How can marketers track cross-device attribution?

There are quite a few categories in the app stores that cross-channel attribution is crucial for them. Travel and shopping are the best examples - users tend to use various devices before making a purchase. However, the Google Stadia launch proves that games, and probably other categories in the future, will need to have better clarity of the entire user journey, across all devices. This for sure will be the next evolution for digital and mobile marketers, and attribution solutions need to support it.

For deterministic models, mapping out user journeys across multiple screens requires:

1. Device ID / Google Play referred 

2. Cookie ID / Customer ID 

For probabilistic matching, various types of fingerprint methods (Battery, size, RAM, wifi, etc) are used.

As an example, an app-to-web conversion path (highly popular in eCommerce and Travel)  looks like this:

One important note - privacy and security play a big part in cross-channel attribution. It is vital to understand the legal implications of following a user across different screens.

The cross device question is key for multiple screens apps and you can address it this way: 

1. Once you are logged in, the tracking system will store every action on our app whatever device you use. 
2. The question is more complex when we want to follow a user from a website (not logged) to an apps (post login). With an analytics partner, you can build different proxys which are good enough to attribute our registrations per UA channels. What we usually do is a funnel with the following steps : 
   • viewed page Y with the specific UTM
   • then registered on our app on desktop or iOS or Android within X minutes. 

Another solution is to propose a code when you register on our website that you can use on mobile (similar to best practises on connected TV or similar to Slack and other web app magic link).