Trapster CEO Explains Why He Sold Their Location-Based Driving App to NAVTEQ [Interview]

Earlier this year, we covered Trapster, a location-based social driving application that lets users submit their information about speed traps, red-light cameras and road hazards. The application was growing rapidly and seemed to have a lock on an interesting new market. Recently, the company sold to NAVTEQ, creators of digital maps and map content. We talked with former Trapster CEO Pete Tenereillo and current employee Sean Farrell about the deal. 1) What was your goal in selling Trapster, currently a free application, to NAVTEQ? To take Trapster to the next level, we need to know actual road vector data. To get there, we had 3 choices: 1) License the data (prohibitively expensive for a free app with millions of downloads) 2) Create the data ourselves, i.e. make a map of the world generated by GPS trace data from Trapster users. There are entire app companies already doing this (would have required a big VC raise and years of engineering work, and we still would have only a limited set of data) 3) Partner with a company who already has this data. Why we need the data: The iPhone and Android and etc map APIs allow manipulation of map tiles and placement of overlays such as traps and animation and polylines, but there is no access to actual vector data, i.e. a navigable map. Trapster currently uses proximity based alerts, and GPS lat/lng-placed overlays using these APIs. The Virtual Radar(R) alert cone follows your GPS position, and narrows and elongates based on your velocity and direction of travel. When the user approaches a reported trap, accident, camera, etc., and the area of the cone intersects the point where the trap is placed, the trap will alert via a spoken warning. The point of the narrowing is to reduce irrelevant alerts. The point of elongation is to alert further in advance when the car is at higher speeds. However this approach falls short in many ways. Here are a couple of them: First, the app sometimes alerts to traps and hazards that are not on the road the user is actually on. For example, if the user is driving along say I-280, and there is a trap reported on an adjacent access road, the trap will alert. This is annoying. Trapster should know the user is on I-280 and only alert to relevant information for I-280. But without vector data it cannot. We tried silly solutions like continuous reverse geocoding and then string-matching street names, but that is inaccurate/unreliable, and doesn't even come close to scaling to the level we need - we would be making 100s of millions of reverse geocode API calls per day! How NAVTEQ data fixes this: With NAVTEQ data we can know what road the user is on, and can match that with traps and incidents on that road. We can also get speed limit, class of road, and lots more. Game changer! Second, the app can only alert as far in advance as the Virtual Radar cone reaches. This is usually set to 1/2 mile or so, and elongates automatically if the user is traveling at freeway speeds, but that is not enough. For example, in rainy flooded San Diego this very morning, Trapster failed me. I drove all the way up to a road closure and had to turn around and go back. Dave, one of my employees, had already reported that road closure in Trapster on his way to work, but by the time I got the alert it was not very useful. There was a big line of cars, a police officer, and a bunch of barricades. I still had to backtrack and go a completely different way. Trapster should have alerted me as soon as it knew I was on that path. But it couldn't know. How NAVTEQ data fixes this: With NAVTEQ data we can do continuous predictive routing, constantly figuring out what path the user is most likely to take, and providing relevant alerts for that path far in advance. This will be amazingly better! There are many examples like this -- basically Trapster is about to get WAY better! 2) Do you foresee major changes to the application itself? Yes, we now have access to countless engineering resources across NAVTEQ, and the funding to staff up right here in San Diego (well, technically Cardiff by the Sea, which is 20 min north of San Diego). 2b) Will the Trapster database and technology be integrated into other products? You had asked about why Trapster sold to NAVTEQ (answered above). This is the other side of the question - why would NAVTEQ buy Trapster. And this is the answer - we have 10 million users who crowdsource real time speed trap, traffic, accident, etc. data. NAVTEQ appears to be totally focused on content, content, content and they're wanting to understand how to support communities in getting to the data they want most. They seem really interested in parking availability data right now. 3) What was your revenue model for Trapster? Are you profitable? Trapster is (and was) pre-revenue. 4) Your service uses Google Maps, We use Bing Maps on BlackBerry, Nokia, and J2ME phones. We use the standard APIs on iPhone and Android, which (completely out of our control) in turn do as you say use Google. Our Web site uses Google, while NAVTEQ has their own proprietary maps system. 4b) Do you foresee challenges in adapting their technology, or will you be able to maintain your current technology? It is not technically possible to access the vector and navigable map data described in "1" above without moving to NAVTEQ maps, so yes we will be changing over to NAVTEQ maps in the apps. 5) Your application is based on crowdsourcing information about speed traps. What other social elements do you include in the service, We already have the Caravan(R) feature, which allows users to see other users on the map in real time (buddies) and send audio messages (no texting or typing needed) to one other car or a group - very useful for that fraternity ski trip caravan up to lake Tahoe - "hey guys, let's pull over for dinner, I'm hungry!" or "slow down, I can't catch up!". We also have the My Trips feature, which allows users to record a trip, share the trip with friends and family, or on Facebook or Twitter. For example, you can share your trip with "Grandma" and she can watch your progress in real time (from any computer) as you travel toward her house for Christmas. As you take geotagged photos and place audio notes, she can see them and participate in the trip, instead of just stressing out about when you will arrive. 5b) Was entering the 'social' game part of the reason that you were acquired by NAVTEQ? Are they planning more social features in their products? Definitely. We had many more social and gaming features planned before we contemplated being acquired. This acquisition gives us the resources we need to accelerate that. Obviously I can't pre-announce features, but it will suffice to say that there will be a lot of cool things coming to Trapster!

Social Pro Daily

Trapster CEO Explains Why He Sold Their Location-Based Driving App to NAVTEQ [Interview]

We talked with former Trapster CEO Pete Tenereillo and current employee Sean Farrell about the sale of their location-based social driving application to NAVTEQ, a large digital maps company.

Neil Vidyarthi

About

Subscriptions

Events

Publications

WORK SMARTER - LEARN, GROW AND BE INSPIRED.

Subscribe today!

Neil Vidyarthi