When Embedding Expertise Breaks the System
Organizations regularly restructure in the name of speed—often by embedding Center of Excellence resources (design, program management, architecture) directly into delivery teams.
It’s a logical move. Fewer handoffs. Faster decisions. Work gets closer to execution.
It also fails more often than people expect.
Years ago, at a Fortune 200 financial services company, I was part of a leadership team that implemented this model at scale. I led Design, Usability, and Content, and we embedded UX resources directly into cross-functional delivery pods.
At first, it worked. Teams moved faster. Decisions came faster. Momentum improved.
Then the cracks showed—and they weren’t subtle.
Where It Broke Down
Local Optimization vs. System Outcomes
Teams became more effective within their pods, but alignment across teams weakened. Dependencies increased. What looked like acceleration at the team level introduced friction at the system level.
Loss of Strategic Throughline
With expertise distributed, prioritization fragmented. Without strong connective tissue to shared outcomes, teams focused on what was in front of them—not what mattered most across the organization.
False Sense of Speed
Delivery appeared faster in the short term. The cost showed up later—in integration, in rework, and in inconsistencies that required cleanup downstream.
Delayed Feedback Loops
We established an Executive Review Board to provide oversight. Teams brought work forward weekly, we reviewed it, and they moved on.
In practice, this slowed things down. Feedback came too late, and we missed opportunities for real-time collaboration that would have improved outcomes earlier.
Unintended Cultural Friction
In an effort to maintain quality, I kept a small group of design leaders centralized. In hindsight, this created distance. Instead of developing leadership within the teams, we reinforced an “us vs. them” dynamic between the center and the pods.
We also never fully established the standards we believed would guide the work—leaving teams to figure out too much on their own.
I was part of these decisions, and I own how they played out.
What We Misunderstood
The problem wasn’t the decision to distribute expertise.
The problem was assuming that structure alone would improve outcomes.
Distributing resources changes where work happens. It does not ensure alignment, consistency, or shared accountability. Without mechanisms to connect teams, reinforce standards, and maintain a clear line of sight to outcomes, distribution creates fragmentation faster than it creates speed.
What It Takes to Make It Work
Organizations that make this model work treat it as an operating model shift—not a structural adjustment.
They:
Maintain clear ownership of system-level outcomes—not just team delivery
Create real-time feedback loops that complement (or replace) periodic reviews
Invest in and uphold shared standards, vision, tools, and enablement across functions so teams can operate consistently and efficiently
Develop and enable leadership within teams—not just at the center
These don’t happen by accident. They require deliberate design and consistent attention.
Closing Thought
Distributing Center of Excellence resources can improve delivery. I’ve seen it work—and I’ve been part of getting it wrong.
The difference isn’t in the org chart. It’s in whether the organization treats distribution as a coordination problem—or assumes that structure is the answer.
Most organizations recognize the symptoms when this breaks. Far fewer understand how to correct it without undoing the model entirely.
That’s the work most organizations underestimate.