Preparing for a single calibration meant moving through 10 screens. I helped collapse the work into two connected screens with every input a leader needs.

Our design team ran a three-day, in-person discovery sprint with one goal: a high-fidelity prototype by day four. Two product partners kept the work anchored in real needs while another designer and I shaped the framework. I then led a team of four designers in translating that framework into an interactive prototype.

To build calibration materials, a people leader had to work across the whole system. A table-based “My Team” list was just the index: from it, they clicked into a feedback flow to review each associate's responses, looked through results on separate screens, then moved through a long profile-building flow to assemble the materials needed for one calibration conversation.

All told, 10 screens for a single conversation. Information was split across all of them, so leaders had to hold the synthesis in their heads: high cognitive load, endless clicking, and judgment formed from scattered context.

The sprint surfaced the core friction: the legacy design fundamentally clashed with the People Leader mental model. The old system relied on table-based navigation, but research showed that leaders don't evaluate a spreadsheet of people: they assess each associate holistically, one person at a time.

That reframed the whole problem. If the experience mirrored how leaders actually think, pulling self-evaluation, peer feedback, and their own perspective into one place per person, the workflow would feel intuitive, and calibration would get more objective as a result.

Now, when a leader opens the My Team page, they see every stat they need for calibration at a glance. From there, a single associate view lets the leader review feedback, results, strengths, and calibration notes in one place instead of jumping between screens.

The biggest interaction decision was replacing the old step-by-step profile flow with a workspace: the evidence stayed visible on the left, while the calibration profile opened beside it on the right. It made the screen denser, but it solved the real problem: leaders needed to interpret performance data and write their point of view at the same time.

We had to make a tradeoff: a 50/50 layout meant less room for each side, but it kept the leader in the evaluation context. The outer tabs kept results, competency feedback, insights, calibration profile, and performance review sticky as the page scrolled. Inside the profile, a second set of sticky tabs let leaders move through the sections they needed to complete without losing their place.

The drawer also carried state. A profile checklist showed where the leader was in the process of gathering and interpreting evidence, and the ready-for-calibration toggle stayed unavailable until the required inputs were complete. Once the profile was marked ready, the form locked; turning the toggle off unlocked it for edits.

Earlier versions relied on hyperlinks, but every link pulled the leader out of the associate view. I helped move the core subsections into tabs so leaders could switch between results, feedback, insights, and profile-writing without losing the person they were evaluating.

Results were standardized into a consistent hierarchy: title, description, then impact. Feedback was organized by competency instead of splitting strengths from opportunities, because competency was the common ground across every feedback source. Within each competency, filters let leaders narrow by feedback type, relationship type, and provider, while expandable sections handled long comments and empty rows stayed visible as “No feedback available.”

The redesign brought a leader's entire calibration prep onto a few connected screens, with the inputs for a fair assessment side by side instead of scattered. Delivered as a high-fidelity prototype in 72 hours (what typically takes weeks of fragmented virtual work), its impact carried well past the sprint:

The biggest learning was that speed did not come from skipping discovery; it came from compressing the distance between research, product decisions, and design execution. The redesign reduced prep from 10 screens to 2, created a 76% efficiency gain for people leaders, and became a reusable sprint model for related PATH workflows.

It also reinforced what makes redesign work stick: not just a cleaner interface, but a better mental model. Once we framed the experience around how leaders actually assess people, the path from ten screens to two became much easier to defend.