Eleven interviews asked for the same feature. Maya builds exactly what they described.
←→ to move through the storySwipe or use the arrows to move through the story
Everyone Loved It For Nine Days — Volume I, Understanding Humans Before Products · Why Humans Change · Failure & recovery · Core. Maya ships a weekly summary that interviews praise and usage abandons after week two, and finds the feature is genuinely useful but attached to nothing that already happens
01 · Opening
Eleven interviews asked for the same feature. Maya builds exactly what they described.
02 · The setup
Eleven interviews. Every one of them asked for a way to see what changed across the week without opening four screens.
Priya: If it just told me what moved, I would look at it every Monday.
03 · The setup
Maya thinks: Eleven for eleven. I have never had a signal this clean.
04 · The evidence
The summary ships. Week one looks like the interviews promised.
Weekly Summary — opens per user — Week 1: 3.1; Week 2: 1.4; Week 3: 0.3; Week 4: 0.2
05 · The evidence
Maya: Nine days. It took nine days to go from three opens to nearly none.
06 · The evidence
Sam: Did it get worse? Slower, uglier, wrong numbers?
Maya: It is exactly the thing they asked for. Nothing changed.
07 · You make the call
The story does not tell you first.
A. The interviews were leading and the demand was not real
B. The feature delivered value but nothing in their week pointed to it
C. The summary was not valuable enough to be worth the click
08 · What happened
Maya goes back to the same eleven people and asks a narrower question: not what they think of it, but when they last opened it.
Maya thinks: They all still say they like it. That is the part that stops making sense.
09 · What happened
Priya: I meant it. I just never think of it. Monday starts and I am already in the standup doc.
10 · What happened
Eleven versions of the same sentence. They wanted it, and their week never contained a moment that pointed at it.
Maya thinks: I built a destination and no road.
11 · What happened
What already happens — Name a thing the user does every week without being asked. The standup doc, the Monday call, the invoice run. It has to exist before your feature does.
What points at you — Ask what in that moment could carry your feature. If the answer is 'nothing, we would add a prompt', you have not found the moment yet.
What happens if you say nothing — A habit you have to keep reminding people about is not a habit. If removing every prompt kills the usage, the trigger was yours and never theirs.
12 · What happened
Sam: So we email it to them every Monday and the problem goes away.
Maya: That buys opens while we keep paying for them. I want the week to do the reminding.
13 · What happened
The summary moves into the standup doc the team already opens every Monday. Nothing new to remember, and the opens hold.
Maya thinks: Same feature. It just stopped needing me to announce it.
14 · Complete
Value earns the first visit and a trigger earns every visit after it — and the trigger has to be something their week already contains. Next: what happens when the product decides for them and calls it a preference.