Most mid-market GPs pay their third-party administrator on asserted complexity, not measured effort. The TPA knows what it does. The GP does not.
That information gap is an asymmetry tax. It is paid every quarter, on a fee that has not been renegotiated in years.
A fund-side data layer flips the asymmetry. The renegotiation pays for the build inside one cycle. The optionality is what compounds.
The problem, in the numbers that matter
TPA fees on a mid-market PE fund family run roughly 8 to 15 basis points of NAV blended. That fee is set at onboarding, reviewed lightly at fund launches, and almost never renegotiated on the basis of what the TPA actually does.
The GP has no independent visibility into time spent, error rates, manual touchpoints, or which functions are genuinely complex versus inherited from a legacy operating model. The TPA does. The asymmetry is the trade.
It is rarely framed as a transparency problem. It is framed as a fee problem. They are the same problem.
Indicative economics
| Line item | £2bn fund family | £5bn fund family |
|---|---|---|
| Current state | ||
| Blended TPA fee at 11 bps of NAV | £2.2M p.a. | £5.5M p.a. |
| GP-side ops oversight (1.5 / 3.0 FTE loaded) | £0.3M p.a. | £0.6M p.a. |
| Current run-rate | £2.5M p.a. | £6.1M p.a. |
| Post-build state, year 2 | ||
| Renegotiated TPA fee (20 to 35 percent reduction) | £1.5M p.a. | £3.8M p.a. |
| Selective insource of NAV review, LP reporting | £0.2M p.a. | £0.4M p.a. |
| Data platform plus instrumentation run cost | £0.4M p.a. | £0.6M p.a. |
| New run-rate | £2.1M p.a. | £4.8M p.a. |
| Recovery | ||
| Annual saving | £0.4M | £1.3M |
| Build cost (one-off, 9 to 12 months) | £0.6M | £1.1M |
| Payback | 18 months | 10 months |
Ranges are illustrative for a fund family with standard PE structures across two to four jurisdictions. Specific numbers move with mandate count, reporting cadence, and how much of the current fee is service versus jurisdictional overhead. A diagnostic call sharpens these to the actual asset.
The optionality, which the saving understates
Switching cost collapses. Today, changing TPA is a nine-month migration that nobody initiates. With the GP holding the golden source of truth, the historical record stays put. A switch becomes a contract decision, not an operational one. That itself is worth basis points at every renewal.
Selective insource becomes credible. Functions where the data shows the TPA adds limited value, LP reporting workflow, capital call drafting, fee calculation review, move back in over 18 to 24 months on the GP's own evidence rather than a consultant's assertion.
LP reporting differentiation. Owning the data layer means custom LP cuts and ESG reporting stop being TPA change requests. The GP runs them on its own timeline. This matters more in 2026 than it did in 2022 as LP data demands sharpen under AIFMD II and SFDR.
Carry-aligned operating leverage. The saving lands in management fee margin, but the deeper effect is that the GP can move on operating questions at its own speed rather than the TPA's. That speed compounds across a fund cycle.
Why this gap exists, and why it will not close on its own
The reason the gap exists is structural, not technical.
The COO of a mid-market GP is running fund admin, technology, compliance, regulatory reporting, LP servicing, and increasingly cyber and data infrastructure. The role expanded faster than the headcount. Building a fund-side data layer is a 12-month programme. The COO does not have a 12-month window. They have today's reporting, this week's audit, this month's regulatory deadline, and this quarter's LP requests.
The TPA, meanwhile, is rationally not incentivised to surface its own time-and-motion picture. Their commercial relationship with the GP depends on the asymmetry. They are not behaving badly. They are behaving like any vendor whose price is calibrated against information their counterparty does not have.
External advisors who could close the gap are typically paid as a percentage of savings, which creates a different problem. The advisor's incentive aligns with finding savings, not with respecting the operating reality of the firm. Vendors get strong-armed in ways that strain relationships the firm needs to maintain. The savings land but the cost shows up in service quality six months later.
What is missing is not analysis. It is a senior advisor who can do the work alongside the COO at a steady senior pace, with the operating credibility to renegotiate without breaking the relationship, and without an incentive structure that pulls toward over-cutting.
That role does not exist in most mid-market firms. The result is a fee schedule that drifts year by year and gets re-examined only when something breaks.
How the build sequences
Months 0 to 3. Stand up the data platform on the GP's cloud and mirror TPA feeds. Instrumentation captures what the TPA actually does, by function, by mandate, by hour.
Months 3 to 6. Six months of operating data produces a time-and-motion picture by function. The renegotiation conversation moves onto the GP's own evidence rather than the TPA's assertions. The fee comes down, typically 20 to 35 percent.
Months 6 to 12. Selective insource of two or three functions where the data shows clear economics. A new fee floor sets the next switching cycle. The optionality is locked in.
The uncomfortable conclusion
The mid-market GPs that will quietly outperform on management fee margin over the next two fund cycles will not be those that found the next great deal or hired the strongest deal team. The deal layer is competitive but not differentiating at the level of fund-level economics.
They will be the ones who decided that the answer to "where is the golden source of truth" cannot remain "inside the TPA." That decision sounds technical. It is not. It is the decision to stop paying the asymmetry tax.
If you are a mid-market COO or CFO, three questions are worth sitting with before the next operating committee.
What did the firm pay in TPA fees this year, all in, including pass-throughs and the GP-side oversight headcount? If the all-in number is not on the tip of the tongue, the gap is already open.
When was the fee schedule last meaningfully renegotiated, and on what basis? If the answer is "at the last fund launch" or "we have never renegotiated," the asymmetry is the price.
Where is the gold copy of the data today? If it lives inside the TPA, the leverage is theirs. Every other answer flows from that.
That is the conversation worth having before the next fee renewal cycle starts pricing this year's run rate.
Arcvale advises global private markets GPs across two service lines: AI as exit infrastructure across the private equity portfolio, and operating model design inside the GP itself. Senior Advisor, NED, and Operating Partner engagements across the investment cycle.
Open a conversationSara Gilbert is the founder of Arcvale Partners. She advises global private markets GPs across two service lines: AI as exit infrastructure across the private equity portfolio, and operating model design inside the GP itself. She operated inside AltaReturn through its $500M acquisition by Vistra and has held senior commercial roles at Northern Trust, FINBOURNE Technology, FIS APAC, and SS&C Technologies. Cross-border working rights across the UK, the EU, and Hong Kong.