Let's Talk About Geolocation Data
By: Bala Gopalakrishnan
Mobile geolocation is a long sought-after data set for publishers, marketers and analysts. In a world where people would sooner leave the house without their shoes than their smartphones, mobile location is the key to understanding a consumer’s physical world behaviour. Despite the potential– accurate and actionable mobile geolocation has been elusive until now. It is finally mobile location data’s time to shine.
Types of Mobile Geo-Location Data
Bid Stream Mobile Geo-Location
This is location data (a latitude/longitude coordinate) provided by real time bidding (RTB) ad requests. Bid stream data is derived from whatever location signal (i.e., Wi-Fi, cell tower, GPS, IP address) is available to the device during ad-serving. The strength of bid stream data is its large scale. Bid stream is often scrubbed (80-90% of bid request location data is inaccurate according to an AdExchanger article) by vetting the number of decimal digits (precision) to filter good location signals. However, this precision of geo-location data is often confused with accuracy.
For example, the geo-location signal received may be as precise as 10 meters in the bid request, but that location sent by the device could be the nearby cell tower and not the actual location of the user. Also, since bid stream is reliant on when an ad is served, it only gives glimpses into a location behaviour and not the whole story.
Beacons are Bluetooth devices that create precise virtual zones into which an app can detect
entry and exit. The accuracy of this type of tracking can be as granular as a few metres. Beacons are the best way to detect location signals in situations where indoor precision is key (i.e., inside store sections). Beacons are still small in scale, as implementing them has been a barrier to entry for retailers.
Always-On In-App Geolocation Data
This data set is far richer and more accurate than bid stream, and higher in volume than beacons. Since this data is collected in-app, mobile app creators can leverage their own deep proprietary technology/algorithms to effectively capture user movements, frequency and dwell time, making this data both accurate and precise. Always-on tracking provides a user’s complete story which can be analyzed to understand physical path to purchase (i.e., where does a user stop directly before your location? Where do they go afterward?), competitive insights (i.e., what other vendors are your customers frequenting?) and lifestyle inferences (i.e., commute patterns and paths). However, apps need to provide a clear value proposition to its audience to entice them to enable the “always-on” location function. With battery drain and sensitivity to user-tracking, this is not often an easy ask. Yet, for apps that are successful, the insights are powerful. FourSquare showed the power of in-app accurate location data by predicting the number of iPhone 7’s to be sold.
Mobile geolocation has always been a sleeping giant. It has just woken up and is going to a be a monster force in marketing analytics and effectiveness.