A backlog of small feature tickets that kept losing on their own turned out to describe one coherent experience. I built it instead of writing it up, and product shipped their version.
Role
Built and shared the prototype.
Period
2026
Team
With product and engineering.
I owned
Synthesizing scattered customer tickets into one experience
The working prototype and every product decision in it
Making the case to product and engineering
The team owned
The production build
Prioritization and release timing
Shipping credit
Aggregated feedback is not a product direction until someone draws it.
The path
01Scattered feature tickets
02synthesized into one drawn prototype
03rebuilt and shipped by product
The problem
Enterprise customers had been asking for a better analytics dashboard for a long time, through separate tickets that each looked small on its own:
An export capability had been removed at some point and never replaced
Chart types were limited
Customers who did not measure on the platform’s default headline metric had no focus metric that fit how they actually worked
What I built
A working prototype of the whole experience: configurable chart types, a focus-metric selector for the underserved segment, export reinstated, and richer interaction inside the charts.
Then I shared it with product and engineering.
What changed
They took the concept, built the production version, and shipped it as a platform release. It closed a batch of customer requests that had been open for a long time.
I did not ship it and do not claim to have. What I own is the prototype and the direction it set. These tickets had lost their prioritization arguments more than once; drawn as one thing, the same requests became a decision the product team could make in a single meeting.
The pattern
That is the forward-deployed loop in one artifact: sit close enough to customers to hear the pattern in scattered requests, build something concrete enough to argue with, and hand it to the people who own the platform so it ships without your name on the commit.
The call
Decided
Build a working prototype of the whole experience rather than write up the requests.
Instead of
A document arguing that the individual tickets were related and should be prioritized together.
What it cost
The prototype's code was thrown away and I am not on the commit that shipped. A written argument would have been faster to produce and would have lost, which is what it had already done several times.