How Near became Southeast Asia’s most valuable marketing tech startup

Facebook knows a lot about you. In fact, in some respects, Facebook may know you better than you know yourself.

However, where platforms like Facebook, YouTube, Amazon, and Google fall short is how our digital footprints tell only part of the story. To get the full picture, there’s a whole host of data from the physical world that these platforms might not necessarily have – and that’s the problem that Anil Mathews, CEO and founder of Singapore-based startup Near, is trying to fix.

Anil Mathews, CEO and founder of Near / Photo credit: Near

Part of the reason why these tech giants have been so successful is that their digital user data is pretty accurate. Google might profile you as a potential car buyer because you’re visiting car-related forums, reading car reviews, and watching Top Gear on repeat. And more likely than not, Google would be right.

However, just because you’re an automobile enthusiast doesn’t necessarily mean you have a driver’s license, let alone the means to buy a brand-new car.

Enter Near, which was founded in 2012. The company has created a platform that uses machine learning to combine digital-world data (what people do online) with anonymized real-world data (what people do in real life). That integrated data gives businesses a more insightful view into people in a specific physical location, allowing for more effective marketing campaigns.

One example of this is gym chain Virgin Active, which wanted to boost customer engagement and walk-ins, especially among students. Near served up targeted mobile ads to students within the vicinity of Virgin Active gyms. This helped increase walk-ins by 82% and reduce the average cost per walk-in by 62%.

“With information on 1.6 billion users across 44 countries, we are the world’s largest source of intelligence on people and places,” Mathews claims.

How the startup has amassed its vast data trove is simple: mobile devices. These gadgets constantly know where people are because they’re connected to telecommunication firms, Wi-Fi providers, as well as app developers. What Near has done is to strike up strategic partnerships with these providers, obtain their diverse data sets, and then “fuse” them together.

The “fusion”, as Mathews calls it, is tricky. For one, there isn’t a standard way for these service providers to generate identifiers for mobile devices. For example, if you purchase a cup of Starbucks coffee via the official Starbucks app on your phone, the app will generate an identifier, the credit card transaction will produce another, and the Wi-Fi you’re connected to will create one more. This means that there are multiple identifiers and data sets for the exact same user and device.

Another issue with data is noise. Without getting too technical, Mathews explains that location data comes in long strings of numbers (think latitudes and longitudes). One wrong digit, and the entire location will be off. This also explains why Google Maps sometimes tells you that you’re in the middle of the ocean, even if you’re really chilling out at the seaside bar.

“Now imagine if you want to target coffee lovers in a very specific coffee shop in a building, but the coffee shop is right next to a burger shop, and the location data is wrong,” Mathews says. “You end up profiling the wrong users, and your brand cannot reach out to the right people.”

Near’s proprietary AI technology can standardize and “clean up” the data, and then merge this disparate information together. A lot of this process happens underneath the hood, so users don’t have to worry about it. That’s because the startup has also developed a software-as-a-service product called Allspark, which is basically a marketing automation tool that allows marketers to visualize and analyze their target audience around location, demographics, interests, and more.

The walled gardens

Near’s previous iteration was Imere Technologies, a location-based mobile advertising platform that was founded in 2009. It used real-time geo-location and consumer behavior to target relevant users, enabling advertisers to reach a massive audience based on where they were and what was around them. A lot of Near’s expertise in “cleaning up” physical-world data was developed around this time as well.

Lost in translation