Nearly 15 per cent of connected-TV impressions in a 45-day analysis by ad verification firm Peer39 either showed direct evidence of invalid traffic (3.8 per cent) or could not be confirmed as valid or invalid using available signals (a further 10.7 per cent). The finding comes from a report titled The Brand Safety Illusion: Why Web-Level Verification Doesn't Travel to CTV, released on September 29, which argues that brand safety classifications applied to CTV impressions can rest on incomplete or self-declared programme information.

The core technical problem, according to Peer39, is structural. Brand safety tools built for the web read the page surrounding an ad. CTV has no page to scan. Buyers therefore rely on fields such as title, genre and rating supplied with the bid request itself — and those fields can be partial, self-declared or absent entirely, limiting what any safety classification can tell a buyer about the content actually running behind an impression.

To measure how thin that metadata can be, Peer39 compared bid requests for the film Man of Steel against authenticated information about the film. Not one of the observed requests carried the complete set of genre and mood descriptors. In 58 per cent of cases, the bid requests supplied only a single genre: Fantasy. Roughly 11 per cent contained no content object at all.

The commercial implication is that a "safe" label on a CTV impression may reflect very little verified information. Two impressions classified identically can carry entirely different levels of scrutiny — one backed by authenticated programme data, one backed by a lone genre tag supplied by the seller.

"When a buyer sees 'safe,' they should know what was actually checked," said Mario Diez, CEO of Peer39. "A partial genre tag can't tell the full story, and an impression that can't be verified shouldn't be treated the same as one that has been cleared. Buyers need those distinctions before they spend."

The report also covers the limits of film and television ratings as a proxy for suitability, and how web-based suitability rules can misread CTV genres. It sets out questions buyers can ask their existing verification providers to establish exactly what their current controls check.

Peer39's recommendation is procedural rather than technological: make authenticated programme and traffic signals available before a bid is placed, when they can actually inform the buying decision, rather than after the impression has run.

For advertisers shifting budgets into CTV, the report's timing matters. The invalid-traffic findings — 3.8 per cent confirmed, 10.7 per cent unverifiable — suggest measurement gaps that would not pass muster in mature web display or video channels, where verification standards and pre-bid signals are far more established. Whether the industry responds with standardized authenticated content metadata for CTV bid requests, as Peer39 urges, will shape how defensible those budgets become as the channel grows.