FMCG · Dairy / Anonymized case study
Assessing video delivery without borrowing a display benchmark
Eight video placements retained 50.30% of gross impressions in the final count, using a scope distinct from the advertiser’s display buy.
Observed window: 24 Aug 2026 to 24 Oct 2026 (UTC)
Snapshot: 25 Oct 2026
38.89%
GIVT excluded, as a share of gross impressions.
The buying question
A mixed-format campaign needs more than one quality summary. Applying the dairy advertiser’s display result to its video buy would blend different placements and delivery patterns. The video placements therefore need their own gross-to-net reconciliation.
What AdProof measured
Across eight video placements, the observed gross count was 161.10 million impressions. After GIVT filtering, 98.44 million remained. The final net count was 81.03 million, equivalent to 50.30% of gross delivery. GIVT exclusions represented 38.89% of gross impressions; duplicates represented 10.81%.
Why separate format scopes matter
These figures describe impression quality and count-once filtering. They are not a video completion rate, a viewability rate, or a measure of unique viewers. A video completion metric would need its own eligible start population and validity rules. Keeping those definitions explicit makes the report useful to both the media buyer and the partner investigating delivery.
Putting the evidence to work
Maintain separate display and video reconciliations, then break the video scope down by placement. Use the same traffic-quality definitions within that analysis. Bring completion and viewability into the review only when their measurement bases are available, rather than dividing unrelated event totals to create a persuasive-looking rate.
Take this into your next campaign review
- Use separate quality summaries for display and video.
- Keep impression quality distinct from completion and viewability.
- Only calculate rates when numerator and denominator share a valid basis.
More from this portfolio
Making a large display delivery gap visiblePut the method into practice
Completion rate and viewability measure different things