MKMKDBMediaKitDB
Now live — founding members: claim your profile free and lock the launch price.

2026-11-02 · MKDB Research

202 media kits, 152 different ways to count an audience

MKDB now holds 202 independent media profiles from 32 countries. Try ranking them by audience size and you hit the real story: 176 publish an audience number — under 152 different labels. Here is what that means for anyone buying media.

The most natural thing to do with a database of media profiles is sort it by audience size and start at the top. So we tried. MKDB passed 202 profiles this week, and the ranking fell apart in our hands — not because the numbers are missing, but because they refuse to sit in the same column. Of the 202 profiles, 176 publish an audience figure. Those 176 figures arrive under 152 different labels. Almost every media kit in the world counts its audience its own way, and that quiet fact shapes everything about how independent media actually gets bought.

Subscribers, downloads, users, readers — pick your unit

The biggest single family is subscribers: the word appears in the audience label of 76 profiles. But even inside that family, one kit counts free e-mail subscribers, another counts paid subscribers, and a third adds its social followers into the same line. Step outside the family and the units multiply. Data Engineering Podcast reports 60,000+ downloads a month. TechCabal reports 206,000+ monthly website users. Other profiles in the database count monthly readers, page views, Instagram followers, listens per episode, print readers per issue — each one honest, each one measured, and almost none of them directly comparable to the next.

An analyst comparing chart data on a laptop

This is not sloppiness. A podcast is downloads in a way a newsletter never will be, and a niche site's monthly users say something its e-mail list cannot. The metric follows the medium. The trouble only starts when a buyer lines the numbers up as if they were the same currency — because 200,000 monthly users and 200,000 subscribers are not the same asset, bought the same way, at anything like the same price.

The second translation: country and language

Geography adds another layer. Of the 202 profiles, 165 list a country base, spread across 32 countries: 88 are based in the United States and 77 elsewhere — from Nigeria to Mexico to Australia. Language follows the same pattern. 137 profiles publish in English only; Spanish appears in the language of 38 profiles and Portuguese in 11, and 12 profiles publish bilingually. So a buyer comparing a Spanish-language newsletter in Mexico with an English-language podcast in the United States is translating twice at once: once across languages, and once across units that were never designed to match.

Why MKDB keeps the label attached to the number

Faced with 152 labels, the tempting fix is to force everything into one invented metric. We do the opposite. Every profile in the database keeps its audience exactly as the media published it — the value, the label that says what was counted, the source kit and the date it was collected. Comparison still works; it just has to happen where it is fair: newsletters against newsletters on subscribers, podcasts against podcasts on downloads, sites against sites on monthly users. And when an owner claims a profile and connects their platform, the Verified Audience badge replaces a typed number with one pulled from the source.

The lesson of the 152 labels is not that independent media are hard to measure. It is that audience was never one thing — and any shortlist that pretends otherwise is ranking media by a number that does not exist. Start with the type, compare like with like, and the database turns from a wall of incompatible figures into what it was built to be: a market you can actually read.

Browse the database behind this post

Every figure here comes from a profile in the open MKDB database — with its source.

Search the database