Been running into a particular situation in which I've been finding cross contamination of hits happening between different websites sharing the same reporting suite but they have different Adobe Launch properties.
The scenario is as follows:
'Website A' has 'Adobe Launch Property A' installed
'Website B' has 'Adobe Launch Property B' installed
Recently I introduced a new eVar in the reporting suite; let's call this 'eVar A'. The only time 'eVar A' is ever set is within 'Adobe Launch Property A'.
'eVar A' has never been set in 'Launch property B' in any rules.
After a day of collecting data and checking Adobe Analytics, out of a sample of approximately 30,000 Instance metrics of 'eVar A', I see 29,999 instances from 'Website A' . However, there is 1 instance of 'eVar A' reporting in 'Website B'.... how could that be possible?
By definition an 'Instance' metric is 'the number of times a dimension was explicitly defined in an image request' (https://experienceleague.adobe.com/docs/analytics/components/metrics/instances.html?lang=en). By that definition it should rule out any persistence scenarios. Also 'Launch Property B' does not have any rules pertaining to 'eVar A'. I would think there would be 0 instances of 'eVar A' on 'Website B'?
Would most appreciate any suggestions as to how this could be possibly happening and possible resolutions to this issue.
Off the top of my head, I'd say that the Website B page where that eVar A came from is using Launch Property A.
When you breakdown that 1 instance of "eVar A from Website B" by other dimensions, e.g. Page, Server, some other eVar, does the resulting report provide any further clues to help you discover what's going on?
Also, check that the attribution and expiration of that eVar have been set correctly.
I'm certain that Website B is not using Launch Property A and never has had it previously installed.
I actually had multiple brand new eVars I created, these were made exclusively for sites with Launch Property A only. However, it seems other sites that have Launch Property B installed still generate those exclusive eVars a handful of times.
Attribution and expiration of eVars might explain for other metrics, but 'Instances' are the number of times a dimension was explicitly defined in an image request. I think that rules out attribution and expiration possibility.
The only common denominator I can think of that is shared between Site A and Site B is they push data to the same reporting suite. I'm wondering could this be a data warehouse issue or Adobe Analytics reporting issue? Is there any settings/configurations I could potentially look at that govern data processing after Adobe receives the data request?
Did you glean any info when you broke down those eVars by other dimensions?
One other thing to check: Processing Rules in Analytics, to see if those eVars could be set there.
It might just be that your users are populating cookies on website B that still remember evars from website A.
Is it possible users in website A in a session can go to website B? If yes then I think and evars though not page level tagged in launch on website B might be picked up from initial cookie from Website A.
You may have to look at when the evar expires or maybe make them hit based(but lose some downstream session data). Also investigate if you create some rule to reset/clear those evars on users entering website B if their previous pages are from website A.