← All work

Building a UX Center of Excellence at Comcast ULearn

SUS on Touched Products
68+
UX Drag Reduced
33%
Completion at Baseline
<20%

The content was technically fine: instructional designers had built it carefully, compliance had approved it. But fewer than one in five employees who started a course finished it, and nobody could say why, because nobody was measuring the experience. ULearn was Comcast’s bet on replacing a legacy SAP system during the Great Resignation. The new UX function had to prove, quickly, that it was worth keeping.

I was brought in to help structure and operationalize that function. Over time, that work evolved into a recognized UX Center of Excellence: a team that could deliver UX work directly, introduce user-centered methods, and teach the broader organization how to apply them.

Using measurement to make UX credible

Completion on started content was below 20%, but completion alone was not enough to win over skeptics. I introduced SUS as a usability benchmark to make UX more measurable and defensible. Untouched products typically scored in the 40s to 50s, while products we directly iterated on could reach 68+ within two to three sprint cycles.

Structuring the function to scale

The UX team had mixed seniority and limited maturity, with six designers total: two senior and four much more junior practitioners. I helped define the function as a Center of Excellence and introduced a tiered engagement model so teams could get the right level of UX support without every request becoming a full redesign effort.

UX Center of Excellence framework — four core focus areas defining how the function partnered with the business.
The CoE framework. Four focus areas that gave stakeholders a shared language for what UX was for, how it partnered, and where it invested.

Reducing UX drag and expanding capability

Early on, UX could add a full sprint to delivery timelines. By reshaping the service model, bringing in stronger external talent, and building enablement paths for lower-complexity work, we reduced that burden to roughly two to three days over six months, a 33% reduction from peak. I also helped extend UX capability beyond the core team through coaching, facilitation, and a move from Adobe XD to Figma.

2 weeks Baseline 1 week 6 weeks in 2–3 days 6 months in
UX drag: the net delivery time UX added to sprints. Started at a full sprint. Halved by week six. Down to a couple of days by month six, and by then fully absorbed into sprint planning and metrics.
Phased UX enablement plan — building shared UX capability across instructional design and adjacent teams.
Phased enablement plan for extending UX thinking into instructional design and adjacent teams. Shared methods, frameworks, facilitation paths.

Results and impact

  • Helped establish Comcast ULearn’s UX function as a recognized Center of Excellence
  • Introduced usability measurement through SUS baselines and iterative testing
  • Improved touched products from typical SUS baselines in the 40s–50s to 68+
  • Reduced UX drag from roughly one sprint to two to three days over six months
  • Reduced UX drag by 33% from peak while the function continued to grow
  • Expanded UX capability beyond the core team through coaching, facilitation, and shared methods

Reflection

This work taught me that early-stage UX functions in enterprise environments earn trust by becoming measurable, understandable, and sustainable. At Comcast ULearn, I helped make UX legible to the business: first through evidence, then through service design, and finally through enablement that let the function scale beyond the boundaries of the team itself.

Still reading?

Send a note. I'll write back.