Data sourced from Dakota Private Markets, the private fund performance platform powered by Dakota. Learn More | Request Access
Every private fund performance database promises the same thing: accurate returns, broad coverage, and reliable benchmarks. Very few of them show you how they get there.
That gap matters more than it looks. Two platforms can show the same fund with two different IRRs, two different vintages, or two different peer groups. If you are underwriting a commitment, defending a track record, or building a benchmark for an investment committee, the quality of the data is the quality of your conclusion.
In this article, we’re walking through the ten things that separate a high-quality private fund performance database from a list of numbers. By the end, you’ll have a checklist you can run against any provider, including the one you use today.
10 Things For a High-Quallity Private Fund Performance Database
1. Named, Fund-Level Data, Not Pooled Averages
Start with the most basic question: can you see the individual funds? Many benchmarking products report anonymized, pooled cohorts. That answers “where does this fund sit against its vintage?” but not “where does it sit against the ten funds I actually consider its competitors?” Named, fund-level records let the comparison be as broad or as narrow as the question requires.
2. A Clear Answer on Where the Data Comes From
Ask every provider how a record gets into the database. Public filings and pension disclosures give independent coverage but can be incomplete or delayed. Manager reporting adds depth but carries the manager’s own framing. The strongest databases draw on both and reconcile one against the other, rather than leaning on a single source.
3. Human Verification on Every Record
Automated scraping is fast and cheap. It is also how a misread filing or a press-release figure ends up in a diligence memo. Look for a database where a research team reviews each record before publication, and ask whether that review happens once or on an ongoing basis.
4. Consistent Metric Definitions
Net IRR, TVPI, DPI, and RVPI look standard, but they are not standardized underneath. Some managers report gross, some net. Some include fees in paid-in capital, some do not. A quality database tells you which basis it uses and applies it consistently, so a 1.6x TVPI on one fund means the same thing as a 1.6x TVPI on the next.
5. Clear As-Of Dates and a Real Update Cadence
Private fund data ages quickly. A fund’s DPI can move meaningfully in two quarters, and a number with no as-of date is a number you cannot defend. Check how often records refresh, whether every metric carries a reporting date, and whether the platform updates continuously or on a lagging quarterly cycle.
Dakota Private Markets holds 159,000+ performance records across 18,000+ private funds, every one reviewed by Dakota’s research team before publication and refreshed continuously rather than on a quarterly lag. Request Access.
6. Accurate Classification by Vintage, Strategy, Geography, and Size
Benchmarks fall apart when funds are tagged inconsistently. A growth equity fund filed as buyout, or a fund assigned the wrong vintage year, quietly distorts every peer group it lands in. Spot-check a few funds you know well and see whether the tagging matches reality.
7. Coverage That Fits How Each Asset Class Actually Works
Breadth matters, but so does treatment. Closed-end private equity, private credit, and real estate funds are measured on IRR and multiples. Hedge funds and evergreen or interval vehicles never close, so they should be measured on annualized returns across time horizons. A database that forces open-end vehicles into a closed-end template, or skips them entirely, leaves gaps in exactly the categories growing fastest.
8. Depth of History Across a Manager’s Fund Series
A single snapshot of a current fund tells you little about a manager. Look for performance records over time and the ability to trace a GP from Fund I through the current raise, with each fund compared against its own vintage-specific peers. That is how you see whether relative performance is improving, holding, or slipping.
9. Custom Peer Groups You Can Show Your Work On
The real test of a database is whether you can build the comparison you actually need. Can you layer vintage, strategy, geography, fund size, and portfolio company sector into one peer group? Can an investment committee or an allocator see exactly which funds are in it and why? A benchmark built on transparent criteria holds up in diligence. A black-box number does not.
10. Context Around the Number
A performance figure rarely answers a research question on its own. Quality platforms connect fund returns to the GP behind them, the portfolio companies inside them, recent transaction activity, and the source documents that back the numbers up. Then they let you export the exact cut you need into Excel or CSV for an IC memo or data room.
Put Your Current Database to the Test
Take three funds you know well and look them up. Check whether they are named, whether the metrics carry a net basis and an as-of date, whether the vintage and strategy tags are right, and whether you can build a peer group around them you would be comfortable defending. If any of those answers come back unclear, your database is making decisions for you that you cannot see.
Dakota Private Markets was built to pass that test. It covers 18,000+ private funds across seven asset classes, with Net IRR, TVPI, DPI, and RVPI standardized at the fund level, evergreen and interval vehicles tracked on annualized returns, and custom benchmarking by vintage, strategy, geography, fund size, and portfolio company sector. Every record is sourced from public filings, LP disclosures, and direct manager reporting, then reviewed by Dakota’s 60-person research team before it is published.
See where your funds actually stand, request access here.
