The accidental history of Niantic’s Pokémon Go, as told by John Hanke

John Hanke put in his time evangelizing location-based gaming and augmented reality. He spent six years refining Ingress as the first AR game from Niantic Labs, which was incubated by Google. So it’s nice to see that work pay off with the success of Pokémon Go, which has generated an estimated $200 million in sales since its launch on July 6, according to Sensor Tower.

Hanke now officially qualifies as a visionary, or someone who sees something that no one else does. He was willing to undergo derision and skepticism during the hard years of investing in Ingress. Now he has put AR on the map, and that notion of AR being a $90 billion business by 2020 (as predicted by Digi-Capital) doesn’t seem as crazy anymore.

Before starting Niantic within Google, Hanke started Keyhole, which was acquired by Google in 2004 and became the foundation for Google Earth. As for the future, Hanke said he is a science fiction fan. “I want contact lenses that transform everything into whatever themed world I want it to be,” he said.

Don’t we all. Hanke spoke at our GamesBeat 2016 event last week in Los Angeles in a fireside chat moderate by Paresh Dave, reporter at the Los Angeles Times.

Here’s an edited transcript of their talk. You can also watch the video embedded below.

GamesBeat: You started at Google working on Earth and Maps and other products. Then you came up with this game called Ingress. What’s the inspiration for that?

John Hanke: I was lucky. I helped start a company called Keyhole that Google bought in 2004. We worked on this earth viewer project, which became Google Earth. I worked on Maps and Earth and those related projects for about six years at Google. That group grew from a few dozen people, about 30 people in 2004, to more than 1,000 engineers and project managers and a couple thousand contractors working on Google Maps in 2010.

I wanted to go do something a bit more entrepreneurial. I had been at several startups before coming to Google. We were thinking about how we’d built this map of the world. We’d helped bring maps to the Android and iPhone, GPS. I grew up as a gamer. People seemed to like to do social things with games. It seemed like there was something interesting at the intersection of maps and mobile phones and gaming.

We formed this little group. We got permission from the higher-ups at Google, primarily Larry at the time, to give it a shot. We started trying to create something interesting in that area. We did Field Trip first, which is a product that’s still out there. It’s about discovering information about historical sites and cultural things in your immediate vicinity. Then, as that product was launching, we started working on Ingress, which is the precursor to Pokémon Go. We incorporated some concepts around location and multiplayer, discovery of local places, and put that out there.

GamesBeat: You described your company’s mission as involving exploration and exercise and real-world social interactions. Those don’t sound like the mission statements you usually hear from a game company. Where does that mission come from, and how does that jive with being a game company?

Hanke: We’re definitely about fun and adventure, but we were in this interesting position of starting this up inside Google. The mentality within the company was, you don’t just copy somebody else. You try to do something innovative. We were looking at the areas we wanted to explore and thinking about what we could add to the universe of products out there that other people weren’t doing. We weren’t really interested in trying to copy somebody else’s success or copy somebody else’s formula. We wanted to explore a new area and see if there was something worthwhile in it.

We came up with a list of — I don’t know what you would call them. Our philosophy of things that would guide our decision-making about what products we wanted to build. One of them was, “If it’s a game you have to go out and move around to play it. You have to go outside and visit new places.” We came with an idea about seeing the world with new eyes, I think is how we put it. But the basic notion was there’s a lot of cool history and lore and unknown secrets in your own neighborhood that you don’t know about. If a product could help you discover those, that would be a good thing.

As a parent, there’s this idea that we wanted to build a product that would encourage exercise, but not in a heavy-handed way. I was struggling with the whole screen time issue with my kids. Trying to pull people away from all these compelling products that we’d built. That was another thing we were trying to achieve.

And then social. We had flagged this idea of “real world social.” This was a post-Facebook world where online social had exploded. People were making connections like wildfire online, updating status, doing all that, but translating that into the real world — we felt like there was an opportunity for a product to help people connect physically, offline. We didn’t know exactly what form that would take, but we felt like there was an opportunity to unlock something there. We had those as our guide rails for the product and we took it from there.

GamesBeat: When it came time to leave Google as they underwent restructuring last year, what was your strategy? Why did Pokémon become the partner you wanted to work with? Why did you mostly take money from strategic investors as opposed to just venture capitalists at that stage?

Hanke: There’s a lot that went into that. To unpack it, we had launched Ingress in 2012. The game was growing. I’d describe it as a cult hit. It’s had 15 million downloads. It was thriving and continues to grow now. We felt like we’d learned a lot about building a game that followed those guidelines we’d set for ourselves — exploration, discovery, exercise, social. We were thinking about, well, how do we package up the parts of that that are working and take it to a wider audience?

