Follower count is the least useful number on a creator’s profile, and it is the one most briefs are built around. It describes the size of a room without saying anything about who is in it or whether they are listening.
Before a brief goes anywhere, a creator should be assessed on things that predict whether the work will land. Most of it can be done from public information in under an hour.
Audience overlap
The first question is whether their audience contains the people the brand needs to reach. Not an adjacent group, not a plausible group. The actual group.
Geography is the most common failure and the easiest to check. A creator with a large following concentrated in markets a brand does not ship to is worth nothing to that brand, however good the content. Ask for audience location data before anything else, and treat reluctance to share it as an answer.
Comment quality, not comment count
Read fifty comments on recent posts. You are looking for evidence that people are responding to the substance rather than performing engagement.
Questions about the product being used, requests for detail, disagreement, people referring back to earlier posts. These indicate an audience that pays attention. Strings of emoji and generic praise indicate an audience that scrolls, or a comment pod.
A creator with a thousand engaged readers is a better commercial bet than one with fifty thousand passive ones, and usually costs a fraction as much.
Consistency of output
Look at the last twenty posts and ask whether the quality is stable. Creators with one viral post and nineteen ordinary ones are being valued on their best day. You are buying an average day.
Consistency also predicts reliability. Someone posting steadily for two years is more likely to deliver on schedule than someone who has posted four times this quarter after a burst last year.
Commercial history
How a creator has handled previous sponsorships tells you most of what you need to know about how they will handle yours.
Find their paid posts. Did the sponsored content look like their normal work, or like an advertisement wearing their clothes? Did the audience react differently to it? Did they disclose the relationship clearly, or bury it? A creator who discloses properly and still gets engagement has an audience that trusts them, which is the entire asset you are renting.
Also check how many sponsorships they run. A feed where every third post is paid has an audience that has learned to skip them.
Signals worth stopping over
Sudden step changes in follower count with no corresponding change in engagement. Engagement rates far outside the norm for the platform and size band, in either direction. Comments in languages inconsistent with the stated audience. Follower growth that continues while posting has stopped.
None of these prove anything alone. Two or three together are enough to move on, because the cost of investigating is higher than the cost of finding someone else.
Turning it into a score
The point of scoring is not precision. It is forcing the same questions to be asked of every candidate so that a decision does not come down to whose content someone happened to enjoy.
Weight audience fit highest, because a perfect creator for the wrong audience produces nothing. Then engagement quality, then consistency, then commercial history. Rate each on a short scale, write one line of justification for each rating, and keep the sheet. When a campaign underperforms, that sheet is how you learn whether the selection was wrong or the brief was.
Then talk to them
If you are on the other side of this, what a buyer checks in a media kit covers what to put in front of us. Scoring narrows a list. It does not make the decision. A conversation reveals whether someone has actually thought about the category, whether they will push back on a brief that will not work, and whether they want to make the thing at all.
Creators who are interested make better work than creators who are available. That does not show up in any metric.