Data sourced from Dakota Private Markets, the private fund performance platform powered by Dakota. Learn More | Request Access
Private fund performance data often arrives in different formats, from quarterly reports and capital account statements to audited financials and manager presentations. Entering the numbers may appear straightforward, but small inconsistencies can materially change how a fund looks against its peers.
A misplaced decimal can turn a 12.0% return into 1,200%. Mixing gross and net returns can make one manager appear stronger than another. Using the wrong reporting date can create a comparison between funds at different points in their development.
For investors, consultants, and manager research teams, accurate benchmarking starts with disciplined data entry. These are some of the most common mistakes to avoid.
1. Mixing Gross and Net Performance
Gross and net returns measure different things. Gross performance typically reflects investment results before management fees, carried interest, and certain fund expenses. Net performance reflects the return received by limited partners after those costs.
The two should never be combined in the same comparison. A gross IRR entered into a dataset of net returns will overstate relative performance and may move a fund into the wrong quartile. Every performance record should clearly identify whether the reported figures are gross or net.
For allocator benchmarking, net performance is generally the more relevant measure because it reflects the investor experience.
2. Using the Wrong Reporting Date
Private fund performance changes from quarter to quarter as valuations move and cash flows occur. A figure reported as of March 31 should not be treated as if it were current through June 30.
The reporting date should be captured with every Net IRR, TVPI, DPI, and RVPI observation. When several reports are available for the same fund, the most recent valid observation for each metric should be used for current comparisons. Older observations can remain in the historical record, but they should not be counted as separate funds in the same benchmark.
3. Confusing TVPI, DPI, and RVPI
The three multiple-based measures answer different questions:
- TVPI measures total value, including both realized distributions and remaining value.
- DPI measures how much capital has actually been distributed to investors.
- RVPI measures the value still held in the fund.
One of the simplest data-quality checks is that TVPI should generally equal DPI plus RVPI, subject to rounding and reporting differences. If the figures do not reconcile, the values may have been entered in the wrong fields or taken from different reporting periods.
Confusing these measures can lead to the wrong conclusion about a fund. A 1.50x TVPI with 0.20x DPI represents a very different liquidity profile from a 1.50x TVPI with 1.10x DPI.
4. Treating Missing Values as Zero
A missing metric is not the same as a reported value of zero. If a fund does not provide DPI, entering 0.00x will incorrectly imply that it has made no distributions. The same issue applies to Net IRR, TVPI, and RVPI.
Missing values should remain blank and be excluded from calculations unless the source explicitly reports zero. Otherwise, median and quartile benchmarks can be pulled downward by data that was never reported in the first place.
5. Entering Percentages and Multiples in the Wrong Format
Percentages are particularly vulnerable to formatting errors. A 12.5% Net IRR may be stored as 12.5 in one system and 0.125 in another. If the required format is not clear, the resulting value may be off by a factor of 100.
Multiples create a different problem. TVPI of 1.25x should be stored as 1.25, not 125 or 1.25%. Standard input rules and range checks can catch many of these errors before they enter the benchmark dataset.
6. Assigning the Wrong Vintage or Strategy
A fund should be compared with peers that had a similar starting point and investment mandate. An incorrect vintage year can place a mature fund beside newly formed vehicles that have had little time to deploy capital or realize investments.
Strategy classification matters just as much. A direct lending fund should not automatically be compared with distressed debt, just as a growth equity fund should not be grouped with buyout funds solely because both fall within private equity.
Use the fund's stated investment strategy and a consistent vintage-year convention, such as the year of its first investment or first capital call. When a fund spans several strategies, document the methodology used to select its primary category.
7. Counting the Same Fund More Than Once
The same fund may appear in a quarterly report, a public pension disclosure, a consultant database, and a manager presentation. Without a consistent fund identifier, those records can be mistaken for separate vehicles.
Duplicate funds distort both sample sizes and percentile calculations. Each fund should have a stable identifier, and repeated performance observations should be organized by reporting date. For current benchmarks, only the latest valid observation for each metric should be included.
8. Failing to Preserve the Source
Every performance figure should be traceable to its source document and reporting period. Without that record, it becomes difficult to resolve discrepancies, verify whether a figure is gross or net, or determine why a number changed.
Source hierarchy also matters. LP capital account statements, audited financials, and LP portfolio reports generally provide stronger evidence of investor-level performance than manager marketing materials. When two sources conflict, the database should retain enough detail to show which figure was selected and why.
A Practical Data-Entry Checklist
Before adding a performance record, confirm:
- The fund name and stable identifier
- The correct vintage year and strategy
- The reporting date
- Whether performance is gross or net
- Whether percentages and multiples use the required format
- Whether TVPI approximately equals DPI plus RVPI
- Whether blank values are truly missing rather than zero
- Whether a more recent observation already exists
- The source document and page where the figures were reported
The Takeaway
Reliable private fund benchmarking depends on consistent inputs. Most data-quality problems do not come from complex calculations. They come from mixing definitions, reporting periods, formats, or fund classifications.
A disciplined entry and review process makes it possible to compare funds across strategies and vintage years with greater confidence. It also gives investment teams a clearer view of what is driving reported performance, how much value has been realized, and where additional diligence may be required.
