How to vet a creator before you pay them: a brand's checklist

Ten minutes of public checks that catch most bad deals, what to ask the creator for, and which red flags are worth walking away from.

FOR BRANDSAug 25, 20264 min read

Most influencer marketing money is wasted before the campaign starts, on creators whose audience was never going to buy anything. The checks that catch this take about ten minutes per creator and need no subscription.

Ten minutes, in order

1. Engagement against their size, not in isolation

A 2% engagement rate means nothing on its own. It is strong for a 500,000-follower Instagram account and weak for a 5,000-follower one, because typical rates fall steeply as accounts grow.

Run the account through the fake follower checker or the relevant engagement calculator, which grade against the benchmark for that tier rather than a flat number. Our benchmarks by follower count lists the typical rates if you want to check by hand.

2. The comment-to-like ratio

This one signal catches most bought engagement. On a real account, comments run at a small but steady fraction of likes. When comments sit under 0.3% of likes, the likes were probably purchased, because paid like packages are cheap and paid comment packages are not.

3. Read twenty comments

Not the count, the content. "🔥🔥" and "nice post" from accounts with no profile picture is a pod or a bot farm. Real audiences ask questions, disagree, and reference things from previous posts.

4. Following versus followers

An account above 10,000 followers that still follows more accounts than follow it usually grew by follow-for-follow. That audience did not choose them and will not act on a recommendation.

5. Post count against audience size

Fewer than twenty posts with tens of thousands of followers means one viral moment, a purchased audience, or a renamed account. All three change what you are buying.

6. Consistency over the last three months

Open their last fifteen posts. Steady output and steady performance is a working channel. Long gaps, or a big drop in recent views, means you would be paying yesterday's rate for today's reach.

7. Reach, not followers, on YouTube

For YouTube, ignore subscriber count and look at average views on recent uploads. A 500,000-subscriber channel averaging 20,000 views is a 20,000-view channel. The subscriber count checker shows both side by side.

8. Audience fit

The highest engagement rate in the world is worthless if the audience is in a country you do not ship to, or an age band that does not buy your product. This is the question public data cannot answer, which brings us to the next section.

What to ask the creator for

Public numbers take you most of the way. Two screenshots close the gap, and any professional creator has them ready:

  1. Reach and impressions for their last ten posts, from Instagram Insights, YouTube Studio or TikTok Analytics. This shows real reach rather than follower count.
  2. Audience demographics: top countries and age bands.

Ask for the screenshot showing the date range, not a cropped number. Hesitation or excuses here is itself the answer. Honest creators send these within a day, because they get asked constantly.

If you are running a larger campaign, ask for results from a past collaboration: link clicks, code redemptions, or a sales figure the brand was happy to share.

Red flags worth walking away from

Yellow flags worth a conversation

Compare the shortlist properly

Once three or four creators pass, compare them on the same rows rather than on impressions. Our channel comparison tool puts two or three YouTube channels side by side with the leader marked on every row; the Instagram and TikTok comparison tools do the same from numbers you enter.

Then price each one from reach rather than followers. Sponsorship pricing covers the method creators use, which is also the method you should be checking their quote against.

One last principle

Scores and checkers narrow a list. They do not make the decision. A creator with a slightly lower engagement rate whose audience is exactly your customer will outperform a higher-scoring account whose audience simply likes their content. Use the numbers to remove the obvious mistakes, then use judgement on what is left.

Try it on your own numbers

More in For brands

Get the weekly creator report

Benchmarks, rate changes and new tools, one email a week.