Skip to content

How to Choose Influencers for a DTC Brand (Without Starting From Follower Count)

7 min read

Most influencer shortlists are built backwards. Someone opens a database, sorts by follower count, filters by category, and exports the top fifty. The list looks impressive and converts badly. Here is the order the decision actually needs to be made in.

Start with your customer, not the creator

Before you look at a single profile, write down who buys your product: roughly how old they are, where they live, what problem they came to you with, and what else they already spend money on. This takes ten minutes and it is the only thing that makes the rest of the process meaningful.

Everything after this is a search for overlap between that description and a creator's audience. Without it you are not selecting creators, you are ranking them — and follower count is the only thing you can rank by, which is exactly the trap.

Judge audience fit before you judge reach

A creator's audience is not the same as their content. A fitness creator can have an audience that is mostly other fitness creators, which is a fine audience and a terrible one to sell protein powder to. Ask for the audience demographics from their platform analytics — age skew, top locations, gender split — and compare them against the customer description you just wrote.

This is also where a smaller creator often wins outright. Fifteen thousand followers who match your customer profile closely is a better buy than three hundred thousand who match it loosely, and it usually costs a fraction as much.

Read engagement honestly

Engagement rate is easy to quote and easy to misread. Two things make it more useful:

  • Calculate it on recent posts, not a lifetime average. A creator who grew fast a year ago and has been coasting since will look healthy on the average and flat on the last ten posts.
  • Read the comments rather than counting them. Real engagement looks like questions, opinions and inside jokes. A wall of single emojis and generic compliments is the signature of an engagement pod or bought interaction.
  • Check whether the engaged accounts look like your customer. Comments from accounts in the creator's own industry are a different signal from comments by people who would actually buy.

The warning signs worth a manual check

A few things are only visible if a person opens the profile:

  • Follower spikes with no corresponding content that explains them.
  • A recent paid post for a direct competitor — sometimes a dealbreaker, sometimes not, but never something you want to discover after launch.
  • Sponsored posts that consistently underperform their organic posts, which suggests the audience tunes out branded content.
  • Comment sections that are disabled or heavily filtered.
  • Wildly inconsistent posting, which usually predicts a missed deadline on your campaign.

Write a brief that produces a usable shortlist

Whether you are searching yourself or handing this to an agency, the brief needs four things and rarely needs more: your category, your budget range, the outcome the campaign has to produce, and any hard constraints such as territories, competitor exclusions or launch dates.

Budget range in particular is worth stating up front even though it feels like negotiating against yourself. It does the opposite — it removes every creator whose rate you cannot meet, which is most of the wasted time in this process.

This is the whole reason a shortlist can be turned around in minutes rather than days. The slow part of influencer sourcing was never the sourcing. It is the discovery calls, scoping documents and internal handovers stacked in front of it.

Decide what you will measure before you launch

Agree the success metric with your team before the first post goes live, and make sure the tracking for it exists. Unique discount codes per creator, dedicated landing pages, or properly tagged links are all fine — what is not fine is deciding after the campaign that you would like to know conversions when nothing was ever set up to record them.

If you only look at one number afterwards, make it cost per result per creator. It is the number that tells you who to run again, and it is the reason a per-creator report is worth more than a campaign-level summary.