A close is a network of dependencies, not a checklist: compute each path's finish and slack before setting a target or fixing a delay.
A private fund's quarter-end close takes as long as its longest chain of dependent tasks, its critical path, and no target date shortens it until that chain is shortened. In the illustrative Northbridge close, thirteen tasks give a signing on business day 18; removing three delays together brings it to day 15, yet each fix on its own saves at most one day, because two paths tie on day 13.
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 →
Most firms set a close target, day 10 or day 15, and then measure how often they miss it. The question that matters is different: which task finishes last, what it waits for, and how much slack every other path has. A close is a network, not a checklist. Treating it as a checklist is how a team can be "substantially complete" on business day 12 and still sign on day 18, which is where the fictional Northbridge Private Capital starts.
Four workstreams, thirteen tasks, durations in business days after quarter-end. The network and durations are illustrative; the three delays are those the case describes.
| Task | Waits for | Days | Finishes |
|---|---|---|---|
| T1 Bank reconciliations | Quarter-end | 3 | 3 |
| T2 Capital activity and facility | T1 | 2 | 5 |
| T3 Administrator draft pack received | Day 6 | 0 | 6 |
| T4 Administrator questions answered | T3, plus 2 days without an owner | 1 | 9 |
| T5 Fees, expenses and allocations reviewed | T4, T2 | 2 | 11 |
| T6 Capital accounts rolled forward | T5 | 2 | 13 |
| T7 Portfolio data received | Due day 5, 3 days late | 0 | 8 |
| T8 Valuation memos and challenge | T7 | 3 | 11 |
| T9 Committee approval and posting | T8 | 2 | 13 |
| T10 Carried interest and performance | T9, T6 | 1 | 14 |
| T11 Statements and investor reports | T10 | 1 | 15 |
| T12 Final review, 14 items that belonged earlier | T11 | 3 | 18 |
| T13 Close certificate signed | T12 | 0 | 18 |
The final review takes 1 day with nothing in it that belongs elsewhere and clears 7 items a day, so the 14 misplaced items add 2 days. At the gate that owns them, where the owner and the evidence already sit, the same items clear at 14 a day: they add 1 day to the fees and allocations review instead.
Finish of a task = MAX(finish of every task it waits for, its earliest start) + its duration
Close day = finish of the last task; slack of a path = finish of the binding path − its own finish
In Excel, one row per task: =MAX(G$5,G9,G10)+D11, where G5 holds the earliest start and G9, G10 the finishes of its predecessors. Copy the column once per scenario, change only the delay inputs, and read the last row.
The books path (T3 to T6) finishes on day 13: draft pack on day 6, questions answered on day 9 after waiting two days for an owner, fees reviewed by day 11, capital accounts by day 13. The valuations path (T7 to T9) also finishes on day 13: data on day 8 instead of 5, memos by day 11, committee by day 13. Carried interest needs both, so it runs on day 14, the statements on day 15, and a three-day final review signs the close on day 18.
| Scenario | Books ready | Valuations ready | Signed | Days saved |
|---|---|---|---|---|
| As run | 13 | 13 | 18 | 0 |
| Valuation data on time | 13 | 10 | 18 | 0 |
| Questions owned on arrival | 11 | 13 | 18 | 0 |
| Items moved to their gates | 14 | 13 | 17 | 1 |
| Data and ownership | 11 | 10 | 16 | 2 |
| All three | 12 | 10 | 15 | 3 |
Getting the portfolio data on time saves three days on the valuations path and nothing on the close, because the books still finish on day 13. Giving the administrator's questions an owner on arrival saves two days on the books path and nothing on the close, because valuations still finish on day 13. Moving the fourteen items to their gates saves one day: the final review shrinks by two, the fees review grows by one. Separately the three fixes save 1 day. Together they save 3, and the close signs on day 15.
This is the trap of a close with two critical paths. Each team that fixes its own delay sees no improvement and concludes the fix did not work. Only a plan that changes both paths in the same quarter shows the gain.
After all three fixes the books path binds on day 12 and valuations have 2 days of slack. The next day saved has to come from the books: an earlier draft pack, a shorter fees review, a capital account roll-forward that starts before the review ends. Spending effort on valuations from here saves nothing, however visible the work. That is the general rule: after every fix, recompute which path binds, because it moves.
A close is as long as its critical path: 18 business days here as run, 15 with three fixes made together, and none of them worth more than a day alone. Build the network, compute each path's finish and slack, and plan changes as a set. The people side of the same peak is worked in sizing a fund operations team for the peak month, and the free workbooks for this case include the close network with its six scenarios and checks.
There is no universal figure; the floor is the network's critical path. In the illustrative close the shortest path after fixes ends on business day 15, against 18 as run. A target set below the critical path, say day 10, cannot be met by effort alone; the dependencies on portfolio data and the administrator's draft pack have to move first.
Because another path finished on the same day. In the illustrative close, portfolio data arriving on time brings valuations from day 13 to day 10, but the books still finish on day 13, so the close still signs on day 18. Fixing the books as well brings it to day 16, and moving the review items to their gates to day 15.
One row per task with its duration and predecessors, and a finish column equal to the latest predecessor finish, or the task's earliest start, plus its duration. The last task's finish is the close day. In the illustrative case that is day 18, with 0 days of slack on either middle path as run and 2 days on valuations once all three fixes are made.
The close described here is the one chapter 6 of The Private Markets COO describes at Northbridge; the companion files take it to a number. 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.