How To Read An Engagement Rate Properly

Every media kit that lands in our inbox quotes an engagement rate, and almost none of them say what it was divided by. That single omission is why two creators with the same headline number can deliver results an order of magnitude apart.

The Denominator Problem

There are three defensible ways to calculate engagement and they produce very different answers. Divide by followers and a large account looks weak. Divide by reach and a creator whose posts travel outside their following looks extraordinary. Divide by impressions and you flatter everyone.

None of these is wrong. What is wrong is quoting one and letting a brand assume another. On the roster every card states the basis in the same place, every time: engagement over reach, measured on the last thirty posts.

Why Thirty Posts

A single viral post can double a ninety day average. A rolling window of thirty posts is long enough to survive one outlier and short enough to reflect what the account is doing now rather than what it did in a better year.

If a creator cannot tell you what the number was divided by, the number is not evidence. It is a claim.

What To Ask For Instead

Ask for saves and shares on the last ten posts. Saves tell you the audience found the post useful enough to come back to, and shares tell you it travelled. Both are harder to inflate than likes and both correlate far better with what a campaign actually returns.

Then ask for the worst performing post of the last month. A creator who will send you that is a creator whose good numbers you can trust.

Running The Numbers Yourself

If you have access to the analytics, the calculation takes a minute. Total interactions across the window, divided by total reach across the same window, times a hundred. Compare it against the category median rather than against another network — a five percent rate on X and a five percent rate on TikTok are not the same achievement.

Everything on Creatrix is published this way, which is why a card occasionally shows a lower rate than the same creator quotes elsewhere. It is the same account. It is just a more honest fraction.