We had considered a variety of options. Different internal ideas, totally internally developed IP. We also looked across the film and gaming world at IP that might be a good match, where we could combine a well-known property with the type of gameplay that we wanted to continue to evolve. In those discussions, Pokémon came up early on. I imagine a lot of you have played Pokémon or are familiar with the franchise. It represented something that was beloved around the world, very well-known to gamers and non-gamers through the animation, and it had this attribute of—the whole backstory of the game is that the trainer is out in this environment that looks like the real world discovering Pokémon. It was perfectly set up to translate that into this real-world augmented reality game that we do.

It was around April 2014. Google did an April Fool’s joke. The Maps team did this mash-up with Pokémon. They did a video that got 18 million views. On the Niantic team we’d been talking about different IP to use and we saw how that took off virally. We reached out to the Pokémon Company at that point and said, “This joke did really well. Let’s build a real game around on that concept.”

They had already been playing Ingress. The CEO of the Pokémon Company, Mr. Ishihara, was I think a level 11 Ingress player when I met him. He was playing with his wife every day. They were into the concept and got it from day one. I’ve given a lot of pitches for different kinds of businesses and partnerships through the years, but this was one where from that very first meeting, it seemed inevitable. Both parties said, “We have to do this.” We got a deal signed shortly after that.

Then, whenever the Alphabet reorganization happened, we were looking at how to continue doing our project, whether to fold that in under Android or one of the other Google divisions or whether to seek outside financing and do it as an independent company. We talked to VCs and our partners at the Pokémon Company, and through them with Nintendo. They were ready and willing to step up with financing and get behind the product in that way, to be an equity investor as well as a contributor of IP and gaming expertise. That seemed like the right way to go.

We could have taken money from venture folks, but to do it and create that kind of close relationship with Pokémon and Nintendo made them comfortable, collaborating with us around this crown jewel property — it made that whole relationship just rock solid on both sides. They’ve been terrific partners.

GamesBeat: How familiar were you with the Pokémon franchise before the April Fool’s joke?

Hanke: I missed it as a gamer. It came after that phase for me. But as a father of three — my oldest son just graduated from high school a few months ago. I have a girl who’ll be a junior next year and a younger son who’s 10. Our house was filled with Pokémon trading cards and guidebooks and videos and a stack of cartridges for several generations of Nintendo handhelds. I knew it mainly through the eyes of my children, as a game property, but it was clear — if you guys have kids who play those games, it’s the kind of game that your kid will spend hours with. People really enjoy that world and being in it. It felt like a super solid thing to build on.

GamesBeat: Taking Ingress as a base for Pokémon, you were saying that you completely scrapped the code. You rewrote the engine for it. What are the advantages and disadvantages to doing that? Is that commonplace, do you think?

Hanke: Many of you are developers or run companies that develop games, yeah? You usually get it wrong the first time. It’s pretty common that you get V1 of a product out there, your users use it, and you see everything that’s wrong with it. You realize all the architectural mistakes you made. You can patch it and improve it up to a point, but it’s often the case that it’s a good idea to put it aside and do it again, learning from everything you did the first time.

With Ingress, we had the benefit of having millions of users play that game. We were able to monitor it on the back end and understand how it was performing, how it was scaling, what the costs to operate the game were. We understood that there were ways to do things in a more efficient way if we were to rebuild the system. We had people on the team that were able to do that, so we undertook to basically do a version two and build our future products on that new infrastructure.

I’m super glad that we did it, because the scale that Pokémon Go hit would have absolutely crushed our first-generation technology. There’s no way we could have handled the influx of users. I’m so impressed by our team and what they’ve accomplished with the technology, because what we have today has held up to — there have been some outages and we’ve had our struggles as we’ve rolled the product out, but it’s held up to a staggering amount of traffic. It was definitely the right business decision in this case to put V1 aside and invest in V2.

GamesBeat: One thing I’m fascinated by is the relationship you’ve created with real-world places. A friend of mine described it as you guys being a virtual real-estate company, where you have this control and power over locations in the real world. But there’s also been — it stretches from basic issues to more complicated ones. Are you going to let people create locations in these vast swathes where Ingress players didn’t create Pokéstops and portals? And complicated issues like the advertising question, or sponsored locations. Can you get into how you view that relationship between locations and virtual objects?

Hanke: We ended up building Ingress and then Pokémon Go around places that we thought would be fun for people to discover. In a sense I guess it’s good that we didn’t have too much adult supervision when we started, because it probably would have sounded like a really bad idea. We were trying to combine things that seemed like they wouldn’t fit — art, historical markers, museums, things like that — with a science fiction game.

