Rapport · Manager dashboard
Rebuilding Rapport around manager decisions.
Fewer than 10% of managers were using the existing product. I rebuilt the experience around the questions they actually needed to answer, built a working first version seven hours after the written plan, then improved the processing behind it.
ReactManager dashboard
PythonProcessing pipeline
PostgreSQLEvidence + checkpoints
The investigation
From the question a manager needs answered, to a first version, to a sturdier system behind it.
A question worth asking. The words behind the answer.
<10% manager adoption at the starting point
Customer chasing an update
Alex Morgan · sample visit
“I was expecting a call yesterday. Can you check the repair status?”Customer · exact words in this example
Check the promise and what happened next before following up.
How the work unfolded
From question to a reliable, usable experience.
- DefineCapture the real manager questions.
- BuildShip a focused first version in 7h.
- IterateAdd processing, integrations and performance.
- ValidateKeep the work that proved useful.
The outcome
A fast first version. A deeper system behind it.
- 28 signals linked to their evidence
- Recoverable processing
- Smaller requests, less stored data
Seven hours from the written plan to a working first version. Subsequent iterations connected 28 signals to source evidence and made processing recoverable, with smaller requests and less stored model-call data.
Key takeawayAnswer the manager’s question first, then strengthen the system behind it.