Companion files

Operational Due Diligence in Private Equity

Assessing Managers Beyond the Investment Case

One Excel workbook, and it takes the sharpest warning in the book and measures it. It is free. Nothing is gated behind a sign-up, and no email address is asked for.

The workbook

Download the ZIP20 KB

Or take the file on its own below.

Refining your rubric makes it worse

With eight domains, one scoring 5, and an approval threshold of 2.00, the average clears whenever the other seven average 1.571 or better. Not perfection elsewhere — just good elsewhere.

Now split those eight domains into sixteen to capture more nuance, as programmes regularly do. The fatal domain's weight falls from 12.5 per cent to 6.2, and the average the rest must hold to bury it rises to 1.800. A team that refined its rubric has made it easier to hide a fatal flaw, not harder. Nobody running that exercise would describe it that way. If the model needs to be stronger, add veto rules — not domains.

The weights do not matter. The structure does.

Three weighting schemes across the same twelve managers — equal, controls-heavy, governance-heavy. Rank correlation against equal weighting: 0.979 and 0.993. Decisions changed: zero out of twelve. Not one recommendation moves.

Now change the model's structure instead — replace pure averaging with veto rules on custody, conflicts and valuation. Five decisions out of twelve change.

Committee time spent debating weights is time that cannot change an outcome. The design decision that determines every outcome — whether the model can say this one issue is disqualifying — is usually taken in a sentence and never revisited.

How far apart can two analysts be?

Chapter 3 asks whether two analysts would draw the materiality line in the same place. The workbook answers it: for each manager, the distance to the nearest threshold, converted into analyst-points — single-notch disagreements needed to cross a boundary.

Four managers sit one point from a different recommendation. Three more sit two. The median is two analyst-points out of eight domains — which is to say two analysts disagreeing by one notch on two domains hand the investment committee different answers on more than half the book.

That is not an argument against scoring. It is an argument for printing the distance to the threshold beside the score. A rating of 2.10 one point from a boundary and a rating of 1.20 eight points from one are not the same statement about a manager, and a report showing only the rating has thrown away the more useful half of the information.

Conventions used throughout

Amber fillan input — you may edit these
Grey filla formula — do not overtype these
Checks sheettwenty-four controls, each stating its own verdict

Every manager and score is illustrative, as everything in the book is. What is not illustrative is the arithmetic, which holds for any weighted-average model with a fatal domain in it. One control tests that a manager with a 5 in a veto domain still reads “Approve” under averaging — because that is the finding.

Opening the file

The workbook opens in Microsoft Excel, LibreOffice Calc, Google Sheets and Numbers. It uses no macros and no add-ins, so nothing needs to be enabled or trusted. If your spreadsheet asks to update links on opening, decline — there are none.

Also by Julian R. Sterling

The other books with companion files. The full list of titles is on the author page.

These files accompany Operational Due Diligence in Private Equity. The book is on Amazon.

If this book helped — or didn’t — a few lines on Amazon are worth more than they look: they are what the next reader goes on. Write a review. The workbook stays free either way.