It was based around this notion that it would be a good thing if people got out of the house and wandered through their part of the city or to the other side of town and discovered things about it that they didn’t already know. The thesis was that it would be fun, that people would enjoy it, and that it would be generally good for cities if people were out using public spaces, maybe running into each other, maybe going to that park that’s near your house that you drive by every day and you never stop there. What would happen if people stopped there and used it? Would the park get better? Would people volunteer to clean it up?

It was admittedly kind of a wacky notion, to try to combine civic engagement, in a way, with a game. We didn’t want to do it in a heavy-handed way — “this is a good thing, go out and do this for your edification.” We called it accidental history. That was the internal moniker for it. Let’s just slip this stuff in. If you stumble across it and you happen to learn something about your community, you’ll enjoy it and it’ll be a good thing, but we won’t make a big deal out of it in the product.

We seeded the initial data set for Ingress with some of the content we had collected for Field Trip, which was a collection of historical markers across the United States, and a data set of all the public artwork where we had photographs and GPS coordinates. Statues in the park, that kind of thing. That was a few hundred thousand locations – museums, libraries. We wanted the game to work in small towns and big cities. And then we asked users of Ingress to find cool places in their neighborhood or city that they thought would be interesting spots to be portals in Ingress, places that had some interesting cultural attribute to them, that were visually distinctive. We asked them to submit those to us to be added to the game.

A lot of gamers really enjoyed that search for cool places, to try to find the unique spots in their neighborhood and get them past our filtering process and get them approved to be in the game. To a lot of early players, that was the game. Through that process, we went from a few hundred thousand seed locations, which kind of inhibited the number of people who could play Ingress in the early days — the map was not populated in many parts of the world. But over the course of two years and then three years, that got built out all around the world. We’ve had tens of millions of locations submitted and many millions of those have been approved and added to the game board. Ingress was ultimately able to be played in more than 200 countries.

That global game data set got created from what people found were interesting things in their neighborhoods. We were able to use that to launch Pokémon Go in all the countries that we’ve launched it, and we’ll continue that roll-out really quickly, without needing to go through the months and years of building out that data set. It seems to be working. People are going out and exploring their cities, going to different places. There has been this whole phenomenon of — sometimes a local business is near one of these Pokéstops and they’re able to apply a lure to it, which causes the Pokéstop to spray out confetti and make Pokémon appear much more frequently.

GamesBeat: You have at least some data that that leads to more purchasing, perhaps, from businesses.

Hanke: We did early studies on that back in the Ingress days. Gamers will memorize the locations of these local businesses, they’ll visit them more frequently, and they’ll make purchases when they’re there. Then you see that ad hoc validation from all the small businesses that have adapted that strategy themselves, just buying lures through in-app purchases and applying them to Pokéstops nearby. Some of them have published ROI metrics around that. There are how-to guides for businesses about how to use Pokémon. That’s really cool. The power, if you will, is in the hands of the users and businesses that are savvy and want to take advantage of that.

GamesBeat: In the long term, do you see — you were saying you see these in-app purchases maybe not being the best way to make money compared to sponsored locations.

Hanke: We do both. But from the beginning we did want to try to grow a business that would be a complement to in-app purchases. I assume a lot of you guys are deeply into IAP and you know the challenges there. I just bumped into the CEO of Kabam. You have this tension, always, between game design and monetization with that model. It’s tough to be disciplined about it and understand where you want to draw the line between monetization and fun gameplay.

I grew up with computer games and video games. I love making them and playing them. We wanted to always preserve the integrity of game design if possible. We felt like it would be great to be able to complement IAP with a sponsorship model that was economically strong, economically viable, to take a bit of the pressure off IAP so that we could protect ourselves from ourselves. [laughs] We wouldn’t have to cave to that pressure to dial it up a bit more.

It’s in its early days. We have run that through Ingress. We have a number of global sponsors in that game. We launched Pokémon Go with McDonald’s in Japan. We’re talking to a bunch of other businesses that want to take advantage of that model for Pokémon Go in other parts of the world. It’s promising. Again, it’s early days. People have been talking about ads in video games for a long time. But this is a case where, because you’re able to draw people to physical locations, you can do something that not too many other forms of advertising can do. We’re excited about it.

GamesBeat: Users and businesses have sort of scooped you on this, but I think you said part of the business strategy was also holding events. How do events fit into the business?

Hanke: I come to this group of people and I know that a lot of you are probably very involved in things like esports. Events and games are becoming a huge deal, and not just in our particular category of games. Games are now socially mainstream. People want to connect with other gamers, not only online, but in the real world. We see that manifesting across all different parts of the industry. We’ve seen esports explode into this huge deal.

With our particular kind of game, it involved people going out and playing in the real world, playing with their friends and meeting new people. Events have always been part of the mix for us. We started with very informal meetups with Ingress, where we had—literally, we had our first meetup with about 45 people. Just observing that people were thrilled to come together and meet other people who were into the same things as they were and share experiences with the game.

