In this article
Google documented the VideoObject creator property on September 24, 2026, while retaining support for author. It also clarified the interaction types supported by interactionStatistic. Official source.
A practical check
On three video pages, compare the declared creator, thumbnail, date, and view count with the visible video and the data actually available. Omit an invented count instead of publishing a misleading zero.
Example: a company posts a 40-second installation video. Its markup can name the organization that produced it and state its real upload date. If the hosting platform shows views but the page has no reliable count, omit that counter.
What the evidence supports
The useful test is consistency between the page and JSON-LD, followed by whether visitors can play the video. Markup does not guarantee a rich result. Date the check and retain the source documentation so that later changes can be verified.
Why these properties matter
VideoObject markup describes a video on a page; it does not replace the playable video, its thumbnail, or explanatory copy. The creator property identifies the person or organization that made the video. A number in interactionStatistic should represent a measured interaction of the stated type. Copying a count from another platform without a reliable update method can leave stale data in the page.
A repeatable check
Choose one recent and one older video. View each page as a visitor, record its title, upload date, thumbnail, and creator, then compare these with the JSON-LD in the rendered HTML. Test the URL with Google's Rich Results Test, and verify that the video and thumbnail URLs actually work. Record the test date and the field you changed. Passing a syntax test does not establish that the underlying facts are accurate.
Example and limit
A small firm embeds a YouTube demonstration on a service page. It can identify its own team as creator if the team produced the footage. It should not reuse the channel's total views as this video's count. If an accurate count cannot be maintained, omitting interactionStatistic is more honest. Even correct markup does not guarantee a video result.
Questions before publication
Does the video need its own page? For an important service demonstration, a page where the video is easy to find and accompanied by a summary can help. Markup should describe the video on that page, not a video merely mentioned somewhere else.
Which count should be published? Only a number with a known source and an update process. Player views, social platform views, and page visits are different measures.
How can the result be checked? After release, inspect the URL in Search Console and the Rich Results Test, then review available video reports. Missing rich presentation alone does not prove the markup is wrong.
