Smartphone users are familiar with consumer-facing map services such as Google Maps, which consistently ranks among the most frequently used apps. But companies depending on maps are far more numerous than map services apps. Countless mobile apps that use location services such as delivery-oriented Instacart and transportation-oriented Uber rely on mapping software, and still more industries use maps to bolster their efficiency, be they credit card companies or advertising firms.
Mapsense helps these kind of companies use geospatial data by providing the technology to render it into an understandable, visual format. Street Fight recently caught up with Mapsense CEO Erez Cohen to talk about the growing need for mapping data in on-demand apps, and increasing competition in the geolocation space.
What technologies have allowed Mapsense to develop its software, and how difficult is to make mapping software?
What we’re doing at Mapsense wouldn’t have been possible five years ago, or three years ago. The technology on the browser — where all our maps are interactive, vector maps, and everything is completely animated and smooth — is achieved by actually sending vector data down to the browser instead of using browser data. Most maps that you play with online are actually pre-rendered on a server, so you’re seeing an image of a map that is sent down to the browser. We’re actually sending the raw vector data — so it’s latitude-longitude and other metadata — to the browser and then using the browser to actually render the map. Instead of rendering happening server side, rendering is happening client-side. This is something that’s just possible within the last couple years with modern browsers that have a little bit more power and allow us to do very interactive things on the client side.
Also, most mapping companies and most GIS companies don’t focus on building their own basemap. A basemap is all the buildings, roads, rivers, highways, states, counties — all the things you’d see on Google Maps. We actually have our own map built on open-source data and some other sources that we’ve spent the last couple years curating and building.
Two big things come out of us building our own basemap. One, we’ve built some hardcore technology to be able to build a vector map of the world and to be able to process all those polygons on the back end. We use the same technology that we used to build a basemap for auto-customer data sets. So, for example, we have several advertising customers who want to create geosensors around every fast food restaurant in the United States. … That’s something we would not be able to do if we didn’t have the technology we built by building our own basemap.
The second thing that’s nice about our rendering our basemap is we have underlying data that’s in the basemap. We have the vector data for every building in the United States. … That’s useful because more and more apps are actually interested in having that underlying data. There are certain apps that want to know every time you land in the airport or every time you enter a bar; we have the polygons for those places. … That is something that’s very hard to build. There are not that many companies that have a basemap of the world. You can probably count them on one or two hands. … Entering that group of companies has been a really hard engineering challenge.
To what extent has the growth of OpenStreetMap enabled Mapsense to do what it does?
We as a company would not have made the strategic decision to go after building the basemap if it weren’t for the OpenStreetMap project. We would have relied on a third party. … Not doing that has been both strategically and technically and from a business perspective very lucrative and a very good decision — and we were only able to do that because of OpenStreetMap because it’s an open-source map data set that’s always being added to and updated.
What types of companies buy your software?
We have done something that I think is different from a lot of other mapping companies. From the very beginning of Mapsense and through today, we’ve stayed very horizontal. We’ve tried explicitly — and it’s worked out — to stay horizontal in that we’re answering questions and location-based questions from companies in a wide variety of verticals.
When we started Mapsense, we were very explicitly only going to collect location data. We do not accept data that does not have some type of location reference. … It’s limiting, but it’s also empowering because we have reduced our problem set to only location-based questions, and the location-based questions that people want answered are pretty common across verticals. There’s a common set of questions you might want to answer. … So, we’ve been able to build a platform that scales across verticals. We’re working with a couple of credit card companies mapping transaction data; mobile advertisers mapping mobile advertising impressions; we’re working with apps that are mapping user interactions on phones; beacon companies. And we haven’t done any customized work for any of these verticals.
How large an impact has the emergence of mobile apps that center on delivery or transportation had on your strategy? Why has the market opportunity increased for maps?
The on-demand, location-based app is the Mapsense market opportunity. … There’s a huge new set of customers and a huge new set of data. This data is coming in real time and it’s orders and orders of magnitude larger and faster than the data that is being mapped by traditional mapping companies. … We have several of these on-demand apps as customers, and we see it as a huge vacuum in the space that we can start to work with these companies.
It’s definitely what we see as a market opportunity: a huge set of apps that are focused on delivering something user-centric where the location of the user or the service or the good is inherently valuable to know and understanding where those locations are (of the user or of the good) can actually improve the experience for the users. For example, connecting drivers and riders or bringing food to someone who wants to eat it or connecting people who want to go on a date. All of these things are location-centric and it doesn’t work if you don’t know the location of those users. … This is the future of mapping, and the industry is much larger than the traditional GIS industry. It’s actually growing at a massive rate, and there are a massive number of new companies that need the services of mapping analytics because they’re generating all this location data.
Are there a number of other companies that render geospatial data and sell it to companies that need digital maps?
The big differentiation of Mapsense is in the magnitude of data that we’ve been built to handle and the fact that we can handle streaming data sets. … Traditional mapping and GIS companies and even some startups out there now just weren’t built to handle hundreds of millions or billions of streaming points every single hour, and we’ve engineered the infrastructure from the ground up to be able to do that. … Traditional GIS companies are not growing to scale with data. Mapsense can support data sets that are changing all the time, so every few seconds you can see a live, updated version of a data set. Traditional GIS companies are a little more static.
Do you see Google’s dominance continuing in consumer mapping apps? Or do you believe other startups will be able to compete?
Google still has the best consumer-facing map. Their dominance there is unparalleled. The reason I don’t know if that’s going to stick is people are going to start using maps less. Apps are becoming increasingly location-centric, but that does not necessarily mean that they have a map on their user interface. When you use Tinder, there’s location going on, but you’re not looking at a map. So that’s an example of a location-centric app that is not map-based, and I think that’s increasing. So I think people are going to start using maps less, and that might bring down the dominance of Google maps. It’s easier for companies, and there are other startups helping apps build location-based services that aren’t necessarily map-based. … It’s really, really valuable to Google to be embedded in Uber or to be embedded in Yelp, and if those apps remove the map from the interface, it removes the dominance of Google as the primary app that everyone sees.
The other thing I would add is that Apple maps is increasingly devoted to putting people on the mapping scene, and they’re getting much, much better, so I actually see the differences between Apple maps and Google maps going down, and I think that’s going to continue to happen. Apple is doubling or tripling the size of the mapping team and they’re growing the devotion to making a competitive map.
Also, a company like Uber now also has primary sources of location data. … Uber drives every street in New York every 24 hours. … This is extremely valuable data. … Imagine if Uber started to use the location data coming out of its drivers to build a better map. I think that’s something that Uber’s thinking about very hard. The consumer-facing map of the world is becoming less and less of a valuable thing.
Joe Zappa is an intern at Street Fight.
This interview has been edited for length and clarity.