Monetizing sensor data requires new thinking

As the transportation industry becomes more dependent on the use of data from connected-vehicle sensors, there's a growing interest in the future business opportunities in the sector. To gain more insight on those opportunities, Special Correspondent Marcus Amick talked with James Hodgson, principal analyst at consultancy ABI Research. Here are edited excerpts.

Q: ABI Research estimates that connected sensor data could generate more than $700 million globally by 2023. What factors are driving that growth?

A: First, you have the technology factors. There's been a real growth in the number of connected cars, and also the number of vehicles with sensors on [them], different kinds of sensors that are equipped to aid us.

At the intersection of these two technology trends, you have this sensor data crowdsourcing paradigm that kind of piggybacks off those two markets.

How, specifically, is crowdsourcing, or this use- generated mapping process, driving the monetization growth in the sector?

It's really been the development of different platforms, and the cooperation between different brands, to begin the whole process of data aggregation from connected cars on the road. Having that kind of standardization has been what's led to the building of these technology platforms with the appropriate analytics to build maps from crowdsource data. Once those platforms are developed, it only makes sense that you make as much money off that platform as you can. That's where the monetization comes from — it's more indirect monetization.

Are there any trends creating new opportunities in the sector?

There is already a high bit of that data on the road … even though the data is what I call "low resolution" or "low bandwidth." For example, if you start your car and drive away, that will provide some low-resolution data from the engine turning over. You can then infer that there is likely to be a parking space opening up wherever that car just turned its engine on. Another important development is the increasing usage of semantically rich sensors [which provide more meaningful data based on contextual cues]. Those are the two most important growth potentials in the short term.

What's an example of where you see growing business opportunities with the use of more semantically rich sensors?

The best example is the camera. You can learn a lot about what's happening in the environment, thanks to having a high-resolution sensor. … There are already a lot of camera sensors for autonomous parking, which are constantly searching for empty parking spaces. Now it would be possible to just share simple geographic coordinates for the location of these spaces without having to upload really rich data images of empty parking spaces.

What are some of the misconceptions about the monetization value of sensor data that need to be considered when trying to build a solid business model in the sector?

Sometimes OEMs can overestimate the value of their data. They are aggregating data, but it's actually really poor quality data. There's also a common misconception that OEMs won't share their data. Very often they will for a cost, but sometimes the cost is too much. It's difficult to build an ROI case, which is why you have to do it with this kind of platform or work with some kind of third-party platform to get a user-generated mapping platform going. Once that's in place, then it's likely to produce some change in the internal corporate structure that will look at it and say, "We built this platform," or "We paid for this platform. How else can we use it?"

How are concerns about privacy affecting this business sector's growth potential?

It's very rare that you have identifiable personal data or even nonidentifiable data. Where privacy concerns come in is where you have what you call direct monetization for anyone who wants to grab data and crowdsource it for the purposes of marketing and advertising. That tends to raise eyebrows.

Is there anything that automakers can do to help better address those privacy concerns among consumers, to create more business opportunities in the sector?

There are always going to be hyperaware and purposely conscious consumers. What OEMs really want to avoid is the idea that this kind of data aggregation doesn't, in any way, interfere with the primary experience that they want to provide, whatever that is. They don't want it to be too in-the-face, in the way of targeted advertising. I think it's going to be a smaller problem than most people anticipate because it's happening already.