Categories

Every LP aims for a top-quartile private equity manager. It’s considered the industry’s gold standard. It’s what justifies committing to a fund for over 10 years.

But here’s a question we don’t ask often enough: Is a manager who hits top quartile once or twice truly better than one who has delivered second-quartile performance consistently for 30 years?

A recent AI-driven performance analysis changed how I think about that trade-off.

The Excel Approach

Before AI, evaluating manager performance was mostly mechanical.

Pull benchmark data from Preqin, PitchBook, or MSCI

Clean messy spreadsheets
Align vintage years
Plot the fund against its peer set.

The conclusion was usually simple:

“PE manager X is in the first or second quartile. Improved or worsened performance over the last X number of funds” Box checked. Move on.

What AI Unlocked

Recently, we used a closed-environment AI platform to evaluate a US middle-market buyout manager’s full performance history alongside every comparable benchmark constituent peer fund across all vintages. That’s 23 private equity managers with 5+ funds each.

At first glance, the result looked average: The manager we were evaluating fell in the second quartile.

But as we expanded the lens across the whole dataset, a different picture emerged.

This manager has raised 10+ funds since the mid-1990s and has placed in the upper half of its benchmark every single time.

Thirty years. Over ten funds. Zero bottom-half vintages.

That isn’t “average” performance. That is institutional-grade consistency.

Another observation that emerged as we analyzed a much larger data set is that many managers in the peer group who flash top-quartile results once or twice fail to maintain them. Many fall into the second, third, or even fourth quartile in later vintages.

This manager never did.

The Insight We Would Have Missed

Without AI, we likely would have stopped at the shallow conclusion: “Second quartile.” And potentially have overlooked one of the most consistently durable track records in the market.

By removing the manual burden of data processing and formatting, our time shifted toward higher-order questioning:

• How did the strategy perform during past market cycles?
• What about COVID, GFC, etc.?
• Where did deal size and sector focus generate excess returns?
• How did evolution occur without drifting from the core discipline?
• How were lessons from previous funds incorporated into later funds to maintain consistency?

The answers converged on three structural strengths:

1. A disciplined approach to portfolio-company selection and portfolio construction.
2. Conservative valuation and leverage at entry.
3. EBITDA growth vs. multiple expansion as a value creation driver.

The consistency wasn’t coincidental. It was designed and refined over time.

The Takeaway

AI doesn’t replace the analyst. It raises the analytical ceiling by widening the ground for judgment and fundamentally reshaping the questions we can pursue.

Top quartile will always matter. But consistency across market cycles may matter even more.

Note: All analysis was conducted in secure, closed environments. Uploaded data is not used for model training or shared externally.