Google Ads raises the video engagement threshold from 5 to 10 seconds
Google Ads' engagements reporting help page now says an in-stream video ad only registers an engagement after 10 seconds of viewing, a full view if the ad runs shorter, or a click. A March 15, 2026 archived copy of the same page set the bar for video ads at 5 seconds. PPC News Feed reported the change on October 10, 2026.
An edit to a help page definition
The Google Ads help article on engagements reporting lists what counts as an engagement for each ad type. Today it says in-stream and video app promotion ads need a viewer to stay for 10 seconds or more, to watch the whole ad when it is under 10 seconds, or to click it. The Wayback Machine snapshot from March 15, 2026 told a different story: its general line for video ads accepted 5 seconds or more of viewing, with a full view required for ads under 10 seconds.
PPC News Feed covered the change on October 10, 2026 and credited Hana Kobzová with spotting it. The current help page carries no date for the edit, so the change sits somewhere between the March 15, 2026 archived copy and the October 10 report.
The new rules, format by format
For in-feed video ads and Masthead ads, an engagement now means 10 seconds of muted viewing or a click through to the full video. YouTube Shorts ads use the same 10 second bar (a full view for shorter ads) or a tap on the call-to-action button. Bumper ads and non-skippable in-stream ads are not measured through views, so their engagements rest on clicks alone. Lightbox and Gmail ads count an engagement when someone expands the ad, and PPC News Feed notes these two formats did not change.
The detail is more nuanced than a single number. In the March copy, in-feed video ads and video app promotion ads already had a 10 second threshold; what moved was the 5 second line covering video ads in general. The page's opening example is also the same in both copies, with Google describing an engagement as "watching a video ad for at least 10 seconds". The current version adds Masthead and YouTube Shorts entries that the March copy did not have, and bumper and non-skippable in-stream ads now get a line of their own. The reported metrics stay the same: engagements, engagement rate (engagements divided by impressions) and average cost per engagement (Avg. CPE).
A dip in the report may not be a dip in performance
Doubling the viewing time needed to count as an engagement means the same campaign, audience and creative can log fewer engagements. PPC News Feed makes the same point, warning that engagement totals could fall when advertisers compare periods measured under the new rules. As a result, engagement rate and Avg. CPE may not be directly comparable across the point where the definition changed. A business that reports YouTube campaigns on those two metrics, without knowing about the new definition, could read the drop as lost performance.
The current help page says nothing on whether Google will restate historical data under the new definition, or whether bidding strategies are affected. By its scope, the page deals with engagement reporting; it makes no statement about conversion definitions. On the creative side the reading is clear: if a viewer has to stay for 10 seconds to count, videos that land the message early and hold attention gain value. We covered an earlier structural change to video campaigns, reach and frequency merging into a single goal, separately.
The sources do not mention Türkiye, so this section is commentary
None of the three sources mentions Türkiye. The help page, the archived copy and the PPC News Feed report give the definition without any country split; there is no date, exception or data specific to accounts in Türkiye. What follows is therefore not a sourced fact but explicitly UNALSOFT's commentary.
Our view: Turkish advertisers and agencies that track YouTube ads by engagement rate or Avg. CPE should add a note to their reports when comparing the old and new periods, since the two may rest on different definitions. The unknowns are worth listing one by one: the exact day the change took effect, whether historical data will be recalculated, whether engagement based bidding strategies are affected, and how much engagement counts shift for accounts in Türkiye are not covered by the sources.
UNALSOFT's take
We read this as two separate jobs. The first is reporting: in accounts that target engagement rate or cost per engagement, periods before and after the definition change should be annotated before they sit side by side on one chart, or the lower numbers can drive the wrong decision. The second is creative: because an in-stream engagement now takes 10 seconds of viewing, the opening seconds of a video have to deliver the message and keep the viewer watching. That is why, in our AI ad films work, we treat planning the edit around those first 10 seconds as a step of its own. With no Türkiye-specific data in the sources, this piece does not forecast a local performance impact.
Does your video ad hold viewers for the first 10 seconds?
A short conversation is enough to review your creatives and YouTube reports against the new engagement definition together.