← Community

Diagnose activation

Find where the first session dies, name one bottleneck, and run one experiment — instead of "improving onboarding." Use when signups do not come back, when a launch looks busy and empty, or before you add another empty-state tour.

---
name: diagnose-activation
category: business
description: >-
  Find where the first session dies, name one bottleneck, and run one
  experiment — instead of "improving onboarding." Use when signups do not
  come back, when a launch looks busy and empty, or before you add another
  empty-state tour.
---

# Diagnose Activation

Activation is **the first time a new account gets the job they signed up
for.** It is not "engagement," not "time in app," and not a completed
profile. If they can leave after one session with the outcome they wanted,
they activated. If they cannot, nothing you add to week four will matter.

Most onboarding work fails because it treats a funnel as a vibe. The
method is a count at each step, then one change at the worst drop.

## Draw the first session as a spine

Four to six steps, named as **user-visible outcomes**, not screens:

1. **Arrived** — they can see the product (signed in, not an error).
2. **Aimed** — they picked or understood the job (a template, a first
   capture, an import).
3. **Acted** — they did the core action once (not "clicked around").
4. **Got the value** — they saw the result of that action (the note in
   the list, the file in the folder, the invite accepted).
5. **Could return** — they have a reason and a way back (email they
   wanted, a bookmark that works, a team that exists).

Your spine will differ. If you cannot write it, you do not know the
product job. Do not start with tooltips.

Count **accounts**, not events, through a recent cohort (a week of
signups, not all history). Where the percentage falls off a cliff is
the bottleneck. There is almost always one cliff. Improving a step that
is already 90% is theatre.

Shared publicly by the author and open for any use. See the Terms.