Illustrative example
A reminder problem that wasn’t really about reminders.
This fictional example shows how a Retention Sprint could move from a broad week-two drop-off problem to a specific behavioral diagnosis and a small experiment. The numbers and product are synthetic. The method is real.
Week-two adherence drop-off
Illustrative example, not a client result
Observed signal
42% of reminder dismissals happen during the morning commute.
The surface problem looks like weak reminders. The behavioral barrier may be the timing and context instead.
Does the moment allow it?
The reminder lands during the morning commute, while the medication is still at home. Most dismissed reminders are never reopened.
Experiment to run
Let people choose a reminder tied to a place or routine instead of a fixed clock time.
The starting evidence
A fictional medication-support app sees healthy setup completion, then a sharp fall in daily confirmations during week two. The team’s current idea is to make reminder messages more urgent.
Support notes add a different clue: people often say they notice the reminder while commuting, dismiss it, and forget to return. Interviews also suggest that the confirmation starts to feel like admin once the novelty wears off.
The sprint does not treat those clues as proof. They are enough to write better hypotheses and choose what evidence the next test should collect.
Target behavior
After receiving a dose prompt, an established patient records or resolves the medication action while it is still relevant.
This is narrower than “use the app more.” It names the person, the moment, and the action the product can help with.
Three directions worth comparing
Routine-linked reminders
Addresses the opportunity barrier by moving the prompt to a moment when the medication is available.
Watch for: Routine changes could make the cue unreliable for some people.
Visible goal progress
Addresses the motivation barrier by connecting today’s action to the reason the person chose at setup.
Watch for: Weak or generic goal language would turn this into decoration.
Frictionless recovery
Addresses capability and shame after a missed dose by giving a clear next step without forcing people through the full flow.
Watch for: Clinical rules have to shape the recovery options before prototyping.
The first experiment
Prototype
During setup, let people attach the dose prompt to an existing routine such as arriving home. Show the chosen routine in the reminder.
Hypothesis
A routine-linked prompt will increase completed or intentionally resolved medication actions because it arrives when the medicine and attention are more likely to be available.
Measure
Compare prompt-to-action completion, time to action, and manual reminder changes against the current fixed-time flow.
Decision rule
Continue only if completion improves without creating more late, unsafe, or confusing actions. Clinical review comes before a live test.
The point is a better decision, not a clever diagnosis.
Your sprint would use your product, your evidence, and your clinical constraints. If the example feels close to a problem your team has, we can talk through the fit.