Metric · Hooked
Cadence-matched cohort retention
Share of a sign-up or activation cohort that does the value event in a later bracket matched to the product's natural cadence (for example week 4 or week 8). Read by segment for curve shape and across cohorts for stacking.
Plain label: cohort retention at the natural cadenceAlso known as: retention curve, week-4 retention, week-8 retention, bracket retention, day-N retention
Pin four choices first. Bounded (exactly day N), unbounded (day N or later) or bracket (a window such as days 22-28, best for weekly products); "active" means the value event, never a login; the denominator (all sign-ups, activated users or payers); and the cohort grouping (start week, channel, plan, segment).
Read the shape per segment. Declining to zero, flattening or a smile (the Stage Fix curve signal). Then compare cohorts at the same age: newer cohorts above older ones is the best sign the product is improving.
Use as a guardrail. Week-2 retention of the treated cohort guards onboarding changes against activating users who then leave.
Amplitude's 2025 data also found that 69% of top day-7 performers were top three-month retainers (benchmark confidence B).
Benchmarks
Every benchmark carries its year and a confidence grade for its source. Use them to calibrate, not as targets.
| Context | Median | Top 25% | Top 10% | Source |
|---|---|---|---|---|
| day-7 return of ALL tracked new users, including drive-by visitors; not comparable with sign-up-based rates | — | 7% | — | Amplitude 2025 Product Benchmark Report, 2025Benchmark confidence B. A reputable dataset. |
| mobile apps, all installs, day-30 (bounded); definitions vary between vendors | 6% | — | — | Aggregators citing Adjust and AppsFlyer, 2025Benchmark confidence C. A secondary source. |
Limits
What we don’t claim
- The Hook Model is a practical design tool, not a validated or complete theory of behavior.Hooked, Introduction
Sources
Hooked · Nir Eyal
Ch.8
The 7% retention rule (2025 Product Benchmark Report) · Amplitude, 2025
Benchmark confidence B. A reputable dataset.What is good retention · Lenny's Newsletter (Lenny Rachitsky and Casey Winters), 2020
Benchmark confidence B. A reputable dataset.
Source IDs: HK-MEA01
Book content is paraphrased, never quoted at length. Last verified October 24, 2026.
Your product
Find the stage your product is stuck in
Gamify It reads your product, picks the one move that matters at its stage, and sizes the test to your traffic. Connecting is free; a Pro subscription starts it.