if someone's shipping something to San Francisco, the card’s IP is in San Francisco, it looks like a low chance of being fraudulent. But if they're on a VPN in San Francisco, that totally changes the dynamic. It could be some guy totally somewhere else. So we've built out additional datasets based on a lot of feedback from customers saying, we love what you're doing. We'd also love to know this. We've built up a lot of expertise in the IP address data space. We've got a lot of the broad ingredients. Let’s see if we can create other VPN detection products. We’re innovating on lots of products like that -- carrier detection, all these things, we’ve really grown our expertise around the data.

Matt Wolach: It's funny how once you put out a product or a piece of the product and you get more customers, they come to you with more needs, more requirements, more wants, and you have to build something else. I've always thought when we're building out our products, if we just had this one thing, then we're going to be good to go and everybody's going to love it. Then once we get that, now there are new things that we need.

Ben Dowling: It's an issue because like I said, initially, we thought we’d just build an API that the market for is maybe a hundred or a thousand hobbyist developers. Then, we're in a world of geolocation data, but even with a VPN product, we said, ok, we can detect VPNs. Each product grows and evolves into a bigger and bigger thing, which is exciting. There's more opportunity, but that also means the challenge is trying to stay focused. You have to think about what you’re really good at and whether each new product is going to help not just this customer but hundreds of others? Weighing these opportunities can be tough as well, but it's a good problem to have.

Matt Wolach: A very good problem to have. That's something that I face with some of my companies. How did you guys determine which avenue to take and how do you determine which feedback to act on and which to say, well, we're not going to do that?

Ben Dowling: We’ve made some really great bets, but I think we've made lots of bad bets as well, where we think, this thing will be huge. This customer is going to love it. And there's gonna be hundreds of others. More often than not, we've been right. That one customer does love it, but it turns out maybe it was too niche.Sometimes we've been like, we'll build this, we know this is an important customer to make happy. We'll create it for them, and it turns out to have much wider appeal. It's hard to always be right about that. You can make educated guesses. Sometimes you'd be surprised in both directions. Also, one thing that's really helped us is putting a framework work around our scope.

We've had requests for all sorts of things. I think one thing that has been useful for us is to say, no, we're focused primarily around IP addresses. So there are things that are one degree removed, right? Like the location, is it a VPN? Is it something else? We can imagine there'll be other data sets that we might launch in the future that give more context around IP addresses.

We don't want to get too far removed from that. We want to be able to cover a broad set of use cases. That helps us leverage what we're already good at. There are ways we can build out the data sets and expertise we have, and it gets easier each time. There could be a big opportunity, that maybe doesn't overlap a lot with our existing skill sets or data sets. So that's been a useful way for us to evaluate opportunities. Could it be useful for lots of customers, or, is this something that we could maybe do, but it distracts us from the core thing that we're focused on.

Matt Wolach: That makes a lot of sense. I think it's funny how you started as a side project and then found out, wow, this thing actually has some legs. I've heard that story a few times. When did you realize this could be your primary business?

Ben Dowling: It’s interesting. Initially there was no expectation. I've launched side projects that I’ve thought were going to be huge and nothing happened, whereas I thought this one would be small and it's been more successful. I had very low expectations. It was useful for me. I thought it could be useful for other developers. I didn't think it could necessarily grow to be a big business and that wasn't the ambition at the start. There are a few inflection points. One was where I started charging customers and they started paying for it. I realized this isn't just like a hobbyist thing. That was a big inflection point. But still then it was like, well, this will be a side project that makes me some money. It was this great, easy to use API that worked and it didn't need such any maintenance or anything. Then there was another inflection point, when we signed up the first 10 customers that wanted to pay us. We thought that was it, and it would cover the cost. But then it continued to grow, and it didn’t taper off. And then was like, wow. Then we start getting the feedback, and people saying, we love what you're doing. Can you improve the data? There was a big opportunity to provide context around IP addresses in a way where the data quality and reliability are really important. And that was a sizable enough opportunity that warranted full-time attention, and building a team and building a business around that.

Matt Wolach: How did you go about growing the business in those early days? What were some of the best things that you did along your path that really helped to get to where you are now?

