90% of Marketers Worry About Principal Media. Only 63% Have It in Contract.

The ANA's March 2026 principal media study found 58% of marketers used principal-based buying in the past year, up from 47% in 2024. Yet only 63% of agency contracts address it, and 90% of marketers say their top concern is whether the recommended inventory is in their best interest. The upfront is where that gap gets locked in for a year.

The mechanic that nobody explains in the upfront pitch

Principal media works like this. A holding company has a principal trading unit. That unit buys inventory from networks, streamers, or programmatic supply at scale, often by committing to volume or term. The unit then resells that inventory to clients at a price the client agrees to. The difference between what the unit paid and what the client paid is the agency's margin. That margin is sometimes called the spread.

In a traditional agency relationship, you pay a fee and the agency acts as your agent. They have a duty to recommend what's best for you. In a principal relationship, the agency is the seller. They have a duty to themselves and their shareholders. Both can be honest. They are not the same thing.

Most upfront pitches don't say which one is happening. The deal slides talk about scaled inventory access or guaranteed value. The transaction structure underneath is the part you have to ask for explicitly.

Why upfront season makes this worse

Upfronts lock in a year of commitments inside a few weeks. The contracts are long, the inventory is bundled, and the negotiations are happening at the same time as the budget season. That is a structural environment where opaque arrangements get embedded faster than anyone can audit them.

According to Brandon Doerrer's reporting at Ad Age, brands are asking more questions about media transparency this year. They are not, in his view, getting answers commensurate with the questions. The upfront concession is bandwidth: a brand can spend 30 days arguing the contract or 30 days arguing the placement. Most pick the placement.

That is how a year of inventory ends up bought through a structure the brand never explicitly approved.

The numbers, lined up

Here is the ANA's 2026 principal media study, in the order that matters:

• 58% of marketers used principal media in the past year (up from 47% in 2024).
• 56% expect to use it in the coming year (up from 41%).
• 74% of principal use happens in television.
• 43% happens in the digital open web.
• 90% say their top concern is whether the recommended principal media is truly in their best interest, up from 79% in 2024.
• 57% have internal company guidelines for principal media.
• 63% have agency contracts that address it.

Adoption is up 11 points. Concern is up 11 points. Governance is up far less than either. That gap is the story.

Marketing Dive's coverage of the report frames the same gap differently: more than a third of contracts don't address principal media, or the marketer is not sure. "Not sure" is doing a lot of work in that data point.

What "in your best interest" actually means

The 90% number gets misread as a trust question, like marketers are worried agencies are stealing. That is not what the question is. It is asking: when the agency recommends an inventory bundle, are they recommending it because it is the best fit for the brand, or because their principal unit owns the spread on it?

Both can be true at once. Sometimes the cheapest way to reach the audience really is through the holdco's principal inventory. The point is not that principal media is bad. The point is that the brand cannot tell, because the spread is not disclosed, and the same agency is on both sides of the recommendation.

Forrester has been more diplomatic about this than I will be. In most cases I have seen, when a structure makes audit harder and disclosure optional, it is because the audit and disclosure would change the recommendation.

The question that does not get answered at the upfront

The angle worth carrying into negotiations is not "are you doing principal." Every holdco has a principal arm, the answer is yes. The question is what arbitrage spread are you keeping. According to AdExchanger's analysis, the standard agency response is some version of "we don't disclose underlying economics," which is not an answer, just a polite refusal.

You will probably not get the answer. What you will get is a tell. Agencies that have nothing to hide tend to push back with documents. Agencies that do tend to push back with framing.

The contract clauses that do the actual work

The ANA report includes governance recommendations, and they are unsexy in a useful way. Four things, in order of how often they actually get included:

1. Explicit per-transaction designation. The contract names whether the agency is acting as principal or agent for each placement, before money moves. If the contract is silent, default to "we authorize neither."
2. A spending cap. Industry guidance referenced across recent coverage suggests 10 to 20% of total budget. Pick a number, write it down, do not let it drift.
3. Pre-approval, not notification. The agency cannot route spend through their principal unit unless you have signed off on that specific decision. Notification after the fact is not approval.
4. Audit rights with a named third party. Not "we may audit." A named auditor, a defined scope, a defined timeline. Without that, audit rights are decorative.

Per the ANA, only 57% of marketers have any internal guideline for this. That means there is a material gap between the average brand's contracting position and the top quartile's. Closing that gap before the upfront paper goes out is roughly four hours of legal review and one strongly worded email.

Connected reading

If you have followed WPP's top 25 clients cutting 9.4%, you already know the holdcos are under fee compression. Principal margin is one of the levers they are pulling to make up the difference. And the measurement concession from Nestlé, Haleon, and Molson Coors is the related symptom: brands are losing their independent ability to verify what's working at the same moment the buying structure is getting harder to audit. Those two trends are not separate.

What I'd actually do this week

If you are inside an upfront negotiation right now, the realistic action is not "rewrite the contract." It is "add three lines to it." Designation per transaction. Spending cap. Pre-approval requirement. The audit clause is the one to fight for, but it is also the one you may have to trade for closing other concessions.

Write the lines yourself. Don't ask the holdco's procurement team to draft them. The contracts that come back with their language already in them are the ones that have nothing in them.

The part that surprised me reading the ANA study back to back with the Ad Age piece is that the brands asking the most questions tend to also be the ones with the strongest contracts. The questions are not making the contracts stronger. The contracts are making the questions safer to ask. If you are in the 37% with no contractual language on principal, the answer to "are you doing principal" is going to be the answer the agency thinks is least likely to lose the account, and you will not be able to tell which one that is.