Passing data values via URL for Adobe Analytics

[montezh2001]

Hello, 

I'm assisting a team with a possible issue they are trying to solve. They're wanting to do is pass various data fields (client id; tenant id, etc) from the their website (which sends a Transactional Message) via URL. They are hoping the data configured in the URL can be sent directly to Adobe Analytics. 

Example: https://experienceleaguecommunities.adobe.com/t5/[clientid/[tenantId] 

I think you can pass the URL as part of the CTX into Adobe Campaign, but my client is wondering is it a way for that data to go directly into Adobe Analytics for Reporting. In Adobe Campaign - this data would somehow be collected as part of the Click data that's capture by the Tracking logs. 

Is there any material you can point me too to assist with this request?

[Jennifer_Dungan · Accepted solution]

The first question I have is do you have URL patterns that would allow you to distinguish client id and tenant id from your folder structure..

Your example is:

https://experienceleaguecommunities.adobe.com/t5/[clientid/[tenantId] 

might you also have

https://experienceleaguecommunities.adobe.com/t5/something/[clientid/[tenantId] 
or
https://experienceleaguecommunities.adobe.com/t5/something/somethingelse/[clientid/[tenantId] 

URLs?

even if client id and tenant id are at the end of the url in that order, how do you distinguish urls that don't have them?

https://experienceleaguecommunities.adobe.com/t5/something/somethingelse

(you don't want to treat "something" like a client id, and "somethingesle" like a tenant id)

It's absolutely possible with a little script to parse URLs and to extract and pass values to Analytics, the trick is how do you parse them reliably....

If maybe there was a specific path structure like /cid/[clientid]/tid/[tenant id] or /cid-[clientid]/tid-[tenantid] that you could use to properly match the pattern against with a regex, that would be a better implementation....

It won't be out-of-the-box like Query String Parameters (?clientid=[clientid]&tenantid=[tenantid]) - there are specific launch rules for getting the value of any query string parameter:

But you can absolutely use JS to parse the URL and extract portions using regex or other string manipulators......

But here is the next question.. do you want all URL variants being recorded (in Adobe and any other analytics tools) to track every client / tenant variant as a unique URL value (this could also potentially have SEO impacts, even with having canonical URLs pointing to the URL without those value - Google reserves the right to ignore your canonical declaration if it feels like something else would be better).

You can of course override the g (page url) param with a canonical value using Launch.. you don't have as much control with other tools (so if you have a backup GA implementation, or FB Pixels, etc.. they will record the URL as seen - most of these can be configured to ignore query string parameters, but not to ignore what looks like actual folder structure....)

[montezh2001]

@Jennifer_Dungan - Thank you for the feedback. To answer your question, I'm not sure how the configutation would be, I was creating an example based on some pre-lim discussions I've had. 

I'll take your information and work with my team to see how they plan on getting the information into AA. This is really good. 

[Jennifer_Dungan]

Ok, I was hoping that was just a random example (if it was that way, I am sure your developers would have a lot of the same questions about how to parse the URL and make sure the correct content loads )

When you have a clearer idea of how its going to work, and you are still looking for help extracting the values, let us know!

[montezh2001]

Just wondering, if would it be easier to have them pass the URL, or just the id fields, through the Transactional Msg ctx, and then provide that information via ETL process into AA?