Ben Dowling: The big initial inflection point was posting about it to Stack Overflow. That really helped get the word out. And initially that wasn't intentional. I built this service that was good for me. I'm a member of the Stack Overflow community. I ask questions, I post answers. I’d say my service will help with this. It wasn't intentionally a growth channel for us, but it quickly became one. We would start to see people come and sign up and say, I found about you on Stack Overflow and when people Googled, how do I do this? We would be the top answer saying, use our service. That drove most of the initial traction. Then I became intentional about it. I was like, how do I actually go out and find Stack Overflow questions that I can answer? Initially I answered the top two or three questions around this. Then I would go and find 10 more and I got more and more niche. So from, how do you do IP geolocation, to, how do I detect the country of the user? There was a long tail of questions and I spent a long time going through and making sure that we had answered all of them.

We got to the point where I had answered all the Stack Overflow questions that were applicable to geolocation, but then there were questions saying, how do I get the ISP of the user over an API? I wondered if we could do that. We would ask existing customers, would this be useful for you? And they’d say, yeah, that'd be great. And we’d build it and answer on Stack Overflow. And that would bring a new influx of users.

Matt Wolach: That's really cool. I love that you really pointed out one of the things that I talk about with my clients. Figure out where your market is. Where are they congregating? Go there and start interacting with them. You did that. Your market is developers. Developers are on Stack Overflow. You went there and showed what you could do to help them. You didn't intend it to be a business at the time, however, that's the exact right way to do it. And that's why it took off for you. I want to ask you now about the flip side, just to make sure we can help some of the people out there who are building companies now. What are some of the steps you took that you wish you would've done differently as you were getting going?

Ben Dowling: It's a great question. It's hard in hindsight because I'm pleased with the trajectory we've had on places where we are. One of the most challenging things in growing a business is trying to figure out what's important that you need to focus on, and what you can worry about later. Because at first, it feels like almost everything's important. There are things that I think people worry about in the early days, that maybe I worried about: We've got to get our corporate structure set up, or we've got to dial in our pricing plan, or we're closing this deal and we've got to make sure we get all the terms.

There are so many things that ended up being a huge waste of time. I remember we closed a customer in China early on. A lot of our customers are self-serve, and this was one of the first enterprise contracts. So this customer said they needed the “chop” on the contract. And I said, what's a chop? It turns out, it’s like a corporate stamp. I thought, I’m not a real business if we don't have the chop, everybody's going to want this chop. I literally ordered a chop for the business. I spent weeks finding out where I can get a chop. I had to pay to FedEx all these documents to China and thought that was really important. No one's ever asked for it since.

It got to a point where I would push back on these demands and be say, we don't have a chop. And people said, no problem. It's amazing how many things are negotiable that you don't even realize, and as I got more confident in doing these bigger deals to say, no, we don't do this.

That was an interesting lesson. I speak to some guys that are starting business where they'll ask, where did you get all your website terms and conditions? I'm speaking to a lawyer. I want to get all this done before I launch my website. But, no one's going to care. These things seem really important, and there are things that are really important, but it's ultimately, are you delivering value to the customer? One thing that served us well was focusing on making sure we have really, really great data, because that's what ultimately helps the customer. Let's make sure our API is fast and reliable.

If we're spending all our time on those things, we're probably doing the right stuff. Time spent chasing down chops and sending contracts -- if people love your products, they're willing to make concessions and say, I can get it past the purchasing guys. There's lots of things that we wasted our time on that we didn't know. But now when I talk to people who've started a business it's easy to see some of these same mistakes. It's like, don't worry about the chop, don't worry about getting your terms and conditions perfect on the website. That stuff you can always fix later, after you make sure you've got something that really delivers on the value that your customers want.

Matt Wolach: I love it. So if I could package up what you just said there, essentially what would have really helped you is somebody who's been there, done that and would say, you need this. You don't need that. Just do this. It sounds like that would have made things so much easier for you in those early days. It sounds like you're helping others do that now.

Ben Dowling: Absolutely. If you've got someone to ask, that's a great way to do it. If you don’t, think about: is this part of your value for the customer? If someone's saying to me, your data sucks and you need to improve it, that's something I absolutely need to jump on. If someone's saying, do you have terms and conditions on the website? Generally customers don't ask those kinds of things. They care about solving their own problem and how you can help with that. A lot of these other things ultimately tend not to matter.

Matt Wolach: Yeah. I help software founders figure out what they need to do and what they don't have to do. I wish that I had had that when I was starting my software companies because I wasted so much time on things that never mattered. If I’d had somebody to ask it would have saved me years of struggle. Ben, this has been really cool. I've learned a lot. Thank you so much for coming on the show.

Ben Dowling: Thank you. It's been great.