Google Signals, still in beta testing, will help track and market to customers across devices.  This is important because the user will interact with digital properties on multiple devices every day.

To manage something you have to be able to measure it.  This whole picture measurement allows you to:

  1. Optimize the complete customer experience, with specific consideration for how they use different devices, and in what order, along a multi-session path to conversion.
  2. Enhance the relevance of ads by device.
  3. Build audiences (for advertising, as well as personalization & testing) that allow treating the user as the same person regardless of the device they are using.

While all of the above benefits are great, we think item 3 is especially powerful, and can be engaged immediately with no additional work.

Relationship George & Independent George are the same person

In “The Pool Guy” Seinfeld TV episode, George reveals he has two distinct personas, Relationship George and Independent George.

Unlike George on Seinfeld who really wanted his different personas to be treated as separate people, we do not want or need to be considered as different people across our devices:  Desktop John, Mobile John, Tablet John, Smart TV John, Xbox John, and so on.

If Desktop John converts, and Google knows that Mobile John & Tablet John are all the same person – with Google Signals engaged – Desktop John getting added to a remarketing audience list for advertising or for personalization / testing with Google Optimize, means that Mobile John & Tablet John get added to that audience too – all treated as the same person.  This happens even if your digital property doesn’t allow users to login to an account.


If you engage Google Signals for your digital property in Google Analytics, some limitations apply – the biggest of which is that Ads Personalization must be engaged for a given user to consider their activity across devices as the same person.   That begs the question, what percentage of users have this setting engaged?  Also, is this setting turned on by default?  If the answer to the later question is “yes”, then indeed most (a vast majority of?) users probably have not turned off this setting.

Some guesswork is involved

“Because of the large volume of data generated by users who have turned on Ads Personalization, Google is able to estimate from that data the cross-device behavior of your entire user base.”

In other words, they are doing some guessing.   But, as we noted when writing about Google Ads store visit conversions, Google has a data collection infrastructure in place that allows it to be pretty good at guessing.

By activating Google signals, you enable Google Analytics to collect additional data about your traffic to drive reporting for cross-device audiences and insights.  This aggregated and anonymized additional data can come from end user location detection, search history, YouTube history, and data from sites that partner with Google.  Such data may be accessed and/or deleted by end users via My Activity.

If you don’t want to collect data for Google signals, you should not engage the related toggle in the Google Analytics admin interface, and also ensure that you have not manually enabled any Advertising Features data collection in your Google Analytics tags.

Other Google Signals limitations

  • Segments in Google Analytics are not compatible with cross-device reports.
  • Sampling threshold for cross-device reports is 250k sessions for the related report date range, whereas standard ad-hoc query sampling starts at over 500k sessions (property level) .
  • Google Data Studio, Big Query, Custom Report/Tables and User-ID views can not use data collected by Google Signals.

Additional Sources

Cross Device reports with data from users who turned on Ads Personalization

Tips for interpreting data in Cross Device reports

Audiences on steroids with Google Signals