00 Methodology

Never a number without a source.

This publication exists to be cited, and a figure is only citable if you can see where it came from, when it was checked, and what it means. This page is the standard every benchmark on the site is held to. If a number cannot meet it, we do not publish the number.

Where the data comes from

Every row draws on one or more of four source types, in this order of preference. Each published figure carries its source type, a link where one exists, and the date it was last verified.

Priority 1

Agency & practitioner benchmarks

Anonymised, aggregated campaign data contributed by working media buyers and agencies. No single advertiser, campaign, or rate is ever identifiable. This practitioner data is the seed of the dataset and the reason it can cover markets and offline channels that public tools miss.

Priority 2

Keyword and CPC data (Ahrefs API)

Real cost-per-click figures by country and vertical, pulled programmatically and dated at the time of pull. Used to triangulate search benchmarks and to sanity-check ranges from other sources.

Priority 3

Platform-published data

Figures disclosed by the platforms themselves — Meta, Google, TikTok and others — via advertising transparency tools, published rate cards, and earnings disclosures. Public, verifiable, and linked wherever a row relies on it.

Priority 4

Attributed third-party studies

Reputable industry research, always named and linked. Attribution is not a courtesy here — it is what makes a figure checkable, and checkable figures are what earn citations. We do not launder third-party numbers as our own.

A later source — anonymised, crowdsourced submissions from readers — will be added once the dataset is established, and will be labelled distinctly so you always know which figures are practitioner-reported rather than agency- or platform-sourced.

What low, mid & high mean

We publish a range, not a single figure, because advertising costs are ranges. A CPM in one market can double between a broad awareness buy and a narrow high-intent one. The three points describe a typical trading range in USD — they are planning guidance, not quoted rates.

Low
The bottom of the typical trading range — efficient buying: strong creative, broad or low-competition audiences, favourable seasonality.
Mid
The representative figure most advertisers should plan around. When you cite one number from a page, cite the mid.
High
The top of the typical range — competitive auctions, narrow or premium audiences, peak season. Not a worst-case ceiling; outliers exist beyond it.

Metrics we report

CPM
Cost per 1,000 impressions.
CPC
Cost per click.
CPV
Cost per view (video).
CPL
Cost per lead.
30s spot
Cost of a single 30-second broadcast spot (TV / radio).
Panel / month
Cost to run one out-of-home panel for a month.

All monetary values are in US dollars. Where an underlying source reports in local currency, we convert at the exchange rate on the verification date and note it. We do not adjust historical figures for later currency movement — the date on the row tells you when it was true.

How often it updates

The dataset is refreshed on a public quarterly cadence. Each refresh re-verifies existing rows, updates the ranges that have moved, and records the date on every row it touches. The first verified release ships in Q3 2026; from Q4 2026 onward, pages carry trend context against the prior quarter, because a trend is more useful — and more citable — than a snapshot.

Keeping this promise is deliberate. Most cost-benchmark content online is published once and left to rot. A visible, kept update schedule is the difference between a reference and a stale blog post — so we treat the cadence as part of the product, not an afterthought, and a public changelog will record what changed each quarter.

Corrections

We would rather be corrected than be wrong. If a figure looks off — against your own campaign data, a platform disclosure, or a study we missed — tell us. We review every credible report, and when we change a number we update its verification date and note the change in the quarterly changelog. Data this transparent only works if the errors are visible too.

Contact

Corrections, source suggestions, and data questions:

data@internationalmediabuying.com

This is an independent publication. How it is sourced, verified, funded, and kept neutral is set out above and on our about page.