Solved

Tracking a user from browser to app

  • 1 March 2023
  • 7 replies
  • 555 views

Userlevel 1
Badge +1

Hi,

I would like to track a user using my webpage (with amplitude running) as the same user opening my app for the first time (with the iOS sdk running).

I can set the deviceID on iOS app to track IDFV, but that value can’t be captured on the webpage, 

is there any other way to attribute the user across both applications ?

 

Best,

Anand 

icon

Best answer by Saish Redkar 1 March 2023, 18:07

View original

7 replies

Userlevel 7
Badge +10

Hi @asivakumar 

The foundational identity resolution mechanism is that if a user signs in on the web app and also on the iOS app, then Amplitude will automatically reconcile both the device ids and attached them to the same user id.

You can read more on this here . Hope this article will help you understand Amplitude’s identity resolution approach.

Userlevel 1
Badge +1

Thanks @Saish Redkar 

I had looked at this document, but it says that Amplitude uses IDFV or random alphanumeric for app and a UUID for browser. I wasn’t clear how it’s able to reconcile between these. 

 

Userlevel 7
Badge +10

As long as the user is using the app anonymously, their usage on both iOS and web app will be shown under 2 separate amplitude IDs and device ids. For this to work, you shouldn’t assign any user id for anonymous users.

Once that user logs in onto both the apps, it should get the designated user id set by you. Amplitude will automatically merge the user’s usage for their separate amplitude IDs ( which were previously shown as 2 users ) under one user_id. In such cases, you should see those under the user profile’s Merged IDs section and also see the IDFV and UUID listed under device IDs for that user.

Ideally, from what I have seen,  teams prefer to instrument their web app events and mobile app events in separate Amplitude project buckets if the app experience is starkly different in both cases.

Userlevel 1
Badge +1

Thanks Saish, 

We look for certain signals on app usage that could indicate fraudulent behavior. My purpose is to identify a user who has depicted fraudulent behavior on the app when he/she shows up on web. At least in one of these channels (App / Web), the user would be in a non-signed in state or likely use different sign-in credentials. Any thoughts on how this can be achieved ?

Really appreciate your feedback on these, thanks again. 

Userlevel 6
Badge +9

Jumping in here @asivakumar to share a resource. (Thank you @Saish Redkar for all the help!) Have you explored our data taxonomy playbook? It includes best practices by industry. We also recommend this Academy course

Userlevel 1
Badge +1

Thank you @Amplitude Admin , will have a look at this

Thanks again @Saish Redkar - really appreciate it !

Badge

Hello 

 

I have a similar problem. I need to track usage of a user moving from web browser on to app. In the browser user is signed in, but in app, there are few initial steps to track where user may not be signed in and we need to know the failures happening in these steps (may users dont reach successful sign in in app). For the users who reach the step of successfully signing in on app, there is no problem in tie-ing the usage between browser and app.  As I tested Amplitude is giving different device ID to browser and the app though usages is from same machine. Any way to get unique machine specific info to tie a user in a non logged in scenario moving between browser to app?

Or any other suggestions to tie the usages on signed in browser user and non signed in app user (same user on same machine) ?

 

Thanks

Kavitha

Reply