Fully realised, partially realised and unrealised: how to define the buckets, why each must be pooled, and what the split tells an investor that the blended figure hides.
Split the deals into three buckets, fully realised, partially realised and unrealised, and compute a pooled IRR and multiple on each bucket's own cash flows, never an average of deal IRRs. On an illustrative eight-deal record the whole track record shows a 10.6 per cent gross IRR and 1.66x; the five realised deals earned 14.3 per cent and 1.93x, and the three deals still held show 5.0 per cent and 1.33x on the firm's own marks.
Worked in full in Private Equity Investor Relations by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
Every limited partner reading a track record asks the same question: how much of this has been turned into cash, and how much rests on valuations the firm sets itself? The split answers it, but only if the buckets are defined honestly and each one is computed on pooled cash flows. This is a performance split by deal status; the accounting split of a single exit into realised gain and unrealised reversal is a different question, worked in realised vs unrealised gain on a fund exit.
A firm's previous fund made eight investments between 2016 and 2020. The valuation date is the end of 2025. Flows are at year end and gross of fees; every figure is illustrative and in $m.
| Deal | Invested | Cost | Proceeds | Value held | Gross IRR | MOIC | Bucket |
|---|---|---|---|---|---|---|---|
| A | 2016 | 40.0 | 100.0 in 2020 | 0.0 | 25.7% | 2.5x | Realised |
| B | 2016 | 35.0 | 70.0 in 2021 | 0.0 | 14.9% | 2.0x | Realised |
| C | 2017 | 50.0 | 30.0 in 2022 | 0.0 | −9.7% | 0.6x | Realised |
| D | 2017 | 45.0 | 120.0 in 2023 | 0.0 | 17.8% | 2.67x | Realised |
| E | 2018 | 30.0 | 66.0 in 2023 | 0.0 | 17.1% | 2.2x | Realised |
| F | 2018 | 55.0 | 40.0 in 2024 | 45.0 | 6.9% | 1.55x | Partial |
| G | 2019 | 60.0 | none | 75.0 | 3.8% | 1.25x | Unrealised |
| H | 2020 | 50.0 | none | 60.0 | 3.7% | 1.2x | Unrealised |
A deal is realised when nothing is left: no residual stake, no escrow of significance, no earn-out still to be paid. A deal that has returned some cash and still holds value is partially realised, and its remaining value is a mark like any other. Deal F returned $40.0m on a $55.0m cost and still carries $45.0m; it belongs with the unrealised deals for the purpose of the question the investor is asking.
Bucket IRR = the rate that sets Σ (proceeds − cost) by date, plus value held at the valuation date, to zero, across all deals in the bucket
Bucket MOIC = (total proceeds + value held) ÷ total cost
In Excel, sum each bucket's flows by year in one column and use =IRR(B2:B11), or =XIRR(values,dates) on actual dates.
| Bucket | Cost | Proceeds | Value held | Gross IRR | MOIC |
|---|---|---|---|---|---|
| Realised (A to E) | 200.0 | 386.0 | 0.0 | 14.3% | 1.93x |
| Partially realised (F) | 55.0 | 40.0 | 45.0 | 6.9% | 1.55x |
| Unrealised (G, H) | 110.0 | 0.0 | 135.0 | 3.8% | 1.23x |
| Partial and unrealised (F to H) | 165.0 | 40.0 | 180.0 | 5.0% | 1.33x |
| Whole record | 365.0 | 426.0 | 180.0 | 10.6% | 1.66x |
The realised deals earned 14.3 per cent and 1.93x. The deals still held, 45.2 per cent of the capital invested, show 5.0 per cent and 1.33x, and the whole record 10.6 per cent. Unrealised value is 29.7 per cent of the record's total value. The investor now knows two things the blended figure hid: the firm's exits have been good, and its more recent deals have not yet shown that they will be.
Note what pooling does. The simple average of the five realised deal IRRs is 13.1 per cent, not 14.3: the average gives Deal C's loss and Deal A's 25.7 per cent one vote each, whatever their size and however long the money was out. A pooled bucket weights each deal by its money and its time, which is what an investor in the fund actually experienced, and the gap between the two can run in either direction. The reason IRRs cannot be averaged is worked in why fund IRRs cannot be averaged.
| Change in marks | Value held | Whole record IRR | Whole record MOIC | Held deals IRR | Held deals MOIC |
|---|---|---|---|---|---|
| −30% | 126.0 | 8.9% | 1.51x | 0.1% | 1.01x |
| −20% | 144.0 | 9.5% | 1.56x | 1.9% | 1.12x |
| −10% | 162.0 | 10.0% | 1.61x | 3.5% | 1.22x |
| 0% | 180.0 | 10.6% | 1.66x | 5.0% | 1.33x |
| +10% | 198.0 | 11.1% | 1.71x | 6.4% | 1.44x |
The realised bucket does not move at all, which is the point of reporting it. Set the held value to zero and the whole record falls to 3.7 per cent and 1.17x: the floor a sceptical investor computes silently.
Report the split even when it is unflattering. A gap between realised and unrealised performance is normal for a younger portfolio. Hiding it is what turns a normal gap into a diligence finding.
Define realised as fully exited, put partial realisations with the held deals, pool each bucket's flows and report IRR and multiple for each next to the whole record. The same discipline, applied to a fund's net figures, is in six ways to state the same fund's return, and the measures can be rebuilt on your own fund in the free workbook for this case.
No. A deal is realised only when nothing of value remains. A partial realisation still carries a mark, so it belongs with the unrealised deals or in its own bucket. In the worked case Deal F returned $40.0m on $55.0m and still holds $45.0m; that $45.0m is a valuation, not cash.
Yes. A deal sold at a loss or written off is fully realised. Excluding it flatters the record: in the worked case the realised bucket shows 14.3 per cent and 1.93x with the loss on Deal C, and 19.0 per cent and 2.37x without it. Diligence teams rebuild the buckets from the cash flows and will find the omission.
Set the held value to zero and recompute on realised cash alone. In the worked case the whole record falls from 10.6 per cent and 1.66x to 3.7 per cent and 1.17x. Nobody expects that outcome, but it shows how much of the record depends on the firm's own marks.
This article is one calculation from Private Equity Investor Relations. 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 453 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.