Device fingerprinting is a technique advertisers use to identify people who have interacted with their adverts. It works by using publicly available characteristics of a device to create a “fingerprint” of the user. This could include device, location and other data relevant to advertisers. This metadata is always sent by internet-connected devices when they connect to a web page or a server.
When a user interacts with a mobile ad, the advertiser redirects the user to the app store through a tracking URL. Once the install is complete, the advertising SDK checks to see if the user’s fingerprint is visible from the app’s install base – attributing the install to that advertiser if it does appear.
Although there are methods of tracking advertising that directly attribute a click to an install based on a device ID, these methods do not necessarily work for all mobile marketers looking to track their efforts.
There are two big reasons for this. The first is that not every advertising network can support integration with advertising identifiers used by major mobile manufacturers (such as Apple’s Identifier For Advertisers). This means marketers and advertisers cannot discern precisely who a user is and whether they download as a result of an ad. Secondly, even if a network does support ID tracking, users can turn the feature off. Although many users don’t do this, the fact that it’s possible means that there is an attribution gap marketers need to fill.
Device fingerprinting exists to solve this problem. Acting as a frontline attribution tactic and a backstop when other methods fail, it’s a useful way for marketers to complete the tracking picture.
Device fingerprinting is a useful attribution tactic used by marketers as a method of last resort to ensure every campaign has a backup option. Adjust’s platform allows you to perform this function with ease, and we can plug the gaps that enable users to dodge typical attribution through our generated user ID. The main ways in which device fingerprinting comes into play include:
To learn more about fingerprinting, as well as other attribution methods, take a look at our documentation.