8 Comments

Excellent article, all-encompassing!

Expand full comment

Thanks!

Expand full comment

I think creating user states tables over time with stable definitions (or even changing definitions as the product evolves) could be game changing for a lot of products. Certainly would have helped me a lot in my previous role.

Also, this was really well written. Thanks!

Expand full comment

Don't ask me - I would have loved to use this in some of my former projects. But I can do it now - the future is bright.

Expand full comment

Hey Timo! Very clear article on the topic!

When it comes to technical implementation, we know that no web or product analytics tool is perfect right out of the box. Sometimes, metrics like "Active Users" in GA4 (the one I know) might not fully align with the specific needs of your product, but they can still be useful to some extent.

Another point to consider is the possibility of tracking different user states using the User ID stored in a first-party cookie within the Data Layer (DL). Depending on how you define the different user statuses, you could adjust the user status parameter in the event user properties dynamically. However, I'm not sure how easy or complex this implementation would be...

Expand full comment

Out-of-the-box "Active users" are rarely helpful since they take any event into account.

I would not try to track user states. As the post describes, user states must be modeled and adapted over time. Therefore, it is easier to do this later than during implementation.

Expand full comment

Thank you Timo!

Although I'm not sure what do you mean implementing "...later than during the implementation". Because you must have some kind of mechanism implemented to know that a "user state" changed/modeled.

Expand full comment

You implement proper event data tracking. Based on this data, you can model the user states later.

Expand full comment