A proposal 80,000 lower on the invoice, priced as a service model rather than a fee: the rework, the reissued notices, the data work and the move itself.
A challenger administrator quotes 540,000 a year against the incumbent’s 620,000. Once the internal hours it creates are costed at 95 an hour, the saving is not 80,000 but 17,488, and the move, at 188,000 all in, takes 10.7 years to pay back. To recover it in three years the cheaper provider would have to get more deliverables right first time than the one it replaces.
Worked in full in The Private Markets COO by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
Chapter 8 of The Private Markets COO says it in a sentence: administrator cost includes internal review, rework, technology, data transformation, audit support, transition risk and the capacity consumed by exceptions, and a lower invoice can create a higher operating cost. It tells the manager to measure right-first-time delivery and to price the service model, not only the fee. It never does the arithmetic. The second of the three cases in the companion files does, for Northbridge Private Capital, the fictional firm the book follows.
The deliverable counts come from workbook 4, the capacity model: 56 NAV packs a year (fourteen vehicles closed four times), 26 capital-event packs (16 calls and 10 distributions), 56 investor statement batches and twelve monthly reporting cycles. Everything else is an input the book does not print, marked ADDED in the workbook, and every one of them can be overwritten.
| A, incumbent | B, challenger | B less A | |
|---|---|---|---|
| Review hours | 612.0 | 612.0 | 0.0 |
| Rework hours | 169.9 | 424.8 | 254.9 |
| Reissues to investors | 1.64 | 4.10 | 2.46 |
| Reissue hours | 65.6 | 164.0 | 98.4 |
| Data transformation and audit support hours | 0.0 | 240.0 | 240.0 |
| Internal hours | 847.5 | 1,440.8 | 593.3 |
| Internal cost at 95 an hour | 80,514 | 136,876 | 56,362 |
| External reissue cost | 4,100 | 10,250 | 6,150 |
| Fee | 620,000 | 540,000 | −80,000 |
| Operating cost of the service | 704,614 | 687,126 | −17,488 |
Rework is the largest line. If every deliverable in the year arrived wrong, the team would spend 2,124 hours correcting them: 56 NAV packs at 24 hours, 26 capital-event packs at 10, 56 statement batches at 8 and 12 cycles at 6. The incumbent gets 8 per cent wrong, which is 169.9 of those hours; the challenger gets 20 per cent wrong, which is 424.8. The reissues follow the same logic on the 82 investor-facing deliverables, the notices and the statements: one in four of the failures escapes, so the incumbent reissues 1.64 a year and the challenger 4.10.
The fee falls by 80,000 and the internal cost rises by 56,362. Together with 6,150 of extra reissue cost, the service is 17,488 cheaper to run, 21.9 per cent of the saving on the invoice. The other 78.1 per cent is paid in hours by people who are not on the procurement committee.
The 80 per cent is an estimate, and the proposal stands or falls on it. The workbook turns it into a price: each point of right-first-time delivery is worth 3,309.30 a year here. Of that, 2,017.80 is rework (1 per cent of the 2,124 hours, at 95) and 1,291.50 is reissues (1 per cent of 82 deliverables, one in four escaping, at 40 hours and 2,500 each, which is 6,300 a reissue).
Part of B’s cost does not depend on its quality at all. The data transformation and the audit support are 240 hours whatever its error rate, 22,800 a year. So of the 80,000 on the invoice, 57,200 is available to pay for quality, and at 3,309.30 a point it buys 17.3 points. Below that, B is dearer than A.
| B right first time | B operating cost | Saving against A | Years to repay 188,000 | Three-year position |
|---|---|---|---|---|
| 74.7% | 704,614 | 0 | never | −188,000 |
| 80.0% | 687,126 | 17,488 | 10.7 | −135,535 |
| 85.0% | 670,580 | 34,035 | 5.5 | −85,895 |
| 90.0% | 654,033 | 50,581 | 3.7 | −36,256 |
| 92.0% | 647,414 | 57,200 | 3.3 | −16,400 |
| 93.7% | 641,948 | 62,667 | 3.0 | 0 |
Changing administrator is not free. The case puts 150,000 of external cost on it and 400 internal hours, 38,000 at 95, so 188,000 all in. At the proposal’s 80 per cent the annual saving of 17,488 repays that in 10.7 years. Few administration contracts, and fewer funds in their investment period, last that long.
The non-obvious row is the fifth. At 92 per cent, the incumbent’s own rate, B is 57,200 a year cheaper and still takes 3.3 years to repay the move. The three-year payback a committee would normally ask for needs 93.7 per cent: a provider that is both cheaper and better than the one it replaces. That is not the proposal on the table, and a proposal that assumes it has assumed the conclusion.
That is why the right-first-time rate belongs in the contract as a measured service level rather than in the pitch as a promise. On these inputs every point below the assumed rate costs 3,309.30 a year, and the difference between the rate a challenger assumes and the rate it delivers is where the saving disappears.
Money is one currency; hours are the other, and the team does not have spare ones. B adds 593.3 internal hours a year, 7.8 per cent of the team’s practical capacity of 7,614 hours in workbook 4. Workbook 4 already has April, the month after the year-end close, at 118 per cent of monthly capacity on the current run rate: 749.6 hours of demand against 634.5.
| April demand, hours | Monthly capacity, hours | Utilisation | |
|---|---|---|---|
| Current run rate, incumbent (workbook 4) | 749.6 | 634.5 | 118.1% |
| B’s extra hours on the quarter-end profile (15% in April) | 838.6 | 634.5 | 132.2% |
| B’s extra hours spread evenly (one twelfth in April) | 799.1 | 634.5 | 125.9% |
The case treats the extra hours as quarter-end work, because rework and reissues arrive with the NAV packs and the statements: 89.0 of them land in April and take it to 132 per cent. Spreading every hour evenly across the year still leaves April at 125.9 per cent. Either way the switch deepens the one month the team already cannot staff. The hours come out of the same people who are meant to be reviewing the administrator’s work, so a provider with a lower right-first-time rate is reviewed by a team with less time to review it.
Open Three_Cases.zip and the file Case_2_The_Cheaper_Administrator.xlsx. On sheet “1. Cost of service” the inputs are amber: deliverable counts and hours in C5:E8, the internal rate in C10, the reissue assumptions in C11:C13, the two proposals in C15:C21, the transition in C23:C25, and the three figures carried over from workbook 4 in C27:C29. The operating costs are C41 and D41, the gap C44, the value of a point C45, the break-even rate C46, the transition cost C47, the payback C48, the three-year rate C49 and the April utilisation C51 and C52. Sheet “2. Checks” runs six controls, among them that B costs exactly what A costs at the break-even rate, and ends ALL OK.
The carried-over figures come from workbook 4: practical capacity in Team!C14, April demand in Activities!O20 on the peak basis, April at 118 per cent in Capacity Results!I5, and the quarter-end profile in Profiles row 7. Replace the ADDED inputs with your own hours and your own rate and the break-even and payback rates recompute; the conclusion to test is whether your challenger’s evidence clears them.
This article is one calculation from The Private Markets COO. The book takes the same case from first principles to the decision, chapter by chapter, and every figure it prints is a live formula in the free companion workbooks.
Get the book on Amazon →Free companion files
Also on Amazon UK · Amazon Germany · Amazon France · Amazon Canada
Reading guide: private equity and private markets → · All 1706 articles →
If this book helped — or didn’t — a few lines on Amazon are worth more than they look: they are what the next reader goes on. Write a review. The workbook stays free either way.