Tracking the Same Person Across Different Devices with Google Analytics

As we continue towards a more customer-centric approach with Universal Analytics, we want to segment users based on their actions and identify those 20% of the customers that make up 80% of your revenue and be able to understand their needs and wants. We can track events, build out segments, and see the devices that they use, but what about finding out how those potential customers use your mobile/web apps (like online stores) across devices?

Last month Google Analytics announced that are rolling out the much-awaited (at least for me) UserID feature in Google Analytics, which will allow you to see how people are using your sites and apps across different devices. The new UserID feature is slowly rolling out to accounts that are using Universal Analytics.

How is it different than what I have now?

Currently, when someone goes to your site and does research, you record them as one visitor but if the then use their tablet to buy something, they show up as different user. So, you have one visitor that does a lot of research on their desktop (but doesn't buy), and one visitor that doesn't do any research, but buys on a tablet. This new userID feature solves that problem.


How does it work?

The overview In Google Analytics, you can define a user with a unique identifier. Once defined, GA will associate the visitor’s activities to that unique user ID. The best way to do this is if you have an authorization system like a login. Then, when the user logs in to your site or app you can send the userID to GA.

Important Note: The unique ID cannot have any PII (personally identifiable information), like an email/SSN/phone number.

After that is all set up each hit will be sent to GA with that userID.

Sometimes you need to set a userID for a visitor after they have already been on your site and that is where Session Unification sometimes comes in. This works when a user starts their first session and will group GA hits with the User-ID you set up and with one that was auto-generated by GA.

Here is how Google describes it:

Session unification allows hits collected before the User-ID is assigned to be associated with the ID, as long as those hits happen within the same session in which a specific ID value is assigned for the first time.

Setting it up

This part is can get a little technical but your dev team or any good analytics agency (*cough* *cough* like us *cough*) will get this easily.

Basically, you need to add new code to your site that pulls the userID from your database and populate it in the GA Code. Instead of regurgitating Justin Cutroni’s great instructions go to his blog post about this and read his implementation guide.

If you are a real technical person/developer you can jump straight to the documentation.

Some things to note:

  1. The new view you create will only show you stats for logged in users
  2. The unique sessions is populated by the amount of unique ID’s it gets
  3. You can only go back 90 days, (just like multi funnels)

Why should I care?

We have grown to become a multi-device species and devices are becoming more and more ingrained in to our decision.

Mobile and Tablet usage continues to rise! In Mary Meeker's yearly report on internet trends she pointed out that mobile and tablet is on the rise and projected to keep growing. People are bringing these devices into the research and buying phase so we need to understand what is really leading to the sales and how users are combining these all these devices. Also the presentation is full of great stats to use for reports.

Where Is The Decision To Buy Formulated?

Where do your customers do research about your products, and where do they buy? Should your site be responsive? Right now we can only tell that by looking at the devices that people buy on, but we are missing the important part -- the research phase.

There is a lot of stuff that is missed when not understanding the full path of the user and how they interact with the site in different ways. A customer might do all the research on his phone and then buy from the desktop, so you’re thinking that desktop is the highest converting device but in reality your tablets are the ones that the important research and decision making are being done on.

Optimizing Your App Based On Features People Use

You can see what features a user interacts with on his mobile, tablet and desktop which will tell you watch features should be optimized.

Check out the following video, it’s a great way to show how LinkedIn uses cross device management to impact how they show ads. I especially love their names for mobile and tablet. They say call the morning coffee time, because everyone is on their phones/phablets while they are in line waiting for coffee and the evening is couch time because people lay on their couches on their tablets and how the users use different features of the app at different times.

Get Real Attribution

Let's say someone comes from and creates an account or downloads your app. You might have a goal set up for that account and maybe value it at a certain amount but lets say after a week they upgrade to a premium account on their mobile phone. Previously, we couldn’t attribute that gizmodo has lead to a premium account but now with UserID we can tell GA that the source was the desktop version.

Or lets say they clicked on an ad on their phone and then converted on desktop, yeah, you got that attribution too!

Use It With The Measurement Protocol

Do you have another device that people are using to access your app? Like an in store kiosk, video game system, or POS system? Well, you can tie it back to that user with the measurement protocol even though you don’t have GA installed on it. To do that use the following measurement protocol parameter:

uid= Example value: as8eknlll Example usage: uid=as8eknlll

Bonus Video

Since we (or at least I) are getting pumped for Google I/O 2014, I thought I’d share the video that I’ve used as my bible while preaching Universal Analytics for the last year. It fine that it is almost a year old, a lot of the stuff just came out. I highly recommend one everyone taking the time to watch this video in it’s entirety.

Want to know more about Universal Analytics? See our infographic about Universal Analytics and the other great features it offers.

What ways do you want to use it? Tell us in the comments.

If you like analytics and data-driven marketing follow me on Google+ and Twitter I talk a lot about that stuff there and occasionally drop some articles about neuro & guerrilla marketing as well.


We love helping marketers like you.

Sign up for our newsletter to receive updates and more: