A fourteen-field administrator scorecard worked month by month, with the critical gate that no average or weighting can replace.
Score the administrator monthly on each field from 1 to 5, average only the non-critical fields, and let any critical field at 2 or below escalate the month whatever the average says. On an illustrative fourteen-field scorecard, June averages 4.55 on its eleven non-critical fields, the best month of the year, and still escalates because payment processing scored 2. An average of all fourteen fields would read 4.36 and hide the failure entirely.
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 →
The scorecard below is the one the companion workbook builds for Northbridge, the fictional firm The Private Markets COO follows: fourteen fields of administrator performance, each scored monthly on a five-point scale. 5 consistently exceeds the agreed standard; 4 meets it with no material issue; 3 is minor misses corrected promptly; 2 is repeated or material misses requiring remediation; 1 is unacceptable performance or unmanaged risk. The scores themselves are illustrative inputs.
Three fields are critical because a single failure in them can move money or misstate a NAV: NAV adjustments after initial delivery, payment or cash-processing errors, and open incidents. The book's rule is that a high overall score cannot offset a critical failure. The arithmetic below shows why that rule has to be built into the formula, not left to whoever reads the dashboard.
| Field | Critical | June |
|---|---|---|
| Close deliverables on time | No | 4 |
| First-pass accuracy | No | 5 |
| Aged reconciliation items | No | 4 |
| NAV adjustments after initial delivery | Yes | 5 |
| Capital notices delivered on time | No | 5 |
| Payment or cash-processing errors | Yes | 2 |
| Investor inquiries within service target | No | 5 |
| Data-feed completeness and timeliness | No | 5 |
| Audit and tax support responsiveness | No | 4 |
| Open incidents and days outstanding | Yes | 4 |
| Recurring root causes | No | 4 |
| Staffing changes and coverage gaps | No | 4 |
| Agreed improvement actions completed | No | 5 |
| Manager dependencies overdue | No | 5 |
Average = sum of the non-critical scores / number of non-critical fields = 50 / 11 = 4.55
Gate = lowest critical score = 2
Status: ESCALATE if the gate is 2 or below; REMEDIATE if any other field is 2 or below or the average is under 3.5; otherwise ON STANDARD.
Excel, with the critical flag in B and June in H: =AVERAGEIF(B2:B15,"N",H2:H15) for the average, =MINIFS(H2:H15,B2:B15,"Y") for the gate, and =IF(gate<=2,"ESCALATE",IF(OR(MINIFS(H2:H15,B2:B15,"N")<=2,avg<3.5),"REMEDIATE","ON STANDARD"))
The two steps answer different questions. The average says how the administrator is performing on the routine work: in June, very well. The gate says whether something happened that the firm cannot tolerate: in June, a payment went wrong. The 3.5 floor and the gate at 2 are the thresholds used here (the companion workbook's status line uses a floor of 3, which gives the same eight statuses); set your own in the oversight policy, but set them before the scores come in.
| Month | Average, 11 fields | Lowest critical | Average, all 14 | Status |
|---|---|---|---|---|
| Jan | 4.00 | 4 | 4.07 | On standard |
| Feb | 3.82 | 4 | 3.86 | On standard |
| Mar | 3.09 | 3 | 3.21 | Remediate |
| Apr | 3.91 | 4 | 3.93 | On standard |
| May | 4.09 | 4 | 4.14 | On standard |
| Jun | 4.55 | 2 | 4.36 | Escalate |
| Jul | 4.09 | 3 | 4.00 | On standard |
| Aug | 4.00 | 3 | 3.93 | On standard |
Six months on standard, one to remediate and one to escalate. March is the month a pure average catches: 3.09, with staffing coverage at 2. June is the month it cannot. Its average is the year's highest, and the month contains the year's most serious event. The two months that follow carry a 3 on a critical field, payment processing in July and open incidents in August: not failures, but the lowest critical score has not returned to 4 since May, and the trend belongs in the quarterly review even though no single month escalates.
| Design | June score | Escalates? |
|---|---|---|
| Eleven non-critical fields, plus the critical gate | 4.55 | Yes |
| All fourteen fields, equal weight | 4.36 | No |
| All fourteen, critical fields weighted three times | 4.15 | No |
| All fourteen, payment processing at 1 | 4.29 | No |
| All fourteen, all three critical fields at 1 | 3.79 | No |
No average can do the gate's job. Even with all three critical fields at the worst score on the scale, June's fourteen-field average is 3.79, above the 3.5 floor. Weighting does not help either: tripling the critical fields brings June down to 4.15, a figure that still looks like a good month. One payment error in a month of excellent close work is a dilution problem, and averaging is a machine for dilution.
A scorecard prices quality; the invoice is the other half of the administrator decision. The two are put together in how much a cheaper fund administrator actually saves, and the fee itself is rebuilt in how fund administration fees are calculated.
The scorecard, with the critical flags, the status formula and the capacity model beside it, is in the free workbooks for this book.
Fields that cover delivery, accuracy and risk: close deliverables on time, first-pass accuracy, aged reconciliations, NAV adjustments after delivery, capital notices, payment errors, investor inquiries, data feeds, audit support, incidents, root causes, staffing, improvement actions and manager dependencies. The worked scorecard uses fourteen, three of them critical, each scored 1 to 5 every month.
Because an average dilutes a single failure. A month with excellent close work and one payment error averages 4.36 across fourteen fields. Even with all three critical fields at 1 the average is 3.79, above a 3.5 floor. Keeping them out and gating on them separately makes a critical score of 2 escalate however good the rest of the month was.
Monthly, with a quarterly review of the trend. Monthly scores catch events like June's payment error; the quarterly view catches drift, such as a lowest critical score that has stayed at 3 for two months. Averaging the year, 3.94 in the worked case, hides both.
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 1340 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.