Tapping into Mobile Operators’ Big Data

I posted a while back about how mobile operators are getting into the business of aggregating and selling customer data to advertisers. And now here we are.

SAP (my employer) is launching a new service that will change the way organisations acquire mobile consumer information. It’s called SAP Consumer Insight 365.

Currently, advertisers can target using individual preferences gathered from consumers, and then delivering content based on that. That information is unreliable in that it’s what people say they do, and say they like.

That’s why Consumer Insight 365 is such an advance. It sources data from major mobile operators across the world. That’s absolutely reliable because it’s based on what people actually do.

This doesn’t mean you’re about to get a bunch of scary messages on your phone from Big Brother saying he knows what you’ve been up to. It does mean that advertisers might soon be delivering more relevant ads to you in a number of ways.

Here’s an example. Let’s say we’re Mall of America in Minnesota, USA. What’s the best way to target our advertising so that we reach our customers? We could do a survey of license plates in the car park to try to work out where people are coming from, but that won’t give us a complete picture. We could hire a team of temp workers with clipboards to conduct a survey inside, but how many people would be willing to respond? And how many of them would tell the truth? Again, any data we could gather would be incomplete and unreliable.

Now, if we look at a mobile network, we can see all the people in the mall as their phones check into the network. If we cross-reference that with a partial postal code, then we can create a heat map of where customers are coming from. We’re not seeing individuals. We are seeing complete, reliable, aggregated data on a demographic scale.

We can gather this data for whatever time period we want, and then put it to use by advertising locally. To take it a step further, we can even check the cell towers along the roadways leading to the mall. Once we learn which routes people take to get there (based on their mobile phones checking in, and cross-referencing that with our check-ins at the mall), then we can advertise on those roads.

If those giant screen billboards are available, we can pay to just show our ads on weekends. The billboard vendor could do a socio-economic profile of the people driving by at different times of day, and display different ads depending on who’s driving by. We can take into account location, browsing patterns, time of activity, device type, gender, age, and demographics. Again, this is information that has previously been unavailable. We can’t stop and ask people as they go by.

Retailers worried about showrooming could see which web sites people visit while they’re browsing in the store.

Mobile operators have been collecting this data for years. So, why haven’t they been able to use it before? It’s the sheer volume of the data. We’re talking about 100 terabytes of data every month, petabytes of data every year. Further, each operator might be storing its data in 20 different databases. Until now, it hasn’t been possible to analyse and use this data in any practical way while maintaining consumer privacy.

That’s exactly why this is an SAP service. It’s powered by HANA, the next-generation in-memory database solution. HANA makes it possible to process the petabytes of data to provide near real time information on mobile user activity.