Global Capability Centres (GCCs)

By replacing assertion with evidence. That means a quality standard defined precisely enough to score, applied consistently across teams and internal clients, and documented so coverage is demonstrable on request — what was inspected, how it was scored, what the trend is. When the parent can reconcile the centre’s numbers with what its own business units experience, reviews stop being audits. The deeper shift is positional: a centre that can prove its standard tomorrow morning holds an asset most internal alternatives cannot match, and the burden of proof it carries becomes the strongest argument for giving it more.

Rationalise them into one governed framework rather than renegotiating them one by one. The work is to define shared measures — turnaround, quality, escalation — once, with service tiers where clients genuinely differ; migrate existing agreements onto the framework; and build measurement that reconciles across the whole book so the centre finally sees its performance the way leadership needs to. Add leading indicators such as queue depth and first-pass yield so risk surfaces before a breach rather than after. The sprawl took years to accrete, but the framework does not: definitions and reconciled measurement come first, and the difficult renegotiations follow with evidence in hand.

Rung by rung, with evidence at every step. A centre justified on rates is permanently renegotiable, because a cheaper alternative can always be modelled. The escape is capability: first prove quality at volume with a scored, auditable standard; that earns end-to-end process ownership; measurable improvement of owned processes earns transformation work; and a centre contributing improvements the parent adopts globally has left the rates conversation for good. None of the rungs can be skipped and none can be claimed — each must be demonstrated operationally. The maturity story is built from proof, which is exactly why it holds when procurement comes asking.

This is the sector nearest my own history. At Publicis Groupe I led Business Excellence across a global delivery operation — 500+ clients, more than 2,000 teams, USD 750M+ in media spend, for brands including Disney, Samsung, Adobe and P&G — the same structural position a GCC holds: critical work delivered at distance, proven continuously to stakeholders with alternatives. The results I bring as method, not anecdote: quality from 95% to 99% across more than 2,000 campaigns and 450 clients, and staged QA that protected more than USD 20 million in billings at a global advertising network. XLRI-trained, Lean Six Sigma Green Belt, 19 years in.

Because distance amplifies doubt. An error onshore is treated as an incident; the same error from the centre becomes evidence in a standing argument about whether the work should have moved at all. It is asymmetric and it is not going to change — so the answer is to build for it. Prevention first: quality designed into the flow so fewer defects reach the parent. Then proof: a scored standard and reconciled reporting that place the error in honest context — one defect against a measured, visible record. A mistake inside a provable system is an exception; the same mistake inside asserted quality is a pattern.

By giving the next layer real machinery to run, not courses to attend. Move the operating standard out of your two or three exceptional managers and into documented processes, explicit quality definitions and a governance cadence — then hand pieces of that system to emerging leaders as genuine ownership: they chair the reviews, hold the numbers, defend the standard. Capability compounds fastest under real accountability on a well-designed system, because the system catches what inexperience misses while judgement develops. The centre gains a bench it can show the parent, and the standard stops being one resignation away from crisis.

Established and scaling centres. The setup phase — location strategy, entity structure, the feasibility case — is specialist work I do not do, and I would rather say so than stretch. My work begins once a centre exists, runs meaningful volume and answers to a parent: that is when quality must become provable, SLAs need governing, the maturity case needs building and the leadership bench needs developing. A young centre approaching that threshold benefits from installing the standard early, before bad habits accrete — but the centre, and the operating reality, have to exist first.

Usually three things the internal team cannot easily create for itself. A standard with independent authority — an excellence function that reports into delivery ends up marking its own homework, and HQ knows it. A view across the whole book — internal teams typically inspect within silos, so systemic defects recurring across many internal clients read as unrelated incidents. And governance that turns findings into design changes rather than reports. The aim is not to replace your team but to hand them a sharper system: I install the standard, the reconciled measurement and the cadence; they run it — with more credibility in front of the parent than they had before.