In 2010, Citymaps was a fledgling idea scribbled on a cocktail napkin at a Manhattan bar. Cofounders Elliot Cohen and Aaron Rudenstine had no experience with maps, but knew they wanted to build a map app people could use on their smartphones. “Why reinvent the wheel?” most people told them. Wasn’t Google Maps good enough?
Their answer, emphatically, was: no. They wanted to build a map embedded with business logos that offered users a more social experience. Cohen and Rudenstine started small, building a prototype map of local businesses across some 60 Manhattan blocks. But making even such a small map proved far more difficult than they’d imagined. “Little did we know at the time that building maps was so hard,” says Rudenstine.
Nonetheless, that prototype helped Citymaps raise its first million dollars in seed funding and the company has grown rapidly since, raising $11 million from investors and expanding to include 80 million locations around the world. In its latest move, Citymaps opened its API to developers this month, enabling them to embed the map in their own applications instead of using Google or Apple Maps. Already, big-name partners like Delivery.com, Comcast, and New York City taxis have signed up.
Rudenstine spoke with Fast Company about the company’s circuitous path to success and lessons learned along the way.
With little money and zero mapping experience, the pair used the Census Bureau’s database of geographic information to build a bare-bones map of New York City. Next, they needed to figure out a way to embed local businesses into the map itself–a feature they knew would distinguish their product from Google Maps. But after wasting $10,000 buying local vendor data that ended up being defunct, they realized the data they needed would require a lot more legwork–literally.
Each morning, Cohen and Rudenstine woke up at 4 a.m. and walked the streets of New York until 8, armed with clipboards. They jotted down the names of every business they passed those early-morning hours before heading off to their respective day jobs. “Our processes those early days were as manual and dumb as you could be. We were doing it by brute force,” says Rudenstine. Yet they were determined to get enough information, laborious as the task might be, to build a prototype for investors. “You need to have a vision you’re incredibly passionate about,” says Rudenstine. “If not, you won’t invest enough.”
But invest they did. Cohen and Rudenstine spent five months walking the city streets those early mornings until they finally gathered enough data to build a prototype. The problem was, they had no idea how to actually turn that information into a tangible map.
Cohen and Rudenstine knew they needed to build a prototype in order to convince investors they were on to something. But finding someone who could translate all their geodata into a map that humans could actually read proved impossible at first. “We went to every computer science program you could think of looking for a computer developer who could build a map from scratch,” says Rudenstine. “We couldn’t find anyone who could do it.”
They kept hearing the same thing from developers: “Why don’t you just use the Google Maps API?” But in order to embed a map with local business logos, Google Maps just wouldn’t work. Exasperated, they posted an ad on Craiglist–and a few days later, a cryptic message arrived in their inbox. “Hey Dude,” the sender wrote. “We know maps. We should talk.”
The message came from a developer named Adam Eskreis, who turned out to be an expert in mapping technologies–and soon, Citymaps’ first hire. Eskreis built the initial prototype that landed the company its first million-dollar investment.
Cohen and Rudenstine are super-careful about who they hire. To date, the company has a total of 17 employees, including a former product designer from Google. “We look for folks who have a lot of drive, vision, and passion,” says Rudenstine. “But even when you check all those boxes, sometimes the cultural fit isn’t there. We’ve had to turn great people away because the fit was [missing].”
Rudenstine stresses how important it is for every employee to work as quickly as possible. “The view that we had from the beginning was that we would never have a lot of money, and so the only way to compete was to move incredibly fast,” he says. “The only way to do that was to hire people who loved working together . . . and as a result, could put in long hours.”
At the same time, in an industry where competitors are leaping into the market left and right, moving fast is the only way for Citymaps to stay relevant. That means giving employees the freedom to make decisions on their own. “We ask people to be incredibly independent. No one is waiting around to be told what to do,” says Rudenstine.
To reach a massive scale of users and local businesses, the predawn clipboard-and-shoeleather method was most certainly not going to fly. Citymaps began thinking strategically about how to scale the Manhattan map to the world. Today the process it uses to build its maps is algorithmic and involves tapping more than 100 data providers for points of interest.
The company created a large-scale data operation that ingests all this data, removes duplicate information, and supports dozens of languages. “It’s one thing to get all that data for the United States. It’s another thing to get that data for the whole world,” says Rudenstine. “You have to get very creative.”
When Citymaps expanded from the U.S. to include locations around the world, its database ballooned from 10 million to 100 million records. The company had to beef up its search tool so that it worked just as fast despite the tenfold increase in records. Not only that, but Citymaps needed a way to keep track of businesses that shut down or opened up around the world to make sure maps were updated in realtime. That’s where creativity really came into play. The company began working with MasterCard, which notified Citymaps when transactions at a business would stop or new ones started popping up, often a good indicator of an either defunct or newly-opened venue.
Citymaps also developed a system for tracking businesses on social media, following 700,000 merchants around the world on Twitter and keeping track of social media activity on Instagram, Yelp, and other sharing and rating sites. All this data is aggregated to give every business on Citymaps a “confidence score” — a number ranking how confident Citymaps is that the venue is still in business. Streamlined systems like this make it possible for a company of 17 employees to manage such a massive and ever-changing body of information.
In 2012, Citymaps was busy scaling in the United States when an announcement rattled the company: Apple had just announced the release of Apple Maps in direct competition with Google. “It doesn’t get bigger for a company in our industry,” says Rudenstein. “It’s moments like that when you need to be creative and agile enough to put aside whatever road map you’ve been working on [no pun intended] and adjust to the circumstances in front of you.”
The company had been moving quickly already, but after Apple’s announcement, it picked up even more speed. All-nighters were a regular occurrence in those days. “The sense of urgency was multiplied by a factor of 100,” says Rudenstine.
These days, in an effort to expand and compete with the likes of Google and Apple, Citymaps is offering its API for use by developers. “Distributing our map API was one of the final pieces of focusing on the map itself,” says Rudenstine.
“In the world we live in today, where things move so fast, your long-term plans shouldn’t be written in stone,” he says. “At best, they should be written in pencil and you better have a really big eraser close by.”