A B2B marketing budget review often ends the same way. Someone brings a published percentage of revenue, the CFO asks why that number applies here, and nobody in the room has an answer that survives the follow-up. The percentage is real. The trouble starts when it moves from the population it was measured on and lands on the company you actually run.
Whose budget is the benchmark describing?
Gartner’s 2026 CMO Spend Survey put marketing budgets at an average of 7.8% of company revenue, up from 7.7% in 2025. That is the figure that circulates, usually without the sentence that gives it meaning: Gartner ran that survey between January and March 2026 among 401 CMOs and other marketing leaders across North America, the UK and Europe, and the vast majority of respondents reported annual revenue above $1 billion.
So the benchmark describes enterprises. If your company is at $20 million, you are reading an average produced by organizations with a different cost base, a brand buyers already recognize, and a sales motion you do not run. Borrowing the percentage imports a ratio and leaves the conditions behind. The benchmark is not wrong. It is an accurate measurement of a population you are not in, and the failure is in the transfer.
The second problem comes from the same source. In the 2025 edition of the survey, Gartner found that 59% of CMOs reported insufficient budget to execute their strategy, an improvement on the 64% who said so the year before. Read that beside the average and the problem is clear: matching the benchmark does not establish that your strategy is funded.
Two credible sources disagree by more than a percentage point.
Forrester, writing up its Budget Planning Survey, 2025, reports that European B2B marketers invest 9% of revenue in marketing. Gartner’s 2026 average is 7.8%. Both are credible and recent, and they sit more than a percentage point apart. They did not measure the same population. Forrester also reports that 83% of B2B marketing decision-makers expect increased investment over the next 12 months, which is expectation data rather than spend data, and Forrester publishes no sample size or margin of error for either of its figures.
Data table
| Published figure | Share of company revenue |
|---|---|
| Gartner 2026 average (mostly $1B+ companies) | 7.8% |
| Forrester 2025 survey (European B2B) | 9% |
Two sources that far apart do not give you a target. They give you a spread, and anyone in the room can produce the other figure.
Price the pipeline your revenue plan requires.
Arithmetic replaces the citation. The number that belongs at the top of this model is the new revenue the plan expects marketing to source, not the whole revenue target. Your average deal size turns that revenue into a number of closed deals, and your recorded conversion rates fix the volume each earlier pipeline stage has to carry. Assign that volume to the channels that have produced it before, and price it at what they have actually cost. What comes out is a budget a finance team can audit line by line, which turns the meeting into a review of a shared model. Most of that chain is demand generation work, so the model is only as good as the stage data you keep.
We have found soft joints in this chain. Every conversion rate in the model is historical. If the plan needs a rate the company has never hit, the model returns a number that is precise and wrong. Hold each rate flat at what you have recorded and let the required volume rise instead of the rate. The ask gets larger and it is visible, which beats burying the same ask inside an optimistic assumption. Check the calendar as well: the deals in next year’s plan come from pipeline created the year before, so the spend lands in one period and the revenue it produces lands in another.
Before anyone picks a campaign, most of the budget is committed.
Allocation arguments run on a smaller slice of the budget than the headline total suggests. In the 2025 edition of Gartner’s CMO Spend Survey, paid media alone accounted for 30.6% of marketing budgets, equal to 2.4% of company revenue. Put salaries and tooling on top and the portion genuinely open for debate is a minority of the total.
That split matters in a review, because it separates decisions from the consequences of decisions already made. A signed media contract is not a campaign choice. It is a commitment until it is changed or expires. Bring the split into the room yourself. It also makes the build-or-buy question tractable, since it is easier to weigh what agencies charge and why once you know which of your own costs are locked.
The AI line item is a readiness question.
One line will draw scrutiny this year. Gartner’s 2026 CMO Spend Survey found CMOs allocating an average of 15.3% of marketing budgets to AI, while only 30% reported mature or fully developed AI readiness and 70% said becoming an AI leader is a critical goal for 2026. A CFO reading those three figures together will ask what the 15.3% is buying, and the honest answer is a readiness answer rather than a number.
