Grouping users into weekly cohorts

Rather than look at users as a single unit, Value Paths groups them according to the week in which they signed up. The upshot of weekly cohorts in visibility.

There's variability in what transpired over the last year of your business. Maybe you ran a promotion that brought in a group of very motivated customers. Maybe Netflix released a documentary about the problem you're solving and you saw a sudden spike in engagement. Maybe you run a seasonal business β€” you see different trends of engagement during a certain part of the year. Digging into the last year's worth of experiences that users have had with your product, you're likely to find peaks and valleys of progress β€” there will be weeks in which a larger percentage of users who were able to make their way through theΒ user timelineΒ than others. And others where a large percentage of users dropped off.

When you're measuring in weekly cohorts, you go from asking "what is my conversion rate?" to "what is my conversion rate for this group of people and why?" You can tell how effectively you're turning signups into customers on a weekly basis, and how many of those customers you're able to keep from one month to the next.

If you aren't measuring in cohorts, you can't see where things got better and where things got worse. Once the peak and valleys of the path are visible, you can start to form theories about why. The line of questioning typically leads to pinpointing the changes caused a particular group of users to make more progress along the user timeline at a particular time.

From: VPP Episode 04 (22.26)