Assessing verification coverage in streaming inventory

A verification result describes only impressions that produced usable measurement signals. Ask for delivered, eligible, measured and excluded impression counts before interpreting viewability, invalid-traffic or brand-suitability findings.

Request those counts from each relevant partner, broken out by app, platform and supply path. Ask which method and metric the counts cover, and why any impressions were excluded.

Read the denominator before comparing coverage rates. Delivered impressions are the broad total; eligible impressions are those that could use the chosen method; measured impressions are those that actually produced measurement. A rate based on eligible impressions answers a different question from one based on all delivered impressions.

Ask for viewability, invalid-traffic and brand-requirement coverage separately where relevant. Compare rates only when their denominators match, and do not project a low invalid-traffic rate in the measured subset onto unmeasured inventory.

For CTV, check whether the app has integrated the IAB Tech Lab Open Measurement SDK (OM SDK). It provides common code and libraries for measurement signals; integrated apps can send signals to the Open Measurement Interface Definition (OMID) API, where measurement providers can place tags to collect them.

OM SDK supports third-party verification of invalid traffic and brand requirements, and standardises measurement signals across mobile apps, web and connected TV. Its availability is not proof that every app or campaign impression is covered; ask whether the specific app has integrated it.

Check the ad-delivery implementation as well as the app. VAST 4.1 supports ad verification with Open Measurement; VAST describes expected video-player behaviour for ad responses, but format support alone does not give a measured-impression count.

For server-side ad insertion (SSAI) inventory, ask whether the measurement implementation passes verification signals to the provider. VAST 4.0 includes a mezzanine file for SSAI and separates media files from verification and interactive code; these features do not establish that impressions were measured.

If a high-spend segment has little coverage or exclusions remain unexplained, ask the app or platform and measurement provider to identify the gap and whether OM SDK integration or a supported VAST Open Measurement implementation is available. Keep the segment separate, seek an alternative method or set a buying limit until the gap is understood.

In price and quality decisions, do not treat measured and unmeasured inventory as interchangeable. State in the final report what was measured and what was not.

Verification Coverage Metrics: Understanding the Denominators

Key Verification Coverage Checks for CTV Inventory

Critical Verification Metrics to Request