The same Gartner survey reports that organizations it classes as AI-ready allocate 21.3% of marketing budgets to AI and report budgets of 8.9% of company revenue, against the 7.8% average. That is an association inside one survey. It does not show that AI spending produces a larger budget, and it does not show that a larger budget produces readiness. Presenting it as cause and effect costs you the credibility of every other number you bring.
Where should the first cut land?
The market’s default is on record. In the 2025 edition of Gartner’s survey, 39% of CMOs planned to cut agency spending and 39% planned labor reductions. That is a description of behavior, not a recommendation.
Our argument runs the other way. Capacity, meaning the people and partners who run the work, keeps producing the evidence the next defense depends on: the recorded conversion rates, the channel costs, the stage data that makes the model auditable. Cut it first and this quarter gets easier while next year’s number gets harder to defend. Cut committed spend that carries no measurement before cutting the capacity that produces the measurement.
A number is defensible when the model behind it can be checked and its weakest assumption is stated by you rather than found by someone else. Neither of those depends on what anyone else spends.
FAQ: Setting and defending a B2B marketing budget
What percentage of revenue should a B2B marketing budget be?
We do not give a figure, because no published percentage tells you what your company should budget. Gartner’s 2026 CMO Spend Survey reports a 7.8% average from 401 marketing leaders whose organizations were mostly above $1 billion in revenue. Forrester’s Budget Planning Survey, 2025, puts European B2B marketers at 9%, from a survey whose sample size Forrester does not publish. Adopt either and you are defending a number that was measured somewhere else.
Why does the benchmark fail to settle the argument even for large enterprises?
Most of the companies it describes say the budget is not enough. In the 2025 edition of the Gartner CMO Spend Survey, 59% of CMOs reported insufficient budget to execute their strategy, down from 64% the year before. Matching that average shows that you keep pace with a group, not that your plan is funded.
What do you bring into the room instead of a benchmark?
The pipeline the revenue plan requires, priced. The new revenue the plan expects marketing to source, at your average deal size, fixes the number of closed deals; your recorded conversion rates set the volume each earlier stage carries. Then assign that volume to the channels that have produced it and price it at their recorded cost.
How do you handle a plan that needs a conversion rate the company has never hit?
Keep the improvement out of the model. Hold each conversion rate flat at the level you have recorded and let the required volume rise to meet the revenue plan. The budget goes up and the reason is visible. An assumed rate improvement hides the same increase inside a formula.
Does putting more of the budget into AI mean a bigger budget?
Nothing in the data shows that. Gartner’s 2026 CMO Spend Survey reports that organizations it classes as AI-ready allocate 21.3% of marketing budgets to AI and report budgets of 8.9% of company revenue, against a 7.8% average. Those figures sit together inside one survey, and nothing in it shows which one drives the other.
Which parts of the budget are committed before planning even starts?
The committed share is larger than it looks. Paid media alone accounted for 30.6% of marketing budgets in the 2025 edition of Gartner’s CMO Spend Survey, equal to 2.4% of company revenue, and salaries and tooling sit on top. Name the fixed costs before the discussion opens.
What does the market cut first, and should you copy it?
In the 2025 edition of the Gartner survey, 39% of CMOs planned to cut agency spending and 39% planned labor reductions. That is what the market does, and we would not copy it. Capacity keeps producing the conversion rates and channel costs the next budget defense runs on, so cutting it first buys the quarter and makes next year’s number harder to defend.
COSEOM Team
COSEOM is a multilingual B2B growth agency. Since 2008 the COSEOM team has run international SEO, generative engine optimization, paid media and demand generation programs for B2B software and technology companies, in English, German, Spanish and Portuguese. This account publishes our research, playbooks and product guides. More about COSEOM