Most rosters get built by sorting creators into tiers and then arguing about the split between them. The argument feels strategic and usually is not. The labels are borrowed conventions, and the properties they are supposed to stand for can be checked directly on each creator instead.

The tiers are conventions, not categories

Micro, mid-tier, and macro are defined at different follower thresholds by nearly every source that publishes them, and the same creator is often described by two of the labels in the same month. Any cutoff you can quote is somebody’s house style.

The labels also hide the platform. A hundred thousand followers means one thing where reach follows subscriptions and something else where it follows recommendations, and different again in a specialist category than in mass entertainment. Ten thousand engaged viewers in a narrow field can carry more commercial weight than ten times that number in a broad one.

What size actually predicts

A few things do move with scale. Larger creators are usually more professionalized, with representation, a contracting process, and experience working to a brief, which makes them slower to agree and more reliable to deliver. Their deals more often separate usage, exclusivity, and amplification into priced components, where smaller ones tend to be a flat fee. Smaller creators carry more variance, because a single post swings further from their own average.

Larger creators also generally carry a heavier sponsorship load, so any one endorsement sits in a longer queue. That is the factor most directly tied to campaign outcome, and it is measurable per creator rather than inferred from a follower count. Authenticity is not on this list. It depends on how selective a creator is and whether the endorsement fits, and creators of every size vary on both.

Build the mix from the campaign’s jobs

A launch that needs one credible moment of scale plus sustained coverage inside specific communities is describing two roles, not one tier. Decide the roles first, then model expected audience overlap so you know whether the next creator buys new reach or more frequency against the same people. Compare the remaining candidates on relevant reach, audience relevance, production quality, reliability, and price. Those properties differ in ways the tier labels do not, and unlike the labels they compare across platforms.