boltKey takeaways
- No single metric describes a private fund; read a time-weighted measure (IRR) next to multiples (TVPI, DPI, RVPI) and a market comparison (PME).
- DPI is the only metric made entirely of cash returned; everything else depends partly on unrealized valuations.
- Always state whether a figure is gross or net of fees, expenses and carried interest, and never compare one fund's gross number with another's net.
- Benchmark against funds of the same strategy, geography and vintage year, because the age of a fund drives its metrics more than almost anything else.
- Subscription credit lines, recallable distributions and valuation policy can move the reported numbers without changing the underlying economics, so disclose them.
Venture and private equity funds are measured with a small set of metrics: IRR (a time-weighted annualized return), multiples such as TVPI, DPI and RVPI (how much value has been created per dollar paid in), and public market equivalents (how the fund did against an index). Each answers a different question, and each can mislead on its own. A credible performance report presents them together, net of fees and carry, with the fund's age and valuation basis made explicit.
Why private fund performance is hard to measure#
A mutual fund has a daily price, so its return is straightforward to compute. A closed-end private fund has none of that. Capital comes in through irregular capital calls, goes out through irregular distributions, and most of the value for most of the fund's life sits in illiquid holdings marked by the manager under a valuation policy.
That creates three problems that every performance metric tries to solve in a different way:
- Timing. A dollar returned in year three is worth more than the same dollar in year nine.
- Realization. Value that has been distributed is certain; value that still sits in the portfolio is an estimate.
- Opportunity cost. A 15% return is impressive in a flat market and mediocre in a strong one.
The core metrics at a glance#
| Metric | Formula | Question it answers | Main weakness |
|---|---|---|---|
| IRR | Discount rate that sets NPV of all cash flows (plus ending NAV) to zero | What annualized return have LPs earned? | Highly sensitive to timing; can be flattered by credit lines and early small wins |
| TVPI | (Cumulative distributions + NAV) / Paid-in capital | How much total value per dollar contributed? | Ignores time; depends on unrealized marks |
| DPI | Cumulative distributions / Paid-in capital | How much cash has come back? | Says little about a young fund |
| RVPI | NAV / Paid-in capital | How much value is still at risk in the portfolio? | Only as good as the valuations |
| MOIC | Total value of investments / Invested capital | How well did the deals do? | Usually gross; excludes fees and carry |
| PME | Fund cash flows compared with an index | Did the fund beat the public market? | Choice of index changes the answer |
Note that TVPI = DPI + RVPI by construction. Detailed walkthroughs live in the dedicated articles on TVPI, DPI, RVPI and PME.
IRR: the time-weighted view#
The internal rate of return is the discount rate that makes the present value of all contributions equal to the present value of all distributions plus the current NAV (treated as a hypothetical final distribution). Because it accounts for timing, IRR rewards managers who return capital quickly.
That same sensitivity makes IRR easy to distort:
- Subscription lines of credit. If a fund borrows to make investments and calls LP capital months later, the LP's money is outstanding for less time, which raises IRR without changing the multiple. Many LPs now ask for IRR both with and without the effect of the facility.
- Early quick wins. A small investment that doubles in a year can post a very high IRR that persists in the since-inception number even after the rest of the portfolio matures at a lower rate.
- Interim IRR on young funds. For the first few years, IRR is mostly a function of fees and valuation policy. It is not a meaningful predictor of final outcome.
Multiples: the value-creation view#
Multiples ignore time and focus on magnitude. A fund at 2.5x TVPI has created two and a half dollars of value per dollar paid in, whether that took six years or twelve. LPs care about both dimensions, which is why IRR and TVPI are almost always shown side by side.
Within the multiples, the split between DPI and RVPI tells you how much of the result is banked. Two funds at 2.0x TVPI are not equivalent if one is at 1.6x DPI and the other at 0.2x DPI.
Gross vs. net: the gap that matters most#
Gross performance measures the investments themselves, before management fees, fund expenses and carried interest. Net performance measures what LPs actually receive. Both are useful: gross shows deal selection skill, net shows what an investor earned. They should never be mixed in a comparison.
Worked example (hypothetical)#
Assume a $100M venture fund in year six. LPs have paid in $85M. Of that, $70M went into portfolio companies and $15M covered management fees, organizational costs and fund expenses. The portfolio is worth $205M in total (exits already realized plus current fair value of remaining holdings). Carried interest of 20% on LP profits has been paid or accrued.
| Item | Amount |
|---|---|
| Invested capital | $70M |
| Gross value of investments | $205M |
| Gross MOIC | 2.93x |
| LP paid-in capital | $85M |
LP profit before carry (205 − 85) | $120M |
| Carried interest at 20% | $24M |
Net value to LPs (205 − 24) | $181M |
| Of which distributed to LPs | $45M |
| Of which remaining NAV to LPs | $136M |
| Net TVPI | 2.13x |
| Net DPI | 0.53x |
| Net RVPI | 1.60x |
The same fund reports 2.93x or 2.13x depending on the basis. Neither is wrong; presenting one without labeling it is. For a deeper look at how carry is computed, see carried interest.
Benchmarking: vintage, strategy and quartiles#
Fund metrics change predictably with age. A three-year-old venture fund will almost always show a low DPI and a TVPI near or slightly above 1.0x, while a ten-year-old fund should be mostly realized. Comparing funds of different ages tells you very little.
The standard approach is to compare a fund with peers of the same vintage year (usually the year of first capital call or first investment), same strategy (seed venture vs. growth equity vs. buyout) and, where possible, same geography. Peer data from commercial benchmark providers is typically presented in quartiles. When citing a quartile ranking, say which data set, which as-of date and which metric (a fund can be top quartile on TVPI and second quartile on IRR).
PME adds another dimension: rather than comparing with other private funds, it compares the fund with what the same cash flows would have earned in a public index.
Common pitfalls in performance reporting#
- Mixing as-of dates. Distributions through the current quarter combined with NAV from the prior quarter overstate or understate value.
- Inconsistent treatment of recallable distributions. If recycled capital is netted against paid-in in one period and grossed up in another, DPI and TVPI drift for no economic reason.
- Ignoring GP commitment. Decide whether metrics are computed on all partners or on fee-paying LPs only, and say which.
- Stale marks. Holding companies at cost for years after a down round inflates RVPI. A disciplined portfolio valuation process protects the credibility of every other number.
- Cherry-picked periods. Since-inception figures are standard for closed-end funds. Showing a flattering sub-period without the full history invites skepticism.
How to present performance to LPs#
A clear performance page for a closed-end fund usually includes:
- Commitments, paid-in capital, cumulative distributions and NAV, all at one as-of date.
- Net IRR, net TVPI, DPI and RVPI, with gross IRR and gross MOIC alongside for context.
- IRR with and without the effect of any subscription facility, if one is used.
- A PME or index comparison with the index named.
- A short note on valuation methodology and any significant changes in marks during the period.
That layout lets an LP see timing, magnitude, realization and market context on one page, which is what a performance discussion actually needs.
How Privatis helps#
Privatis computes IRR, TVPI, DPI and RVPI directly from the fund's double-entry general ledger, capital call and distribution records and quarterly NAV, so every metric ties to the same as-of date. Investor reporting uses ILPA-style templates, LPs see their own capital account statements and performance in the LP portal, and the audit trail records every change to the underlying valuations and cash flows.