Activation: The First Session

By the end of this lesson, you should be able to: find a candidate activation moment from data instead of guessing one, say why the strongest-looking candidate is usually the wrong one to build on, explain what the analysis can and can't prove, and pick an activation window that fits the product rather than the calendar.

The step where half your users disappear

Last lesson ended on a number worth sitting with. Of the users paid social brought to Basket, only 46.4% ever did anything at all, against roughly 83% from every other channel. They signed up, opened the app, and left.

That's activation, and it's the least glamorous part of the funnel. Acquisition gets the budget, retention gets the strategy decks, and the step in between quietly decides how much of either matters.

Here's the same cut across all five channels, measured as "did anything past opening the app in the first seven days":

Four channels within four points of each other, and one at less than half. Basket's largest source of new users is also the one where two thirds of them never begin.

Finding the moment instead of guessing it

Every product team has a story about its activation moment. Somebody read that a social network found "seven friends in ten days" and now your standup has an opinion about the magic number.

You can find yours from the data, and the procedure is simple enough to describe in an interview.

1. List candidate early actions. Cheap ones and committed ones. Don't prejudge which matters.
2. Fix a window. For Basket, the first seven days from signup.
3. Fix an outcome. Here, still active in week four.
4. For each candidate, split the cohort into users who did it in the window and users who didn't, then compare the outcome.

Run that over 17,804 new Basket users who have a full 28 days of history behind them.