A CTV genre filter can look precise while describing the underlying program poorly. Wurl’s October 28 report offers concrete reasons to ask what a buyer can actually inspect: 89% of examined ad requests lacked a usable program name, and 35% of inventory available for its comparison carried a declared genre that did not align with its detected content.
Those figures use different denominators. They are vendor findings, not an estimate of every CTV impression, and disagreement with a detection method does not establish fraud or an unsafe placement. The useful lesson is to evaluate content signals as part of the purchase, rather than treating a targeting label as evidence of the program itself.
Separate absence from disagreement
Our recommended supplier review uses three categories. A missing signal means the buyer cannot identify the content from that field. A broad signal provides some context but may not support a narrow buying rule. A conflicting signal requires investigation because different descriptions may refer to different levels or moments of a program.
Ask for a table of field coverage with the sample period, supply scope and denominator written beside each percentage. Keep request-level coverage separate from delivered-impression reporting. A beautifully presented overall average is difficult to use if the planned deal contains a different mix of publishers or if only a subset can be matched to underlying programming.
Turn the benchmark into a bounded audit
Before changing the entire plan, select a manageable sample from the supply you intend to buy. Ask the supplier to show the declared label, the available program evidence and the method used to compare them. Document how multiple genres, broad channel classifications and changes within a program are handled. Record unmatched cases instead of silently dropping them.
This is our proposed audit process, not a claim that every supplier offers these fields. If an evidence item is unavailable, mark it unavailable and decide whether that uncertainty is acceptable for the campaign. Do not turn an absence of proof into a confident content classification. Equally, do not call a methodological disagreement proof of misconduct.
Specify the response to uncertainty
Build a small decision table for the campaign. For a strict suitability constraint, identify what evidence must be present before buying. For a broader reach objective, define what uncertainty can be accepted and what must be reported afterward. Route ambiguous cases to a named reviewer rather than letting each buyer improvise a different interpretation.
Consider a hypothetical crime documentary carrying a broad entertainment label. The planning question is which buying condition the evidence satisfies, not whether one word is inherently good or bad. A genre does not by itself explain the surrounding scene or the advertiser’s requirements. Ask how the classification relates to the particular decision you need to make.
Measure the cost of stricter rules
A tighter rule can change available reach, price and the composition of supply. Test the proposed rule on a limited part of the plan before making a blanket exclusion. Compare eligible inventory, delivered reach, cost and the share of placements whose context can be explained. Keep business outcomes in view, using the same attribution definitions across comparisons.
Agree a review date and retain the exception log. Expand only if the improvement in decision confidence justifies the operational effort and commercial trade-offs. Wurl sells content-intelligence services, so its report deserves the same scrutiny as any supplier-led evidence. Its findings are a reason to demand clearer denominators and inspectable samples; they are not a ready-made calculation of wasted budget or a guarantee that buying enrichment will improve sales.
