Design Operations

From Ad-Hoc to Automated: Architecting the DesignOps Infrastructure that Scaled Dell’s Creative Velocity.

When I joined Dell, our 16-person design organization had no intake process, no capacity planning, and no single source of truth for work in flight. I built the team's design operations practice from scratch — process documentation, a Workfront implementation, formalized review cadences, and component governance. Our creative team was drowning in high-volume, low-complexity requests. Designers were bogged down by manual status updates, scattered feedback loops, and inconsistent asset handoffs. 40% of design time was lost to administrative logistics, resulting in burnout and delayed project timelines.

The design organization was 16 people strong: a director, a design manager, a principal UX designer, five senior UX designers, four UI/UX designers, a junior designer, two researchers, and two UX/front-end developers.

The talent was there. The infrastructure wasn't.

Work arrived through every channel imaginable — email, hallway conversations, meeting asides. There was no intake form, no project brief, no RACI. Nobody could answer basic questions: What's in the queue? Who has capacity? What shipped last month? Stakeholders had no visibility into their requests, and designers had no protection from scope creep or competing priorities.

Discovery: diagnosing before prescribing

Rather than importing a playbook, I started with listening.

Internal interviews. I sat down with every member of the design team and asked the same core questions: What's keeping you from completing work? What's confusing? What's broken that isn't a tool or hardware problem? That last framing mattered — it pushed the conversation past "we need better software" toward the process gaps underneath.

Stakeholder interviews. I also interviewed our external stakeholders, focusing on one measure: how confident are you that you know the status of the projects you've requested? The answer was a consistent, uncomfortable "not very."

Two truths emerged. Designers were losing time to ambiguity — unclear requests, unclear priorities, unclear ownership. And stakeholders had lost trust in the system because there was no system to trust.

I focused on four foundational moves, sequenced so each one reinforced the next.

1. Documented the end-to-end process
   Before changing anything, I mapped how work actually flowed — from intake through backlog, active work, review, and delivery. This became the shared reference point for every conversation that followed. You can't fix a process nobody can see.
2. Implemented Workfront as the single source of truth
   Partnering with our Workfront administrator (who handled system configuration), I designed the project templates — structured by project type and role — so that every request entered the same front door with the same required information. Intake, briefs, and assignments now lived in one place, visible to everyone.
3. Formalized the review cadence
   Reviews had been ad hoc: scheduled whenever, repeated indefinitely. I established a hard cadence with fixed dates and capped reviews at three per project. If alignment wasn't reached by the third review, the stakeholder and designer escalated together to design management — which doubled as a component governance checkpoint. The cap changed behavior: stakeholders came to reviews prepared, and feedback got consolidated instead of trickling in.
4. Created a component governance checklist
   Every project had to build with existing AEM components by default. If a designer believed a new component was genuinely needed, there was now a defined process for proposing, evaluating, and creating it — protecting the design system from one-off sprawl while keeping a clear path for legitimate evolution.

Adoption and pushback

The team's response was telling: almost no internal resistance. The pain was so universally felt that the fix sold itself.

The friction came from stakeholders reluctant to learn a new tool. Rather than running Workfront alongside the old systems indefinitely — which would have quietly killed adoption — we deprecated TFS and Clarizen for creative projects entirely. With one system as the only path to getting design work done, adoption followed. The validation came later: the entire company eventually moved to Workfront, extending what began as a design team initiative into enterprise infrastructure.

Outcomes

Speed:

• DAM requests and image retouching turnaround dropped from 1 week to 3 days
• New component creation dropped from 4–8 weeks to 2–4 weeks, depending on complexity

Visibility and trust:

• Every project now had a documented intake, brief, and owner
• Stakeholders could self-serve project status instead of chasing designers for updates
• Design management gained real capacity data for planning and prioritization

Durability:
The Workfront implementation outlived the initiative, scaling from one team's workflow to the company-wide standard