The problem
Two creators, the same market, nine times the difference
Every roster starts with the same question: how many people who could actually walk into a branch will see this? Follower count cannot answer it, and picking on follower count is how most influencer budgets get wasted.
A creator with 25,000 followers delivered nine times the local audience of a creator with 13,000. The smallest creator on the roster had the worst local reach of all.
Same campaign, same city, same month
The evidence
The whole roster, ranked two ways
Sorted by follower count. Watch the local audience column refuse to follow along.
Followers against verified local audience
In-market share applied to reachThe same six, with the numbers behind them
Instagram, San Diego market| Creator | Followers | In-market | Local audience | Authenticity | Engagement |
|---|---|---|---|---|---|
| Creator A | 102,503 | 48.0% | 49,161 | 81.2% | 2.19% |
| Creator B | 49,427 | 40.4% | 19,984 | 87.9% | 1.21% |
| Creator C | 25,013 | 54.1% | 13,532 | 82.3% | 1.51% |
| Creator D | 13,424 | 11.1% | 1,486 | 78.4% | 1.31% |
| Creator E | 11,999 | 30.2% | 3,621 | 85.8% | 1.06% |
| Creator F | 5,676 | 10.2% | 581 | 79.7% | 0.94% |
Read the C and D rows
Creator C has almost twice the following of Creator D and reaches nine times as many local people. On a follower-count spreadsheet those two look like the same buy at a similar price. They are not remotely the same buy.
Small is not the answer either
Creator F is the smallest on the roster and the weakest locally, at 10.2%. "Use micro-influencers" is the same mistake as "use big ones." The only thing that predicts local reach is measuring local reach.
What we pull
Every candidate, before anyone is contacted
This is the full read on one creator. We run it on every name that gets considered, not just the ones that make the list, and the reasoning is documented per creator on the approval list you receive.
Creator C
Instagram · San Diego · family and lifestyle
Audience by city
Audience by age
Audience interests
Audience by country
Audience by gender
Output
Why this creator got approved
Better than half her audience is in the target city, her credibility clears the threshold, and her top two interest categories are family and relationships, which is the profile a membership campaign is trying to reach. The follower count was the least interesting thing about her.
What the location number really means
54.1% is the share the data can place in San Diego. The true share is higher, because a follower only gets placed in a city when their account gives the platform enough signal. Quiet accounts and private accounts are real local people who never get counted, so we report the number we can defend rather than the flattering one.
The screen
What a creator has to clear
Thresholds are set with the client before sourcing starts, so the roster is built against a written standard instead of a taste call. These are typical for a local membership campaign.
In-market audience
Minimum verified share inside the target market. Set per campaign, and reported against at wrap.
Follower credibility
Share of the audience reading as real humans. Screens out bought followers and engagement pods.
Category interest
Share engaging with content in the categories chosen for the brand, picked from more than thirty.
Brand safety
A human reads back through the feed. No threshold replaces someone actually looking at what a creator posts.
Beyond geography
The same method works on an affinity audience rather than a map. For a credit union serving one employer group or trade, we build against interest and behaviour categories and set floors by state or by the markets the branches serve.
Audience overlap is deliberate
Most media buying chases unique reach. On a local roster we want some overlap, so the same person sees several trusted faces saying the same thing inside one window. That repetition is what moves someone, and it is why the roster is chosen as a set rather than one creator at a time.