GamesBeat: What did they do at the meetups?

Hanke: The difference between our game and just a meetup about a game is that you can get together and actually play the game. You’re physically coming together. You’re playing together. You’re catching Pokémon or harvesting items from Pokéstops, or in Ingress you’re harvesting things from portals and battling between the two factions. But then you can go for beers or food afterward.

It provides a great combination of gameplay and real live social hanging out. I think of it as a fun, very inexpensive alternative to going to a movie or just going to a restaurant or a bar. It’s more active than either of those. It provides a lot of open space for conversation and hanging out. It’s a great social option.

Players of Ingress liked it. We formalized those events and came up with a name for them. They’re called “Anomalies.” We started organizing them in cities around the world every month, on a monthly cadence. We settled on this pattern where on one weekend we’ll do an event in Asia, Europe, and the U.S. We pick three cities, one in each of those parts of the world, and every month we do a global event. Those now draw sometimes in excess of 10,000 people, a range of 5,000 to 15,000 people at each site. We’ve started live-streaming those. Our last event in Tokyo, Niconico had more than 60,000 people viewing the stream of the event and we had north of 10,000 people on site.

We wanted to do that with Pokémon Go. That was one of the things we wanted to transplant from our experience with Ingress and replicate with Pokémon Go. For those of you that have followed the Pokémon Go media thing, you’d be aware that users started spontaneously organizing events long before we ever got to it. A couple of weeks after the game launched there was an event with 9,000 people in San Francisco. Professional sports teams have organized Pokémon Go events in their stadiums. College campuses have done the same. Players have beaten us to the punch in terms of self-organizing these.

We’re supportive of that. These are fan-organized events. They’re not companies exploiting the brand at this point. It’s great. We probably will do our own events at some point in time. Pokémon of course has had tournaments and national championships and international championships for a long time. They’re definitely familiar with that phenomenon. Bringing Pokémon Go into that tradition of having Pokémon events is going to be great. But at the scale we’re at and the number of people that are getting together, logistically it’s pretty challenging. We have to figure out how to do these things safely and find the right venues and get the right staff in place. I’m looking forward to that.

GamesBeat: On the hardware side, do you think the future in augmented reality is this combination of smartphones with things like the pins Nintendo is going to start selling for Pokémon Go? Is there something specific that you see AR hardware evolving into? What do you want that would make the experience you’re trying to achieve better?

Hanke: I’m a science fiction fan. I want the contact lenses that transform everything into whatever themed world I want it to be. In Rainbows End Vernor Vinge described a world like that. A lot of people see that archetype as something we’d like to have. It’s probably going to get created within our lifetimes. I’m looking forward to it.

When we started Niantic, there was a watch project underway at Google. There was what ultimately became Google Glass. I personally see an evolution of wearables — things like watches that you carry with you, Fitbit, Pokémon Go Plus — towards glasses, and glasses evolving from things that are maybe not super sleek and sexy in the beginning to models that are more and more so over time.

To me it’s a continuum. It’s not a binary thing, where there’s no AR one day and then AR arrives the next when some particular device gets launched. You can do a fun AR experience with just a phone. Pokémon Go proves that. But it’ll be better with devices. I’m looking forward to the Pokémon Go Plus launch, because that’s going to give people a way to play the game and not look at their phone all the time. They can look around them and see the interesting places that we’re trying to help people discover.

But we’ll definitely take advantage of other kinds of AR devices as they become prevalent out in the market. I’m excited about investing in that area, about building prototypes and bringing our gaming products to that hardware as it comes to market. Maybe we’ll be helping some of that hardware be successful in the marketplace. It’s a really exciting time for anyone making games, anticipating the kinds of experiences we’ll be able to create as those devices transform people’s experience out in the world.

GamesBeat: Back to software for the last question. Part of what’s made Pokémon Go is this aspect of having to figure it out yourself. There aren’t a lot of instructions. You talk to your friends and learn about it. How do you maintain that going forward?

Hanke: Augmented reality, that real-world gaming concept — you can embed interesting hidden things out in the world. They can have all kinds of different behaviors. You can find them by yourself, interact with them, interact with other people. The gameplay mechanics — that’s one thing I’m most excited about. We’ve only just scratched the surface. There’s this whole new set of game interactions that we can start experimenting with and building products around. There’s ample room for products that are going to surprise users that are very different from what we’ve done with Ingress or Pokémon Go.

There’s a ton of way you can take that into different kinds of games and build different kinds of experiences. And as new hardware comes online it’s going to be super fun to see what people create. We’ll take a stab at it, but I hope other people in the industry jump in and create fun games that I can go out and play.