If you’ve ever used Google Maps, Waze, Apple Maps, TomTom, Garmin, or shared your location with any app on your smartphone, then you were sharing and using GPS data. GPS was invented by the U.S. military in the 1970s, and eventually declassified and made available for civilian use in 1983. For more about the history of GPS, check out my other post, What the FAQ - Who Owns GPS.
In the world of location targeting though, GPS is a much more accurate way of finding people. In 2020 in the U.S. alone there are an estimated 275M smartphones, each capable of collecting and sharing GPS signals.
Unlike IP addresses, which can get down your zip code, GPS is accurate down to 3 meters. This level of accuracy allows advertisers to target people not just based on what town they’re in, but what store they’ve been to, or where they are right now.
That might sound creepy, but like IP’s not sharing your home address, there are some built in guardrails with GPS too. To start, each mobile device has their own ID, similar to an IP address. On Apple devices these are called IDFA (Identifier For Advertisers). These IDFA’s are likely changing or going away soon, but for the sake of today’s discussion we’ll stick with them. With mobile devices each one has an unique number that identifies that device, but doesn’t share personal information like name, phone number, or address.
When it comes to GPS, these are coordinates that look something like this: 40.7501909,-73.9940959.
If you search that address in Google you’ll see that’s for Madison Square Garden. So how does Google know that?
Companies like Google, GroundTruth, FourSquare, and others have mapped out the real world and know what GPS coordinates align with which real-world locations. Therefore when they see a GPS signal they know if that signal is associated with Madison Square Garden, a Walmart, McDonalds, coffee shop, or any number of business locations.
Based on this information advertisers can target people based on their offline behavior. If you are selling cars it can be very valuable to know who’s recently been to an auto lot. If you’re a fast food chain you may want to target people who’ve recently been to your competitors and drive them to your store. CPG brands may want to target people who shop at stores where their products are sold. The list of potential uses can go on and on.
Aside from where a user has been, advertisers can set custom geographic targets known as radial targeting, proximity targeting, and geo-fencing. In these situations an advertiser picks a central point and can target anyone who is within a designated distance. These distances can range from a few meters, to 100 miles or more.
This can be more valuable than Zip code targeting because some businesses may cater to only people within 5 miles of their location. Targeting the whole zip code could still result in a lot of wasted impressions.
Advertisers get this data from a wide range of mobile apps. For example, a news app may need your location data to show you the local weather or local news. Once they have that data they can pass it on to advertisers who are buying ad space on their app.
As of late 2020 the majority of smartphone users share their location services with at least one app. Some apps like Maps or Weather may require location services to work correctly. Others may only need it for advertising purposes. Apple is trying to make it clearer for users when an app is collecting their data and what they’re using that data for, so keep an eye out for new OS updates in 2021.