An investment committee meets on a Tuesday morning to approve the quarterly portfolio valuation. Fifteen holdings to review in an hour. The numbers get presented, no one goes back to the underlying file, the sign-off lands before lunch. Six weeks later, the auditor asks a simple question about one line — why the multiple moved from one quarter to the next — and no one in the room remembers seeing the change, let alone challenging it. This is not a skills problem. It is a format problem: the committee reviewed a result, not a reasoning.
Sign-off is not a review
Approving a number and challenging the reasoning behind it are two different activities, and the first never substitutes for the second. A committee that only checks that the portfolio's total fair value "looks about right" quarter to quarter protects appearances, not the valuation itself: an offset between two lines can hide a material error on each. A useful review looks at the path — the aggregate used, the multiple applied, the walk from enterprise value to the fund's ownership share — not just the final number at the bottom of the page.
The questions that actually move a file
Five points concentrate most of a valuation's risk, and they are the ones a committee should systematically reopen rather than assume settled:
- The aggregate used. Normalised or reported EBITDA, over which trailing twelve months, with which adjustments — a change in aggregate without a change in the business needs an explicit justification.
- The comparable set's composition. A peer that drops out of the set — an IPO, a delisting, a shift in profile — and gets replaced without a note attached moves the multiple without anyone actually deciding to. The selection framework matters as much as the multiple it produces (see comparable company selection).
- Where the illiquidity discount sits. Applied to the multiple or to the holding's fair value, the same DLOM does not produce the same result at an identical rate (see DLOM: where to apply the discount).
- How hybrid instruments are handled. Capitalised PIK interest, a warrant, a ratchet clause: each carries its own treatment convention, and conflating exposure with cash flow distorts the multiple presented.
- Quarter-to-quarter consistency. A number that moves with nothing changed in the underlying business is the most reliable signal that an assumption shifted without being documented.
Take a worked example
A holding at a 40% ownership stake shows, last quarter, a retained EBITDA of 5,000K€, a comparable multiple of 8.00x, and net debt of 12,000K€. Enterprise value comes to 40,000K€, equity to 28,000K€, and the holding's fair value to 11,200K€.
This quarter, the analyst applies a 6.50x multiple: one comparable left the set after its IPO and was replaced with a smaller peer, with no note attached to the file. Enterprise value drops to 32,500K€, equity to 20,500K€, and the holding's fair value to 8,200K€ — a 26.8% drop in one quarter, with EBITDA unchanged.
| Multiple | Enterprise value | Equity | Holding FV (40%) | |
|---|---|---|---|---|
| Prior quarter | 8.00x | 40,000K€ | 28,000K€ | 11,200K€ |
| This quarter, before pushback | 6.50x | 32,500K€ | 20,500K€ | 8,200K€ |
| This quarter, after pushback | 7.20x | 36,000K€ | 24,000K€ | 9,600K€ |
The committee's question — why did this multiple change — forces the analyst back to the new set, to document the size gap against the peer that was dropped, and to land on an adjusted 7.20x multiple instead of the raw 6.50x. The bottom line is still a decline, 14.3% instead of 26.8%, but it is now an explained decline, not the artefact of an undocumented peer swap.
What happens once the pushback lands
A file already circulated to committee never gets fixed in place. Once a number has been presented, changing it within the same version erases the record of what was actually shown in the room — the auditor can no longer reconstruct what the committee approved. A correction that comes out of committee pushback should open a new version, timestamped and attributable, separate from the one first presented. That is what a sealed valuation campaign's Freeze & Lock enforces: approved numbers stay exactly as shown, and any later revision carries its own date and author instead of overwriting what came before.
Closing the loop without rewriting the record
A committee that systematically asks these five questions changes what the meeting is for: it stops being a checkpoint and becomes the moment the file actually gets tested. In Jolv's valuation campaigns, every step of the path — aggregate, multiple, the walk to ownership share, discount — is timestamped and attributable from the moment the file is created, and a correction made after distribution opens a dated version instead of overwriting the previous one. The valuation methods applied to each instrument family are detailed on the valuation methods page, and access terms are on the pricing page.
Sources
- IPEV Board, International Private Equity and Venture Capital Valuation Guidelines, December 2025 edition, "Application of the Guidelines", §1.6 and §2.7
- Commission Delegated Regulation (EU) No 231/2013 of 19 December 2012 supplementing Directive 2011/61/EU, OJ L 83, 22.3.2013, Articles 67 and 70
- AMF, Summary of SPOT inspections on the valuation of unlisted holdings in private equity management companies, December 2018 (in French)