Every fund holding an illiquid position is already running a valuation policy. Most just haven't written it down. The marks happen. A method gets used, then used again because it was used last time. The policy exists as habit — right up until someone with subpoena power asks to see it on paper.

That's the moment it becomes real. An examiner, an auditor, or a large LP's operational due diligence team asks a deceptively simple question: how do you value the things that don't trade? And the honest answer at a lot of funds — "consistently, but it's not formally documented" — is the wrong answer to give any of them.

· · ·

What a private fund valuation policy has to cover

A credible valuation policy isn't long, but it has to be specific. It defines the fair-value hierarchy as it applies to your book. It states the methodology you use for each kind of asset you actually hold — not a textbook list, your list. It names who approves marks and how often. It says what evidence supports a mark and where that evidence lives. And it commits to doing the same thing next quarter that you did this one, because consistency is the whole game.

The reason this document is so often missing is that it sits in an awkward gap. It's a legal-sounding artifact, so the finance team assumes counsel owns it. It's about your specific positions, so counsel assumes finance owns it. It's nobody's Tuesday. So it doesn't exist until it has to.

You don't have a valuation problem. You have a documentation problem — the marks are defensible, the policy explaining them just isn't written down yet.
· · ·

Level 1, Level 2, and the one that actually matters

Level 1 is a screen price — there's nothing to argue about. Level 2 leans on observable inputs. Level 3 is where the scrutiny lives: assets with no active market, valued on models and judgment. That's the private credit loan, the early-stage position, the real asset. An examiner isn't worried that you marked a Treasury correctly. They want to know that your Level 3 marks follow a method you set in advance and applied evenly — not a number that drifted toward whatever the quarter needed.

Which means the valuation policy and the marks have to tell the same story. A policy that says "income approach for credit positions" while a memo quietly uses comparable companies is worse than no policy at all.

· · ·

From blank page to a defensible draft

This is the kind of document a fund should never start from a blank page, because the inputs already exist: your asset classes, your strategy, your jurisdiction, your actual positions. FundOS uses them. Its SEC Exam Readiness tool generates a valuation policy draft from your fund's profile — the hierarchy, the methodology by asset type, the committee and approval cadence — as a starting document you edit, not invent.

And it goes one layer further, where the real exposure is. The Valuation Support feature produces a Level 3 valuation memo per position — company overview, methodology, concluded value, sensitivity — drawn from the deal, its pricing, and comparable data already in the system. So the policy and the marks are generated from the same source, and they agree by construction. When the examiner asks how you value what doesn't trade, the policy and the memo behind every Level 3 mark say the same thing.

· · ·

You already have a valuation policy. The only question is whether it's a written, defensible document — or something you'll be drafting at midnight the week the examiner's letter arrives.


R
Ravi Chachra
Founder, Kela

Ravi founded Kela so the documents a fund needs under scrutiny — valuation policies included — are generated from the fund's own data, not drafted from a blank page at midnight.