Principal Associate, Experience Design · PATH
Calibration sessions were held inconsistently across the enterprise. I designed a scalable system that brought structure, trust, and shared context into the room.
My role
Lead product designer
Timeline
Capital One · Dec 2024 – June 2025
Platform
Enterprise web platform
Directional signal
The problem
Scaling something human
Calibration is where performance decisions actually get made. Leaders gather, sometimes with a room full of peers, to align on ratings, surface standouts, and identify development gaps. It's high stakes, politically charged, and deeply dependent on the quality of information available in the room.
We'd proven with the pilot that better feedback inputs led to better calibration conversations. Now we needed to build a system that could hold that quality at 70,000-person scale, across business lines, seniority levels, and calibration models that looked very different from team to team.
My role: lead design for the end-to-end calibration ecosystem within PATH, our performance platform: from how sessions are created and managed to what leaders see in the room when decisions are being made.
01
Who's actually in the room
We started with deep discovery, not just on the calibration session itself, but on all the people moving through it. The experience looked radically different depending on whether you were a first-time people leader, a senior VP running a large org, or an HR partner facilitating the session.
We could have designed one calibration screen for “the user.” But calibration isn't one job. A people leader defending ratings, an HRBP coaching that leader, and a facilitator keeping the room fair each want different things from the same screen. So we split into three personas instead of averaging them into one, then narrowed scope to what a single MVP could actually serve.
Once the personas were clear, we mapped each job-to-be-done across the calibration phases: prep, live calibration, and post-calibration. That helped us decide which capabilities belonged in the first release and which needed to be descoped, so the MVP could focus on the moments where better structure would change the quality of the session.
02
Before the room: Prep for calibrations
Before PATH, calibration sessions were set up manually by HR business partners in Google Sheets. Each group had to be assembled by hand: which teams should calibrate together, which associates belonged in each session, who needed to attend, and when the conversation should happen.
We created a smarter setup system that let HR partners select the teams they wanted to calibrate together, then automatically generated session groups by level and job family. From there, they could edit the details that still needed human judgment: attendees, calibrated associates, session timing, and any edge cases before the room began.
03
In the room: live calibrations
Live calibrations are the hard part. We built this experience for two core personas: people leaders making talent decisions and facilitators responsible for keeping the session structured, fair, and moving. The room has multiple leaders, one shared screen, limited time, and years of performance data that need to resolve into actionable decisions.
The distribution view gave rooms a shared anchor. The individual view surfaced 360 feedback alongside performance data, so leaders could have real conversations instead of debating from memory. Status tracking let sessions pause and resume without losing ground.
One consolidated place to view live calibration data, including associate calibration profiles and ratings in a pop-out drawer format.
Drag-and-drop movement so people leaders and facilitators could move associates during live conversation, instead of managing the room manually in Google Sheets.
Distribution data that helped people leaders and session facilitators track each session by percentage and pie chart, making it clear whether the calibration met requirements set by senior org leaders.
04
Early signal from the field
We ran iterative research throughout the build, not just usability studies, but ongoing conversations with people leaders and HR partners as the experience took shape. The early feedback validated our core bets and surfaced edge cases we hadn't anticipated.
Once this experience launched, leaders described calibrations as feeling "more grounded": less debate about facts, more real conversation about development. HR partners noted fewer escalations post-session. The signal was clear enough to accelerate the enterprise rollout.
Outcome
From pilot to platform
PATH launched as Capital One's first enterprise performance management platform, built on the foundation the pilot established. The calibration ecosystem was the next main addition to the platform alongside the 360 feedback experience.
0 out of 0
Calibration sessions reported fewer disputes and stronger post-session alignment
↑ 0+
employees across Capital One, scaled from an 800-person pilot
0%
HR partners reported reduced escalations and cleaner outcomes in sessions using PATH
Directional figures from the pilot and early rollout; the 70,000+ reflects full enterprise scale.
What I learned
What enterprise scale taught me
Building this ecosystem taught me what it means to design at enterprise scale, where the system has to hold for every edge case, not just the happy path.
The work that mattered most wasn't the interface. It was the alignment work: getting HR, engineering, product, and business stakeholders onto the same model of what calibration should do and for whom. Design was the medium for that conversation.
I also learned that launching isn't the end. The most valuable research happened after PATH was in the field, when real users surfaced problems we never would have caught in a lab.
Once the platform existed, the next challenge was flow: helping leaders prepare for high-stakes conversations without stitching context together themselves.