# Substantiating technical performance claims before publication
A pump manufacturer prints "99.9% uptime" on a spec sheet with no test report behind it. Six months later a competitor files a complaint, a regulator asks for substantiation, and the company has nothing to show. The claim gets retracted, the spec sheets get reprinted, the trade show graphics get pulled, and the sales team spends a quarter explaining the walk-back to distributors. This is not a hypothetical: it is the standard failure pattern behind advertising enforcement actions in industrial markets.
This lesson builds the evidence file standard that prevents it: what documentation a performance claim needs before it goes on a datasheet, a booth banner, or a product page.
Consumer advertising rules were largely written with retail products in mind. But B2B industrial claims are covered by the same core frameworks:
The common thread: a specific, measurable claim ("99.9% uptime," "reduces energy use by 30%," "MTBF of 50,000 hours") is treated as a factual assertion, not marketing color. Regulators and courts expect it to be backed by evidence that existed *when the claim was made*.
Not all evidence is equal. A credible evidence file typically includes:
1. Test methodology, documented in advance. Was the test designed to prove the claim, or does the claim describe what a test happened to show? Regulators look for whether the protocol (test conditions, load, duration, environment) was defined before results came in.
2. Sample size and statistical basis. A claim based on three units tested for a week is not equivalent to a claim based on 200 units tracked over a year. If your "99.9% uptime" figure comes from five pilot installations, the claim needs qualifying language (see below) or a much larger dataset.
3. Third-party verification where feasible. Independent labs (such as UL, TÜV, Intertek) or accredited certification bodies add credibility that in-house testing does not. For safety- or compliance-linked claims, third-party testing is often mandatory, not optional.
4. Reproducibility. Can someone else run the same test and get comparable results? Internal-only data that cannot be replicated is weak evidence in a dispute.
5. Currency. A test report from 2019 does not substantiate a claim about a product revised in 2025. Substantiation must match the current version of the product being sold.
A useful reference for how substantiation standards are applied in practice is the FTC's guidance on advertising and marketing basics, which, while consumer-facing, sets out the same "reasonable basis" test used in B2B enforcement.
Disclaimers and qualifiers do real legal work when they are specific and proximate to the claim. Compare:
The second version tells the buyer exactly what conditions produced the number, which is what regulators and courts actually check. A disclaimer that hides the limitation in 6-point font at the bottom of a trade show banner generally fails a fair-treatment test anyway, because the claim in the headline is what a "reasonable buyer" takes away.
Fair-treatment principle: both the FTC framework and the EU's Unfair Commercial Practices Directive judge claims by their "net impressionimpressionThe total number of times an ad or piece of content is displayed, regardless of clicks. Each display counts as one impression, even to the same person.Voir la définition complète →," meaning what an average customer reasonably understands, not just what the fine print technically says.
Spec sheets and datasheets. These are treated as durable, referenced documents. If a distributor quotes your datasheet number to a customer two years from now, you are still on the hook for it. Version and date every datasheet, and retire old ones actively when specs change.
Trade show graphics. Booth banners and floor graphics often get created by a marketing agency working from an old press release, not the current test file. Build a sign-off step: no performance number goes on a printed graphic without a compliance or engineering initials.
Comparative claims. "50% more efficient than Model X from Competitor Y" is a comparative advertising claim, and it invites the fastest complaints because Competitor Y has standing to challenge it directly. Comparative claims need the tightest evidence file: same test conditions applied to both products, ideally third-party run.
Before any performance number is approved for external use, run it through four questions:
1. Is there a written test report dated before this claim was drafted?
2. Does the sample size support the precision of the number ("99.9%" implies a much larger dataset than "roughly 99%")?
3. Is the qualifying condition (load, environment, duration) stated next to the claim, not buried in a footnote three pages away?
4. Has legal or compliance signed off on this specific version of the asset (not a prior version)?
If any answer is no, the claim gets softened, qualified, or held until the evidence exists.
Vérification des acquis
1. What does the FTC's 'prior substantiation' doctrine require of a company making an objective performance claim?
2. Why does the lesson emphasize that B2B industrial claims carry meaningful legal exposure, contrary to a common assumption?
3. A manufacturer wants to claim a pump has an 'MTBF of 50,000 hours.' What distinguishes this from a vague marketing statement like 'built to last'?
4. Select ALL correct answers about consequences described in the pump manufacturer example of an unsubstantiated claim.
Sélectionnez toutes les réponses correctes.
5. Select ALL correct answers about when sector-specific regulatory overlays may apply to a technical performance claim, beyond general deceptive-practices law.
Sélectionnez toutes les réponses correctes.
The mistake most manufacturers make is treating substantiation as a one-time legal review before a big launch. The stronger model is a living file, maintained per product line:
This also protects against a subtler risk: internal drift, where an engineer's cautious internal estimate ("we're seeing around 99.9% in early pilots") becomes a hard external marketing claim by the time it reaches a datasheet, without anyone re-checking the sample size behind it.
🎬 [VIDEO: "How the FTC Evaluates Advertising Claims" - youtube.com - search for FTC or American Bar Association explainer videos on advertising substantiation standards, useful for seeing how regulators actually apply the "reasonable basis" test to specific claim types]