What engagement rate ranking measures
Meta estimates how likely people are to react, comment, share or click on your ad, then compares that estimate against every other ad that competed for the same audience. The result is reported as a bucket rather than a score: above average, average, or one of the below-average bands. It is one of three ad relevance diagnostics, alongside quality ranking and conversion rate ranking.
Two details change how you read it. It is relative, so your ad can decline in ranking without changing at all if stronger competitors enter the auction. And it only appears once an ad has served enough impressions for Meta to make a reliable estimate — before that the column shows as unavailable, which is not a bad result.
Why engagement rate ranking matters
The same estimate that produces this diagnostic feeds the auction. An ad people are expected to engage with can win impressions at a lower price than an ad they are expected to ignore, so a weak ranking usually shows up as a rising cost per thousand impressions before it shows up anywhere else.
Its real value, though, is that it tells you which part of the ad to fix. Reporting rarely separates a creative problem from an audience problem or a landing page problem. This column, read next to the other two diagnostics, points at the creative and the hook specifically.
Common mistakes with engagement rate ranking
The first is treating it as a score to be maximised. It is a diagnostic, not a target. Ads with a modest ranking that deliver enquiries at a comfortable cost do not need fixing, and I have watched advertisers rebuild a profitable ad because a column said average.
The second is reacting to a single ranking in isolation. Meta intends the three diagnostics to be read together: weak engagement with healthy conversion ranking points at the opening of the creative, while all three weak at once usually means the audience is wrong rather than the ad.
The third is comparing rankings between accounts or between countries. The comparison set is the ads competing for your audience, so a ranking earned in a small Nepali auction and one earned in Australia are not measuring against the same field.
How to act on it
Start with the first moments of the ad, because that is what the estimate mostly reflects. Change the opening frame, the first line of the caption or the format — vertical video instead of a static image, a customer speaking instead of a product shot. Test the new version against the current one rather than editing in place, so you can see whether the ranking and the cost per result moved together.
If several honest creative attempts fail to shift it, stop editing and look at targeting. An audience that has already seen the ad repeatedly will disengage whatever you show them, so check frequency before commissioning new work. The production side of that testing cycle is covered under Meta ad creative.