Global Capability Center Performance Tuning
An independent, buyer-held read of an underperforming captive, run on the institution’s own evidence, ending in a board-grade verdict: fix, expand, or repatriate.
Buyer-held · a live captive · no center to sell
The center looks full. That is not the same as working
A captive global capability center is justified on labor arbitrage. The arbitrage erodes as wages inflate, and agentic AI substitutes for headcount in the work the center holds, so the value case weakens over the life of the center. The metrics a center reports measure activity, not value delivered. The offering is an independent, buyer-held read of the existing center, ending in a board-grade verdict: fix, expand, or repatriate.
- The arbitrage that justified the center is eroding. A captive is built on labor arbitrage. Offshore wages inflate over time, and agentic AI substitutes for the headcount the center was built to hold. The saving that justified the center narrows across its life. A case priced on yesterday’s arbitrage prices a position that has moved.
- Activity is not value. A center reports utilization, seats, tickets, and cost per person. Each measures activity; none measures value delivered. A center can look busy while its case quietly erodes. The board sees a full center and assumes a working one.
- The institution owns the failure. A captive is the institution’s own operation. There is no vendor to blame for a weak result. A gap in the parent’s own readiness to run and govern the work surfaces as the center’s problem. The accountability sits fully with the parent, and the failure has nowhere else to go.
- The categories that decide the outcome are institutional. Akholi’s five-category framework, drawn from the study of 150 outsourcing engagements, is institutional and structural, and follows the work into any delivery model, including a captive. In that study, more than half the institutions, 77 of 150, also ran their own global capability centers, and most were shifting more work in. The same institution that must prepare and govern outsourced work owns the readiness a captive demands.
- No advisor will tell the board to bring the work home. Nearly every firm that could assess the center also builds, operates, expands, or buys captives. The only instrument on offer, the maturity model, points one way: up. It has no rung for winding down or bringing work home. Repatriation is the outcome the market cannot write, so the board hears every option except the one that ends the center.
The read in brief
The Captive Read
The Captive Read, phase by phase
Frame — the executive-sponsor kickoff
The read opens with the executive sponsor. Akholi and the sponsor fix the board question the read must answer. The sponsor sets the institution’s own adequacy bar, the standard the center must clear to justify its cost and location. The conditions for repatriation are named up front. Repatriate is a live outcome from the start.
Instrument — build the buyer-held read
Akholi stands up a buyer-held read of the center. The read measures true output and cost-per-unit, above utilization, seats, and tickets. Every finding rests on the institution’s own evidence, not the center’s activity dashboard. The center is engaged and interviewed throughout. Every later phase runs on this read.
Diagnose — the five categories on the institution’s evidence
Akholi scores the captive across all five failure categories on the institution’s own evidence. Client and retained-organization preparation: the institution’s own readiness to run and govern the work. The center’s own delivery: the vendor role, now owned in-house. Co-management: the seam between the parent and its center. AI technology: the agentic content of the work and its controls. Third-party and fourth-party dependencies: what the center itself relies on. The read fuses two questions: whether the value is real against the cost, and whether the institution could survive the center failing.
Reprice — whole-life cost against the counterfactual
Akholi establishes what the center’s work should cost delivered agentically, and reads the whole-life cost against the counterfactual of not owning the center. Stranded capacity in a fixed headcount base is exposed. The reprice is the institution’s own, not the center’s.
Decide — the verdict against the bar
Akholi returns the verdict against the adequacy bar. It resolves to one of three doors. Fix: recover on measured terms, re-scope the mandate, rebuild governance and decision rights. Expand: only where the value case clears the bar. Repatriate: a governed decision to bring the work home or move it out, independent of any party bidding to receive it. The sponsor holds the verdict and the decision.
The Captive Read fits an institution where
- A captive global capability center is live and underperforming against its original case.
- The board carries accountability for a center it owns and depends on.
- The value case rests on a labor arbitrage the parent has not independently re-priced.
- The center’s own metrics report activity, and the board wants a read on value.
- Repatriation is worth putting on the table, even where the institution means to keep the center.
The Practice
All offerings →GCC Planning
An independent, buyer-held test of the forward build decision, before the capital is committed.
Outsourcing Performance Tuning
A buyer-held read of a live, underperforming outsourcing engagement across all five categories.
The Practice
Independent reads and board-grade verdicts across the outsourcing and GCC lifecycle.