← Back to work
Sprout Social Case Study · 03 2023 → 2024

Working inside out and outside in.

A hard accessibility deadline landed on my Design Systems team in the middle of a company-wide AI push. We got the bug work done. Meanwhile, I pitched the org structure that meant we wouldn't need to do this again — expanding Design Systems into a UI Foundations org with two teams instead of one.

Role
Sr. Director
UI Foundations
Timeframe
2023 — 2024
Focus
Org design
Accessibility program
Cross-team feedback loops
Outcome
UI Foundations org
Design Systems + UI Integrity
Deadline met
UI Foundations org
Team A
Design Systems
inside out
Components, tokens, patterns, docs. The assets that scale.
← →
feedback loop
← →
Team B
UI Integrity
outside in
Migration support, a11y guidance, universal patterns.
01 / The setup

Two mandates hitting at once.

Sprout Social was pushing to integrate AI into every product area — a company-wide directive with real momentum behind it. At the same time, we had a hard external deadline for accessibility compliance across the whole product suite.

Both were real. Neither was going to move. And the Design Systems team — my team — was asked to close the accessibility gap. Feature by feature, product area by product area, bug by bug.

02 / The ask

The wrong team, for the right reason.

On paper it was an ask we couldn't refuse. No one else in the org had the accessibility expertise, and no other team had the bandwidth to hit the volume of bugs on the calendar. So we said yes.

But it was the wrong use of a horizontal team. Design Systems exists to scale — to make components, patterns, and guidance that a hundred product teams can use. Sending it into individual product areas to hunt individual bugs is the definition of unscaling.

We did the work anyway. The accessibility deadline was non-negotiable and the imperative was clear. But while we were fixing bugs, I was drafting a memo about why we shouldn't be.

03 / The proposal

One org. Two teams. Two orientations.

I pitched expanding Design Systems into UI Foundations — an org that would hold the horizontal work under one roof, with two teams shaped for two different jobs.

Design Systems would keep working inside out: building the assets that scale, out to the whole product. Central ownership, shared language, the buffet the company self-serves from.

UI Integrity would work outside in: going into product areas to help teams migrate to newer components, giving accessibility guidance, spotting the shared experiences (nav, notifications, cross-product patterns) that should become system patterns, and closing the feedback loop back to Design Systems.

Neither team could exist alone. DS without Integrity means the system ships and nobody adopts it well. Integrity without DS means great advocacy for components that don't exist yet.

04 / How the two teams work

Two shapes for two jobs.

Team A · inside out
Design Systems
Center of the product, working out.
  • Components, tokens, patterns, documentation.
  • The language the whole product is built in.
  • Central ownership, contribution paths for subsystems.
  • The buffet the whole company self-serves from.
Team B · outside in
UI Integrity
Edges of the product, working in.
  • Migration support for product teams adopting new components.
  • Accessibility office hours and remediation guidance.
  • Spotting universal patterns (nav, notifications, shared modals) and pulling them into the system.
  • The feedback loop from the edges of the product back to the center.

"Working inside out, and outside in."

— The strategy that let one org do two jobs without doing either badly
05 / What happened

The pitch landed.

The expansion got approval. UI Integrity spun up as a sibling team to Design Systems, both under UI Foundations. The accessibility deadline was met, and — critically — the next time a company-wide product initiative landed, Design Systems didn't get pulled off its scale work. UI Integrity was the right home for cross-product execution.

The feedback loop showed up in the data too: annual user survey engagement went up, and users reported clearer understanding of both how to use the system and how to get help.

Bigger picture, the two-team model gave Sprout a durable shape for horizontal UX work — one that could absorb the next unexpected ask without breaking the team that had to say yes to it.

06 / What I learned

Notes from the pitch.

Sometimes the right move is to do the wrong work — first. Refusing the accessibility ask would have been correct on paper and disastrous in practice. Do the work. Then fix the reason it landed on you.
An org proposal is a design deliverable. Structuring a team is a design problem: inputs, outputs, feedback loops, adoption. My best pitch for UI Foundations came out of treating it like a design brief — not a headcount request.
"No other team could do this" is a diagnosis, not a defense. If a certain kind of ask keeps landing on the wrong team, the ask isn't the problem. The org is. That's leverage — if you're willing to name it.
Two smaller teams that fit their work beat one bigger team that doesn't. UI Integrity is smaller than an expanded Design Systems would have been. But it's shaped for its work, and that shape matters more than the headcount.
Next case study
04 · Gusto + Sprout Social
Turning table-side service into a help-yourself buffet.
Support process at scale
2021 → 2024