TL;DR. The known-and-knowable evidence standard is the rule that decides which facts can influence a Fair Value mark and which can't. The 2025 edition of the IPEV Guidelines expanded the explanation of how to interpret it. Founders who build a contemporaneous, timestamped evidence trail at round close make every honest fact easy to count; founders who don't leave the Fund to work out what was knowable, and risk their best evidence being discounted when the macro turns.
There are two kinds of evidence a Fair Value committee considers. One counts. The other doesn't. The IPEV Guidelines set out the rule that distinguishes them — the 2022 edition added it to the commentary and the December 2025 edition expanded the explanation — and the rule is stricter than founders typically realise.
This is the fourth piece in the IPEV Founder Series. The first three covered the frame (Why your Series A pitch is really a Fair Value defence in disguise), the calibration discipline (Calibration: the discipline that turns a number into a defensible number), and the maintainable earnings reconciliation (Maintainable earnings: closing the gap between your accounts and the Fund's number). This piece deals with the evidentiary plumbing under all three. For the Guidelines themselves and what the December 2025 edition changed, see our IPEV guidelines 2025 explainer.
The rule is this: at any Fair Value measurement date, the only evidence that can influence the mark is evidence that was known, or knowable, at that date. Specifically:
Known evidence. Facts that were demonstrably in the possession of market participants at the measurement date. The strongest form: dated, timestamped, in the round-close data room or in a regulatory filing.
Knowable evidence. Facts that existed at the measurement date and would reasonably have been available to the valuer through routine inquiry or due diligence. The bar is lower than known: the fact need not have been in anyone's file, but it must have existed and been obtainable at the date.
Hindsight evidence. Information that becomes known, or events that occur, after the measurement date. The 2025 edition calls these subsequent events and says they are not known or knowable at the measurement date. A customer that signed in Q3 is a Q3 event; what can inform a Q2 mark is what was known or knowable at the end of Q2, such as the state of the negotiation.
The discipline matters because a Fair Value committee has to defend its mark against an auditor who runs the same test. Evidence that fails the known-and-knowable test is excluded — even when everyone in the room knows the evidence is real and would have been counted under a more permissive standard.
A fact that only came into being after the measurement date cannot rescue your valuation at that date, however true and material. A fact that existed and was obtainable at the date does count — but the ones that are easiest to defend are those already recorded, dated, before you need them.
The known-or-knowable paragraphs entered the commentary in the 2022 edition. The December 2025 edition, in effect for quarterly reporting periods beginning on or after 1 April 2026, expanded the explanation of how to interpret them. Two added sentences carry most of the change:
The edition does not add a requirement to document excluded evidence, a requirement that evidence be contemporaneous, or a ranking of evidence types. Those are good practice, not IPEV rules. The Guidelines do say the valuer's rationale for significant judgements should be documented, and recording what evidence was considered, what was set aside and why gives the auditor a trail. A later extract of records that existed at the date can still show what was knowable, but a dated snapshot taken at the time is far easier to defend than a reconstruction.
The single most common failure mode is the assumption that what matters is the substance of the evidence, not its timing. It isn't. The known-or-knowable test makes timing the gating question. Three patterns produce avoidable losses:
The retrospective brag. A founder marks a soft quarter with the argument: "but we signed a £2m enterprise customer two weeks after quarter end." The Fund's valuation committee cannot count this. The customer became known after the measurement date.
The "everybody knew" claim. A founder argues that a strategic partnership was effectively in place by the measurement date even though the contract was signed afterwards. The question is what existed and was obtainable at the date — the state of the negotiation, the term sheet, the board papers. A dated artefact answers it; collective recollection rarely persuades an auditor.
The reconstructed pipeline. A founder produces a pipeline view six months after the measurement date that shows what was in the pipeline at the date. It can still show what was knowable if it is built from records that existed at the date, but it invites challenge. A snapshot taken at quarter end does not.
The remedy is operational: the strongest founders run their internal evidence cadence the same way a Fund runs its valuation cadence. Every material customer win, partnership signature, regulatory milestone, and engineering achievement is captured the day it happens, dated, and lodged in the appropriate data-room folder.
Five categories of evidence carry the most weight at a Fair Value committee. Founders who instrument these contemporaneously give the Fund the strongest possible defence at every subsequent mark.
| Category | What it is | What "contemporaneous" looks like |
|---|---|---|
| Customer evidence | Signed contracts, dated invoices, cohort retention, NRR | Customer master with creation dates; cohort cuts produced quarterly; signed contracts in the data room |
| Pipeline evidence | Named opportunities with stage, value, probability | Pipeline snapshot taken on the last day of each quarter and lodged |
| Technology evidence | Architecture diagrams, IP filings, key-engineer commits | Dated architecture documents; patent application receipts; commit history exported quarterly |
| Brand evidence | Trademarks, search-volume data, third-party brand mentions | Trademark certificates; dated Google Trends exports; press archive |
| Organisational evidence | Key-employee retention, hiring funnel, EMI scheme docs | Quarterly headcount and tenure reports; signed employment contracts in the data room |
For founders, the implementation question is not "what evidence is true" but "what evidence is captured." The best evidence is the evidence that exists as a timestamped artefact at the moment the fact becomes true.
