When a working feature isn't working.
At FishyVisions, I spent two years developing Quely, a SaaS platform that helps software teams clarify requirements and collaborate around Jira tickets.
One of the features I built was a real-time chat. Initially, it occupied its own page. It worked technically, but people rarely used it.
The problem wasn't the chat itself. It was how we had integrated it into the workflow. Users needed to discuss Jira tickets while looking at them, but our interface forced them to switch between the two.
Rather than adding more functionality, we reconsidered the experience. We studied how products like Slack and Discord approached chat, explored ideas in Miro and held a design jam with our product manager and designer. We sketched three alternatives, discussed them with the wider team and gathered feedback.
The designer and I then refined the new direction together, and I implemented it.
The redesigned interface kept the chat overview visible without getting in the way. Users could follow conversations and work with Jira tickets simultaneously, without constantly switching context.
- 01
Jira ticket
Where the work happens
- 02
Chat page
A separate tab
- 03
Discussion
Away from the ticket
- 04
Update ticket
Back in Jira
switch back and re-read the ticket
The change made the feature much more useful.
What stayed with me was that the solution wasn't more engineering. It was understanding why people weren't using something we'd already built.
NEXT DECISION