Marketing Claim Evidence: What an Inventory of Capability Claims Found

Could your public product claims withstand immediate regulatory scrutiny today? If your marketing pages make concrete promises about software output, speed, or performance, unverified statements create measurable exposure now.

With regulatory bodies like the Federal Trade Commission enforcing civil penalties up to $53,088 per violation—and roughly 670 companies formally notified of potential actions—the threshold for published support is shifting rapidly. In particular, regulators have made clear that AI-involved sponsored content demands two separate disclosures: the commercial relationship and the specific AI involvement. One badge or statement does not satisfy both.

On 2026-08-20, we ran an inventory across 8 company marketing pages. We inventoried 93 discrete capability claims, with a median of 11 per page. Of those 93 claims:

• 70 claims had measurable direct references or did not present immediate substantiation gaps.
• 0 claims cited a source that failed to measure what the statement asserted; when reference links were present, their topical alignment was exact.
• 23 claims would face a substantiation question if challenged, primarily due to absent benchmark documentation.
• 63 claims promised broad functional scope without disclosing the specific tested boundaries.
• 8 of 8 pages carried at least one claim where scope or substantiation remained undefined.
• 0 pages published claims where none carried high exposure.

This follows an earlier batch on 2026-08-19 that inventoried 368 claims across 16 companies. In our experience reading hundreds of live marketing pages, teams rarely fabricate statistics. Instead, the persistent issue is the gap between breadth claimed and breadth tested. A product benchmark might show a 40% reduction in processing time across three specific tabular datasets. By the time that metric reaches the website header, it often reads as an absolute acceleration across all data types. The underlying test was accurate for what was measured, but silent on the broader environment promised to buyers.

Here is our direct view on evidence management: evidence presence is fundamentally different from evidence sufficiency. A five-star badge or a peer-review directory award demonstrates customer satisfaction, yet teams routinely deploy these badges as substantiation for technical efficiency claims. Review aggregates substantiate sentiment; they do not substantiate algorithmic throughput or quantitative performance.

We hold our own content to these parameters. When we scanned knightbyrd.com, we found 2 claims of our own that faced the exact same scoping question regarding unstated test parameters. We logged them and adjusted the published references.

It is important to state our method limitations clearly. These counts are derived from visible text on a single public page. Example output inside product screenshots or interactive walkthroughs can register as a factual assertion, meaning an automated scan provides an initial diagnostic inventory rather than an absolute finding. The ledger records every row so internal teams can inspect each assertion individually.

Reviewing published statements across marketing assets should be an immediate, actionable priority for legal and growth leads before campaign launches. A free scan is available at https://nexus.knightbyrd.com/veris to review raw claim counts across your landing pages.

VERIS is an evidence inventory: it records what was claimed and what evidence was published alongside it. It does not determine whether a claim is adequate, and it is not legal advice.