We put our whole team on a skill-will matrix — and backed it with delivery data

Every engineering org keeps a version of this spreadsheet: names, roles, products, and some private judgment about who's flying and who's struggling. Ours had 26 people across four products, with a skill-will code per person — high/low effort crossed with high/low skill.

We moved it into UltiPad in an afternoon. The roster isn't a separate list to maintain: members and their product assignments come straight from workspace permissions, so the team page can't drift from reality. Admins overlay the judgment — role, skills, and a performance band — and the bands are stripped server-side for everyone else. Sensitive data stays sensitive.

The matrix view

One toggle renders the four-quadrant chart: stars top-right, coaching cases mapped honestly, every chip clickable to edit. Filter by product to see a single team's shape. For a manager, this is the annual-review conversation made visible year-round — and updateable in two clicks when someone grows.

Judgment, meet evidence

The bands are opinions. So UltiPad puts delivery evidence beside them: for each member, how many owned ideas actually reached Done, the on-time rate against due dates, median cycle time from In Progress to delivered, rework signals (work bouncing backwards from QA), and current in-flight load — all computed from stage transitions the tool already records. Nobody fills in a timesheet.

When you open someone's edit dialog to set a band, their scorecard is right there. The tool doesn't rate people — sample sizes are small and context matters — but it makes sure the human doing the rating is looking at what shipped, not at who talks loudest in standup.