Engagement rate sounds like a defined quantity, the way temperature is. It is not. It is a ratio, and the industry never agreed on the denominator.
The three common ones
- By followers — interactions divided by follower count. The oldest, and the one most quoted in media kits.
- By reach — interactions divided by the number of accounts that actually saw it. The most honest, and the one that needs insights access.
- By impressions — interactions divided by total views, including repeat views. Always the lowest of the three.
The same post can be at 8% by reach and 2% by followers. Neither number is wrong. Quoting one without saying which is.
Which to use
By reach, for judging your own work. It answers the question you actually have: of the people who saw this, how many cared enough to do something? Follower count is irrelevant to that, and it is the number most inflated by inactive accounts.
By followers, only when someone else demands it — brand briefs still ask for it, because it is the only one you can compute from the outside.
Comparing across accounts is mostly pointless
A 50,000-follower account will almost always show a lower rate than a 500-follower one. That is arithmetic about audience size, not a statement about content quality.
Compare an account against itself, over time. That is the comparison where the denominator stays honest.