CDP + GA4

[BillKVMotorCo]

Has anyone found a CDP that has an in-depth integration with GA4? For instance, I don't just want to know that there was a VDP view. I want to know the key events that occurred on that VDP view from the CDP itself rather than having to go into GA4 and find it.

 

[Ryan Everson]

Are you wanting to click on a user profile and see all of the events associated with the user? That's typically how a CDP would work.

Might be worth looking into PostHog, it's a very inexpensive (and often free) web analytics platform that is a little more advanced than GA4 and offers some CDP-like functionality.

PostHog - How developers build successful products

PostHog is the only all-in-one platform for product analytics, feature flags, session replays, experiments, and surveys that's built for developers.

posthog.com

It would require installing their code and setting up events, conversions, etc though since they are a competitor to GA4.

 

[LocalWerks]

Has anyone found a CDP that has an in-depth integration with GA4? For instance, I don't just want to know that there was a VDP view. I want to know the key events that occurred on that VDP view from the CDP itself rather than having to go into GA4 and find it.

Do you want to see on which VDPs a Key Event occurred or do you want to see Key Events that occurred from users who viewed a certain VDP?

 

[BillKVMotorCo]

I would like to be able to see what key events a customer fired inside the CDP.

 

[Ryan Everson]

I would like to be able to see what key events a customer fired inside the CDP.

Posthog or Matamo would be your cheapest/easiest options - they are both GA4 alternatives that offer the ability to view a single user's journey and events (I'd lean towards Posthog). They offer a little better experience than GA4's user explorer reports.

But if you want a true full-blown CDP, you could use something like Hightouch to ingest your GA4 data through BigQuery (using GA4's native connector).

You will also have to decide how deep you want to go down the identity resolution rabbit hole.

 

[BillKVMotorCo]

Posthog or Matamo would be your cheapest/easiest options - they are both GA4 alternatives that offer the ability to view a single user's journey and events. They offer a little better experience than GA4's user explorer reports.

If you want a true CDP, you could use something like Hightouch to ingest your GA4 data through BigQuery.

Would I be able to see these actions by a specific user within my CDP, though? That's more what I'm looking for.

It is good to know that I have a way of seeing it by user even without knowing who that user is, though.

 

[Ryan Everson]

Would I be able to see these actions by a specific user within my CDP, though? That's more what I'm looking for.

It is good to know that I have a way of seeing it by user even without knowing who that user is, though.

Yeah that would involve identity resolution using a mix of deterministic and probabilistic matching.

Usually you'd start by adding a URL parameter to all CRM communications to connect your first-party customer data with their device(s). When a customer clicks a tagged link, all their past anonymous data can be backfilled.

Lead form submissions are also an obvious way to identify users.

Once you go beyond those two simple deterministic methods, that's when you'd need probabilistic matching - using techniques like identity graphs and anonymous visitor identification.

In automotive, a CDP is a big lift - especially with the lack of open APIs. Fullpath probably has the best turn-key solution for automotive right now, but they're also limited by the closed-off ecosystems from automotive vendors.

And that's just to get the data ingested and unified. To actually generate an ROI, you then have to activate the data.

 

[BillKVMotorCo]

Yeah that would involve identity resolution using a mix of deterministic and probabilistic matching.

Usually you'd start by adding a URL parameter to all CRM communications to connect your first-party customer data with their device(s). When a customer clicks a tagged link, all their past anonymous data can be backfilled.

Lead form submissions are also an obvious way to identify users.

Once you go beyond those two simple deterministic methods, that's when you'd need probabilistic matching - using techniques like identity graphs and anonymous visitor identification.

In automotive, a CDP is a big lift - especially with the lack of open APIs. Fullpath probably has the best turn-key solution for automotive right now, but they're also limited by the closed-off ecosystems from automotive vendors.

And that's just to get the data ingested and unified. To actually generate an ROI, you then have to activate the data.

Tell me you're smarter than me without telling me you're smarter than me. Seriously, you're good at this stuff, dude. Wow.