In-article and in-gallery ads survey

The research question

Participants viewed mock newspaper articles and image galleries carrying ads between paragraphs or inside a gallery slideshow, at several sizes and in static and animated versions, alongside a no-ad experience and a Standard-compliant experience as reference points.

Method and sample

9,076 ad-blocking and ad-filtering participants aged 16+ from the US, Germany and France, on desktop. Ratings covered disruptiveness, intrusiveness, annoyance, enjoyment and overall satisfaction with the page.

Main findings

• —Animated in-content ads were rated worst in both settings, on every dimension.
• —In this study, in-content ads larger than the 840 × 150 size tested were rated clearly negatively; smaller in-content formats and ads placed at the end of an image gallery performed better, though still not as well as the existing Standard-compliant reference. This describes this study’s own tested sizes, not a general rule — the Standard’s own in-content height limit has since been raised (from 150px to 250px) by the later 2024 300 × 250 study.
• —Interrupting a gallery mid-slideshow was rated far worse than an ad at the end of the gallery.
• —Five formats cleared the 35% disruptiveness criterion: static 728 × 90, 840 × 150 and 300 × 250 between paragraphs, and static 300 × 250 and 468 × 400 at the end of a gallery.
• —The static 300 × 250 between paragraphs was flagged as an edge case — it met the threshold but sat close to rejected formats in the fuller analysis, and younger participants rated it significantly worse.
• —In this study, pages using Acceptable Ads formats and pages with no ads at all rated very closely on how obstructed users felt by the page — one of several dimensions tested, and the one the report highlights for this comparison specifically.
• —Around a third of participants had heard of Acceptable Ads; once explained, most viewed the concept positively.

Clearing the threshold makes a format eligible for consideration. Any change to the Standard is a separate Committee decision, published for a month of user feedback.

Limitations and context

• —Only one ad unit was shown per page, to isolate the effect of position; real pages usually carry more.
• —Only non-native ads were tested; results may not transfer to native in-content formats.
• —Contextual fit between ad and article content was not tested.
• —Three markets only, and a mock-page environment rather than live websites.