Example. A founder's Series B is priced in Q1 at a 12× ARR multiple. In Q3, the public-market SaaS basket has compressed 18%. The Fund's valuation committee considers a partial write-down. The founder produces a Q3 cohort-retention snapshot showing 132% net revenue retention, up from 118% at the round close — dated, exported from the company's data warehouse on the last day of Q3, lodged in the data room. The Fund holds the mark flat: the valuer judges that the retained-customer expansion offsets the basket-multiple compression. A competing portfolio company with the same retention but no Q3 snapshot is marked down 14%. The retention was real in both cases. Only one was documented at the date.
Five operational practices give a founder an evidence base that makes the known-or-knowable test easy to apply, without disrupting normal operations.
1. Quarterly evidence drop. On the last business day of each quarter, the company exports a standard set of artefacts — cohort retention, ARR by customer cohort, pipeline snapshot, headcount and tenure, key-engineer commits, patent and trademark filings, customer NPS — and lodges them in the data room with the date in the file name. Total time: 30-60 minutes per quarter once the cadence is established.
2. Customer master discipline. Every customer record has a creation date, a contract-signed date, and a first-invoice date. These three dates are immutable once set. The discipline is on the date fields, not on the customer information.
3. Pipeline snapshot. A read-only copy of the sales pipeline is captured at quarter end and never edited. The snapshot itself is the evidence — not the live pipeline view.
4. Engineering velocity export. Commit history, story-point burndown, deployment frequency — all exported and lodged. This evidence supports both technology valuation and organisational capital claims.
5. Brand evidence archive. Google Trends export for branded queries, third-party press mentions, trademark certificates, SEO ranking snapshots. Each is captured at the moment it becomes true and dated.
The five practices together cost the company roughly half a day per quarter and produce an evidence base that stands up to an auditor's known-or-knowable questions. For a founder preparing for a Series B or exit, the cumulative effect across eight quarters is decisive.
The auditor checks three things at each annual audit of a Fund's portfolio. Founders who understand the audit perspective build packs that pass cleanly.
Audit test 1 — Evidence date integrity. Were the artefacts in the data room demonstrably created on or before the measurement date? File metadata, audit logs, and version histories are checked. A document created after the measurement date draws closer scrutiny: it may still evidence conditions at the date, but the auditor will want to see the underlying records.
Audit test 2 — Evidence completeness. Does the evidence base cover all the categories the valuation technique requires? A DCF that relies on a 25% gross margin needs the gross margin evidence in the data room. The absence of evidence on a load-bearing input is itself a flag.
Audit test 3 — Evidence consistency. Is the evidence consistent across the valuation pack? Customer counts in the cohort retention export should match customer counts in the management accounts. Inconsistencies trigger a deeper audit and often a re-mark.
Opagio Intangibles records a company's intangible assets in the Value Drivers Register, with the evidence behind each one, so the company can show the Fund what it held at each measurement date. That evidence informs the Fund's valuer; it does not replace the valuer's judgement on what was known or knowable, or on fair value.
For founders, the practical implication is that adopting an evidence cadence is a multi-year operating discipline, not a documentation push at round close. Funds notice the difference within two quarters.
Warning. The single most expensive evidence failure is the absence of cohort retention data at the round close. Cohort retention is the load-bearing input for nearly every SaaS Fair Value mark. A founder who closes a round without a clean cohort-retention export gives up 10-20% of the valuation range on day one.
This is the fourth piece in the IPEV Founder Series and closes the four-part frame.
Together they map onto the four lenses of the Round Readiness Diagnostic, which scores founders against these four areas in eight minutes and routes each gap to the operating evidence that closes it.
Further reading. Known and Knowable Evidence, Data Room, Round-Ready Academy Lesson 5: Building the Series A Asset Register.
Next step. Run the Round Readiness Diagnostic. The diagnostic surfaces the five evidence categories above as discrete gaps and routes each to the operating cadence that closes it.
Mark Hillier is Co-Founder and CCO of Opagio. Thirty years of commercial growth and private equity exit preparation. Opagio builds the intangible asset evidence platform that institutional investors expect. About the team →
Calibration is the IPEV discipline that converts a transaction price into an ongoing defensible Fair Value. The IPEV Guidelines say calibration is required by accounting standards. The founders who supply calibration evidence at the round close — multiples, peer comparables, drift indicators — give the Fund the bridge it needs to defend the entry valuation at every subsequent quarter.
Read more →
Maintainable earnings is the figure the Fund actually values you on, and it is rarely the EBITDA at the bottom of your management accounts. The IPEV multiples guidance asks the valuer to be satisfied that the earnings used are a reasonable estimate of maintainable earnings. Founders who supply the reconciliation defend a higher number; founders who don't, watch the Fund build a lower one.
Read more →
Sophisticated investors run a Fair Value model on you for years after the round closes — and the offer at term sheet reflects that model, not the pitch deck. Founders who build the evidence base before the room walk in defending a number, not optimising one. The 2025 IPEV Valuation Guidelines expanded the guidance on how funds evidence that number; the founders who notice will price their rounds at the top of the range.
Read more →Get the latest insights on intangible asset growth and productivity delivered to your inbox.
Take the free intangible asset assessment to see where your business stands across Opagio 12.