Data sourced from Dakota Private Markets, the private fund performance platform powered by Dakota. Learn More | Request Access
A GP raising a $400 million lower-middle-market buyout fund sends a deck showing top-quartile TVPI against "private equity" broadly. The number looks great. It also tells the allocator nothing about how the fund performed against other lower-middle-market buyout managers in the same vintage, which is the comparison that actually predicts whether this GP can execute the strategy again. Benchmarking at the wrong altitude is one of the most common, least discussed errors in fund evaluation, and it runs in both directions: GPs cherry-pick the broad comparison that flatters them, and allocators sometimes apply strategy-level rigor to a decision that's actually about asset-class allocation.
Strategy-level benchmarking asks: how did this fund perform against managers pursuing the same strategy, in the same vintage, at a comparable scale? Buyout against buyout, growth equity against growth equity, direct lending against direct lending. This is the comparison that isolates manager skill from strategy tailwinds. If lower-middle-market buyout as a category returned 18% net IRR in the 2019 vintage and your fund returned 22%, that's a real signal. If the category returned 14% and your fund returned 15%, top-quartile buyout-versus-buyout, that 100 basis points of "outperformance" against all of private equity is doing a lot of unearned work.
Asset-class-level benchmarking asks a different question entirely: how did private equity, private credit, or real assets perform as a category, independent of which sub-strategy or manager delivered it? This is the right lens for a pension staff member setting a target allocation to private equity as a portfolio sleeve, or a consultant advising on the split between private credit and public fixed income. Asset-class comparisons are built for top-down portfolio construction, not manager selection. They tell you whether the category is worth the illiquidity premium and fee load, not which manager inside it is skilled.
The flattering rollup. A manager whose specific strategy underperformed its peer set reaches for the broader asset-class number instead, because private equity as a whole looks better than direct lending did that vintage. The deck isn't lying, exactly. It's answering a question nobody asked.
The wrong-altitude diligence request. An allocator building a search for a single sub-strategy, say infrastructure debt, pulls a real assets asset-class benchmark because it's the dataset on hand. Every manager in the search then looks either uniformly strong or uniformly weak against a category that includes strategies with entirely different risk and return profiles. The search produces a shortlist that says more about the benchmark than the managers.
The portfolio-construction shortcut. A consultant recommending a shift in the private markets allocation cites strategy-level dispersion, "top-quartile venture returned 3x, bottom-quartile returned 0.8x," as if that variance argues against the asset class itself. It argues against manager selection risk within venture. It says nothing about whether private equity as a category still merits its target weight relative to public markets.
Three signals it's worth pausing to check which lens you're actually looking at:
Bottom line: strategy-level benchmarks tell you if the manager is good, asset-class benchmarks tell you if the category is worth the allocation, and conflating the two is how mediocre funds get mistaken for top performers and good strategies get mistaken for bad categories.
Dakota Private Markets gives allocators and GPs fund-level performance data (Net IRR, TVPI, DPI, RVPI) benchmarked at the strategy, sub-strategy, and vintage-year level, not just rolled up to the asset class. That's the difference between a comparison that flatters a pitch deck and one that survives diligence.