One display placement
Separating invalid traffic from a 119.5 million impression headline
A high-volume display placement retained 68.46% of gross impressions after traffic-quality and count-once filtering.
Customer measurement stories
What independent measurement revealed across FMCG, streaming media, and fashion campaigns. Every story starts with observed delivery and ends with a practical lesson for the next media review.
Customer and partner names are withheld. Each portfolio has two focused studies; some examine different aspects of the same campaign.
14
Detailed studies
7
Customer portfolios
3
Markets represented
25 Oct 2026
Measurement snapshot
Portfolio 01 / FMCG
Display and in-app video measured separately, so invalid traffic and repeat counting do not disappear inside a combined delivery total.
One display placement
A high-volume display placement retained 68.46% of gross impressions after traffic-quality and count-once filtering.
One in-app video placement
In an in-app video placement, duplicate impressions represented 31.02% of gross delivery, compared with 11.18% removed by GIVT filtering.
Portfolio 02 / FMCG
A second FMCG customer, with separate display and in-app studies. Results belong to this portfolio and are not blended with other advertisers.
One display placement
For a second FMCG customer, 160.60 million gross display impressions became 129.03 million net impressions after filtering.
Three in-app placements
Three in-app placements for the same FMCG customer retained 50.08% of gross impressions, with both traffic quality and duplicates contributing to the gap.
Portfolio 03 / FMCG
A third FMCG customer, with distinct display and CTV delivery scopes. Each study follows the same gross-to-net measurement framework.
One display placement
A third FMCG customer’s display placement retained 28.17% of gross impressions; GIVT filtering accounted for nearly all of the difference.
Three CTV placements
Three CTV placements for the same FMCG customer retained 59.95% of gross impressions, using a scope distinct from the customer’s display buy.
59.95%
Net impressions retained · of gross impressions
32.96M gross impressions in scope
Read the studyPortfolio 04 / FMCG · Food & spices
A food business advertising across streaming and gaming inventory, with a separate quality assessment for each buying context.
Four streaming placements
Across four streaming placements, 179.69 million gross impressions became 112.51 million net impressions under the same filtering rules.
62.62%
Net impressions retained · of gross impressions
179.69M gross impressions in scope
Read the studyFive gaming placements
Five gaming placements retained 63.45% of gross impressions after filtering, with distinct contributions from GIVT and duplicates.
Portfolio 05 / FMCG · Dairy
A dairy advertiser with display and video placements across several sources. Each format gets its own measurement basis.
Four display placements
Four display placements recorded 227.69 million gross impressions, with 56.44 million remaining in the final net count.
24.79%
Net impressions retained · of gross impressions
227.69M gross impressions in scope
Read the studyEight video placements
Eight video placements retained 50.30% of gross impressions in the final count, using a scope distinct from the advertiser’s display buy.
Portfolio 06 / Streaming media
A streaming publisher measuring video and overlay delivery for multiple advertiser campaigns, without blending their format-specific results.
Three advertiser video campaigns
Across three advertiser video campaigns, a streaming publisher’s measured net count was 93.82% of gross impressions.
93.82%
Net impressions retained · of gross impressions
15.34M gross impressions in scope
Read the studyTwo overlay campaigns
Two overlay campaigns retained 99.98% of gross impressions. Their result is reported separately from video delivery.
Portfolio 07 / Fashion & retail
A fashion retailer using independent measurement to distinguish invalid traffic, duplicate impressions, and the final count on a display campaign.
One display campaign
A fashion retailer’s display campaign retained 96.53% of gross impressions after filtering, establishing a clear delivery baseline.
96.53%
Net impressions retained · of gross impressions
270.32K gross impressions in scope
Read the studyThe same display campaign, viewed through count-once rules
For the same fashion display campaign, duplicate impressions were a larger adjustment than GIVT exclusions.
Each study states its selected campaign or placement scope and observed UTC dates. The snapshot includes complete days through 24 October 2026. Percentages use gross impressions as their denominator unless explicitly stated otherwise.
GIVT means general invalid traffic. Final net applies the recorded traffic-quality and count-once filters. Duplicates describe repeated impression counting, not audience frequency. Recorded exclusions do not establish financial savings, refunds, or sales uplift.
These are observed campaign measurements, not controlled before-and-after experiments. Studies can overlap, so their totals should not be added together. Customer identities, campaign names, and partner identities are withheld.
See how AdProof can help you reconcile delivery, investigate exclusions, and compare reports on the same basis.
Request a demo