For policy violations, we measure the number of pieces of content (such as posts, photos, videos or comments) we restored after we originally took action on them.
By “restore,” we mean returning content that we previously removed or removing a cover from content that we previously covered with a warning.
We report content that we restored in response to appeals as well as content we restored that wasn’t directly appealed. We restore content without an appeal for a few reasons, including:
When we made a mistake in removing multiple posts of the same content. In this case, we only need one person to appeal our decision to restore all of the posts.
When we identify an error in our review and restore the content before the person who posted it appeals.
When we remove posts containing links we identify as malicious, and then learn the link isn't harmful anymore. In this case, we can restore the posts. This is particularly true with spam.
It might be tempting to read this metric as an indicator of the mistakes we make in taking action on content. However, as in the example of malicious links noted above, restoring a post doesn't necessarily mean a mistake was made.
We report the total amount of content that Meta restored in each quarter — for example, January 1 through March 31. Keep in mind that this means that the numbers can't be compared directly to content actioned or to appeals for the same quarter. For example, some restored content may have been appealed within the previous time period, and some content appealed may be restored in the next time period.
A piece of content can be any number of things including a post, photo, video or comment. How we count individual pieces of content can be complex and has evolved over time. Learn more about our content actioned metric.
How we measure appeals on accounts, Pages, Groups and Events
The Community Standards Enforcement Report does not currently include any metrics for accounts, pages, groups and events we restored.