<b>Why creator income studies systematically overstate earnings</b>

<b>Why creator income studies systematically overstate earnings</b>

Thesis: most published RPM and brand-deal benchmarks suffer from survivorship bias, and the correction is larger than most people assume.

Context. Earnings datasets are usually built from creators who agreed to share figures — typically via a platform, an MCN, or a rate-card aggregator. Creators who churned out of the niche, or never monetized, are absent by construction.

Findings. When a 2024 analysis re-weighted a self-reported sponsorship dataset to include creators who had gone dormant, median estimated annual income fell by roughly 40–60% depending on tier. The headline averages didn't move much; the medians collapsed, because the dropouts clustered at the bottom.

Caveats. The re-weighting itself rests on assumptions about who 'counts' as an active creator, and dormancy is not the same as zero income. Self-reported figures are also unaudited — there is no tax-return validation in any public dataset I'm aware of.

Implications. Treat any cross-creator average as an upper bound on the typical experience, not a central estimate. Medians and percentiles are more honest than means here.

What we still don't know: the true denominator — how many people attempted monetization and quietly failed — because failed creators don't fill out surveys.