A brand sets aside money to advertise. By the time any of it reaches a news publisher, that money has passed through a planning team, a buying team, an automated auction, and several fee layers. Each layer takes a cut or shapes a choice, and each choice tends to favor channels that are easy to measure and cheap to buy.
That is the core answer: a media dollar lands where the agency chain finds it easiest and most profitable to place it. News inventory is often neither. Understanding the chain — the plan, the buy, the fees, and the rebates — explains why publishers sit at the end of a long, discounted pipeline.
What does a media buying agency actually do?
A media buying agency purchases advertising space on a brand's behalf and tries to make sure those placements reach the right audience at the right price. According to Darkroom Agency's guide to media buying, the job combines audience research, inventory negotiation, flighting (scheduling ads over a campaign period), real-time optimization, brand-safety controls, and measurement.
Planning and buying are distinct jobs that feed each other. Planning decides what to say, where to say it, and when. Buying executes that plan, manages placements, and tunes the campaign while it runs. In modern digital buying, much of the execution is automated: an auction fires when a person loads a page, software bids on the impression, and the winner's ad serves in milliseconds.
For a news publisher, this matters at the moment of the auction. If the brand's targeting rules, allow lists, or verification tools screen out news content — often lumped in with "unsafe" categories — the publisher never even enters the bidding. The exclusion happens before price is discussed.
How do agencies charge for all this?
Agency fees come in several standard shapes, and each shapes incentives differently. According to GoCardless's breakdown of agency payment models, the most traditional is the commission model, in which the agency takes a cut — usually described as around 15 percent — of whatever the client spends on media. On a $20,000 television buy, the agency would keep $3,000 and pass $17,000 to the network.
Other common models include hourly billing, flat retainers, fixed project fees, and value-based arrangements that tie payment to performance metrics such as sales or page views. Darkroom Agency's guide adds a hybrid structure — a base retainer plus performance incentives or a small percentage of spend — and notes a recurring watch-out for the commission model: brands should ensure transparency on pass-through costs, the money the agency bills onward for media, data, and tools.
The incentive problem is structural. Under a percentage-of-spend fee, an agency earns more when the budget grows, regardless of where the money performs best. Under a flat retainer, the agency earns the same whether it places media in premium news environments or the cheapest available inventory. Neither model pays the agency extra for steering dollars toward journalism.
What happens between the media plan and the publisher?
Once a plan is approved, the buying team converts it into orders. In traditional media, that means an insertion order — a document committing the brand to specific placements at agreed prices and dates. In digital media, it more often means configuring campaigns in bidding platforms and letting auctions decide.
Along the way, several parties touch the money. Agencies negotiate bulk rates and private marketplace deals, which Darkroom Agency describes as unlocking better rates than most brands can secure alone. Verification vendors charge to check that ads are viewable and fraud-free. Data providers charge for audience segments. Each transaction is small; stacked together, they consume a meaningful share of the original budget before a publisher invoices anyone. We covered a connected angle in What publisher AI licensing deals contain — and what they quietly give away.
The discounted end of the chain is the predictable result. A news publisher typically competes in open auctions against cheaper inventory, subject to brand-safety filters that may exclude news outright, and receives whatever remains after the fee stack. The publisher's own sales team, where one exists, competes for direct deals that skip some of these layers — but direct deals require the agency to do more work for the same fee.
Why does news inventory fare worse than other channels?
Three forces work against news placements, and none of them requires anyone to dislike journalism.
- Measurability. Performance channels — search, paid social, retail media — report conversions directly. A brand can see what a dollar bought. News advertising builds awareness and association, which are harder to attribute, so automated budget shifts favor the measurable channel.
- Brand-safety defaults. Verification tools classify hard-news coverage alongside genuinely unsafe content. Avoiding the category is the low-effort choice; refining the blocklist takes work nobody is paid extra to do.
- Price pressure. Open-auction dynamics push prices down, and news publishers, often with unsold inventory, accept lower rates to fill pages. Cheap inventory attracts volume; premium environments require an argument.
None of this is hidden, exactly. It is simply distributed: each decision in the chain looks reasonable on its own, and no single decision-maker owns the outcome. The compounding effect shows up in publisher revenue lines, not in any agency's report.
What this means for readers and publishers
Our analysis is that the agency chain is best understood as a series of small, locally rational choices that add up to a systemic bias away from news. The fees are disclosed in contracts; the steering is not disclosed anywhere, because it emerges from incentives rather than decisions.
For readers, the practical consequence is familiar: fewer ads funding the journalism they read, and more pressure on subscriptions and other revenue. Publishers have responded by building direct sales teams, joining private marketplaces that guarantee higher floors, and pursuing entirely different revenue lines — reader revenue, licensing, and commerce. The related economics of these alternatives are covered in our pieces on where programmatic ad money actually goes before publishers get paid and how the loss of classifieds broke the newspaper model. For related coverage, see Where programmatic ad money actually goes before publishers get paid.
For brands, the useful step is a question rather than a directive: ask the agency how much of the media budget reaches the publisher, what the pass-through costs are, and whether news environments are excluded by default or by choice. Darkroom Agency's own guidance flags transparency on pass-through costs as the watch-out for percentage-of-spend deals — the same question, asked of the whole chain, is where accountability starts.
Where the agency chain goes next
Two pressures are reshaping the pipeline. First, automation keeps moving decisions from humans to algorithms, which makes the steering faster and less visible. Second, privacy changes and the decline of third-party tracking are pushing buyers back toward context — the content an ad actually appears next to — which is one argument in news's favor.
What the evidence establishes is the mechanics: fees at the agency layer, discounts and filters at the trading layer, and measurability bias at the planning layer. What remains unknown is how much of the news-advertising gap each contributes, because no party in the chain publishes a full accounting. Until one does, the pipeline stays long, and the publisher stays at its end.
