IVT after FraudSensor, viewability still unmeasured
First dated IVT window after add-on 44. September 8 was 0.388 percent. September 9 so far was 1.222 percent. Viewable impressions stayed 0.
Updated 2026-09-10. This page was a plan on 2026-09-09. We will not fill it with unmeasured zeros. After add-on 44 was attached, Beeswax started returning non-zero invalid impressions. Viewability did not.
Source: live performance_agg pull on 2026-09-09, Eastern wall clock, the zone this seat requests. Creatives 12970 through 12979 carried HUMAN FraudSensor Banner tag (add-on 44, vendor 7, $0.05 CPM).
| Eastern day | Impressions | Invalid impressions | Strict IVT |
|---|---|---|---|
| Sep 3-7 (pre-tag, complete days) | (served) | 0 | Unmeasured zero |
| Sep 8 | 5,671 | 22 | 0.388% |
| Sep 9 (partial at pull) | 4,174 | 51 | 1.222% |
Anonymized main-seat brand-name display flights we can cite: 0.523 percent acceptable, 1.544 percent intervention. September 8 is at the acceptable line. September 9 so far sits between acceptable and intervention. That is not a remnant-category benchmark.
What this table still cannot say
Viewable impressions stayed 0. Viewability percent stayed null. Add-on 44 is IVT, not IAS or DoubleVerify. Do not publish a viewability rate from this window.
IVT is not written into impression_log_rollups. A Neon-only read still looks like "this seat has no IVT." Judge from the exchange report. Parse label-cased keys. Strict IVT uses invalid_impressions / impressions. measurable_invalid_impressions is not the numerator.
IVT shows in Beeswax first is the narrative. How to judge display traffic is still IVT plus click-to-page-view, never CTR. The earlier remnant picture without a vendor remains the Supply Bench scorecard.