I am testing webhooks with sGTM. I see that by default a “client_id” is generated on each request (via cookie), which is referring to the Data Client ID. I have also enabled the power-up of Stape User ID. These are the two formats that I am seeing:
Data Client ID (client_id): dcid.1.1714481809622.426939670
Stape User ID (x-stape-user-id): 46e0e1ee94649918e4cc64fcf79d1888
What is the difference between these 2 variables? In which situations would we want to use one over the other? Does it make sense to use both variables?
As far as I understand, the Data Client ID is constructed with the unix timestamp and a random number. And the Stape User ID is constructed by the hashed combination of User IP, User Agent, TLS settings and Hostname.
As far as I understand, the Data Client ID is constructed with the unix timestamp and a random number. And the Stape User ID is constructed by the hashed combination of User IP, User Agent, TLS settings and Hostname.
precisely so. Usage depends on your objective, given it’s webhook we’re talking about - both seem irrelevant in most cases.
Well, I think for a webhook setup where we will receive a GET request for a Purchase event with minimal data (user id & purchase amount) and enrich this data with the data stored in Stape Store (from previous user events), it is crucial to have a user identifier, right? Or am I missing something here?
@hustleou I still don’t think either is useful for you in a webhooks scenario, you should use something solid for ‘stitching’ such as customer’s email or your internal user_id.
@Chris_Bradley you can return the value of x-stape-user-id (or any other piece of data on the server) back to the web IF you need to, just FYI
Stape User ID can change if the user changes devices or browsers for example. While the IP is the same, the User Agent will change, and therefore the Stape User ID will change.
But then if I use some sort of internal “user_id”, how can I track this user across different devices or browsers? How is the user_id persistent in this case? (if there is no logged users)
‘internal user_id’ is usually assumed to be an actual id of the logged in user, if that’s not the case - there’s no dramatic difference between the two, as you’ve figured it out already
But to what exact headers parameters (names) are these referring to? For example, for the User IP, which one does it use?
Cf-Connecting-Ip
X-Forwarded-For
X-Geo-Ipaddress
I have Cloudflare as CDN (instead of Stape CDN in my container) and each one of these 3 request headers has a different value. The correct one in this case is “Cf-Connecting-Ip”. I wonder if the X-Stape-User-Id in this situation is actually being built correctly.
Ok, got it! Then the only way to make the X-Stape-User-Id consistent is to use the Stape CDN. Because if I use Cloudflare, probably the Real IP will not be grabbed correctly, as in the “X-Forwarded-For” header, it is showing a different IP than the Real IP.
