From ad-hoc to operational: building a UX delivery practice
When I moved into leadership, UX entered the team through informal, request-driven work — with no consistent intake, research, onboarding, or accessibility checkpoints. I designed and rolled out an end-to-end delivery practice that made the team's work faster, more consistent, and measurably more influential on the products we shipped.
🔒 Some specifics are generalized to respect confidentiality. Happy to walk through the details in conversation.
Context
I lead product design for a team of 5+ designers supporting 8+ enterprise identity products across regulated, public-sector environments. The team was talented — but under-served by process. Work arrived informally, priorities shifted without a shared framework, and there was no repeatable path from a new request to a delivered, validated, accessible design.
The challenge
Informality has a cost. Without a defined intake or shared cadence:
- The team absorbed whatever came its way, with no consistent prioritization.
- Research happened inconsistently, so decisions leaned on assumption over evidence.
- Accessibility was checked late, if at all — risky in a public-sector, compliance-heavy domain.
- Onboarding a designer to a new product was tribal knowledge, not a repeatable step.
The result was inconsistent quality, slower delivery, and UX that struggled to influence decisions — because it wasn't embedded early enough to matter.
How I led the change
I'd started closing these gaps before I had the title — building our design system around the company's branding standards and piloting improvements with my manager's support. When I stepped into leadership, I formalized the work into a single, end-to-end practice and drove its adoption.
I defined a repeatable pipeline that carried work from intake all the way to delivery, with research and accessibility built in rather than bolted on:
- 1Intake
- 2Preliminary research
- 3UX onboarding
- 4Agile delivery
- 5Intermittent research
- 6Accessibility evaluation
- 7Delivery
I backed the pipeline with governance and steering documentation so it was repeatable, standardized UX/UI across North America against our global design system, and established formal partnerships with the teams UX depends on — DevOps, Product, Training, and Hardware — so design was plugged into the wider delivery machine, not operating beside it.
The outcome
With the practice in place, the team's output didn't just get more consistent — it got more influential.
- Major UX research delivery grew 120% (5 → 11 projects in a year).
- That research shaped real decisions: ~40% drove new feature development, ~40% supported customer requirements, ~20% validated UX patterns ahead of build.
- Accessibility evaluation and documentation nearly doubled (12 → 23 deliverables), strengthening compliance and customer-facing RFP responses.
The informal function had become a repeatable delivery engine — a well-oiled machine.
What it taught me
Process isn't bureaucracy when it's designed well — it's what frees a team to do its best work. Give people direction, patterns, and consistency, and quality stops being heroic and starts being the default.