How much should you spend on marketing?

The honest answer is that the percentage matters less than the split, and the split matters less than whether anyone can tell what the spend did.

Every budget conversation starts with a benchmark and ends with a negotiation. The benchmarks are worth knowing, mostly so you can stop arguing about them and get to the questions that actually decide the outcome.

The benchmarks, briefly

Across most industries, marketing budgets sit somewhere between 5 and 12 percent of revenue, with B2B at the lower end and consumer at the higher. Early D2C brands routinely run far above that, sometimes 25 percent or more, because they are buying a customer base rather than defending one. Mature B2B companies with strong referral flow can hold below 5 percent for years and be perfectly healthy.

Use the benchmark as a sanity check, not a target. A number that is defensible against your own pipeline maths will survive scrutiny better than one borrowed from a survey of companies you do not resemble.

The split that decides the outcome

The more consequential decision is how much goes to demand capture and how much to demand creation. Capture is search, retargeting, marketplaces: harvesting intent that already exists. Creation is brand, content, PR, social: making intent exist in the first place.

Capture-only budgets look excellent for about eighteen months and then plateau, because you cannot harvest more demand than the market generates. Creation-only budgets look wasteful for about the same period and then start showing up in branded search and in the cost of everything else. Our longer argument is in brand versus performance.

What to fund first, by stage

• Pre-product-market-fit. Fund learning: a small search budget, one credible content surface, and analytics you trust. Do not fund awareness for a proposition still being rewritten.
• Early traction. Fund capture hard, because intent already exists and is cheap to serve. Keep a fifth aside for creation so the plateau arrives later.
• Scaling. Shift towards creation deliberately. The signal to watch is capture costs rising while volume stalls.
• Mature. Defend distinctiveness, fund retention and category coverage. This is where budget quietly drifts into maintaining reporting rather than buying anything.

The part that gets underfunded every time

Measurement. It is not glamorous, it produces no assets, and it is the first line cut when the number needs to come down. It is also the only reason anyone can tell whether the other 95 percent worked. Budget for it explicitly, including the analytics implementation and someone’s time to interpret it. See the measurement gap.

How to defend the number

Boards rarely reject a marketing budget because it is large. They reject it because it is unaccountable. Three things make it defensible:

• Name what each tranche buys. This much buys pipeline this quarter, this much buys demand next year, this much buys the ability to tell the difference.
• State what you will stop. A budget with no kill criteria reads as a wish.
• Show a leading indicator. Pipeline lags. Share of search moves earlier and is cheap to track, which makes it useful evidence between board meetings.

The cut that does the least damage

When the budget must fall, cut breadth before depth. Fewer channels done properly beats every channel done thinly, because thin presence in six places produces no memory anywhere. Protect the work that compounds — distinctive assets, owned content, search position — and cut the work that stops the day you stop paying for it. That order is unpopular, because the compounding work is the least visible this quarter, and it is still the right order.

What do companies actually spend?

The most quoted benchmark is Gartner's annual CMO Spend Survey. Its 2025 edition, a survey of 402 marketing leaders in North America, the UK and Europe at mostly large companies, found marketing budgets flat at 7.7 percent of overall company revenue, the same as the year before. The average hides a wide spread: half the respondents reported budgets of 6 percent or less. Paid media took the largest share, about 30.6 percent of the total marketing budget, and 59 percent of CMOs said they did not have enough budget to execute their strategy.

Two cautions before using that number. It describes large companies in mature markets, so a young Indian D2C brand buying its first customers will rightly spend a far higher share, and a long-established B2B firm with strong referrals can spend less. And a percentage of revenue says nothing about the split, which is where most of the outcome is decided.

How to build the number from the bottom up

A percentage of revenue is a check, not a method. The defensible budget is built from the task. Start from the target, say a number of new customers, and work backwards through the conversion rates you can evidence to the reach and demand you need. Price the demand you can capture from search and existing channels first, because it is the cheapest. Price the demand you will have to create next, because it is the slowest. Then add the fixed costs people forget: creative, production, tools and measurement.

Compare the total with the benchmark for your stage. If yours is far above it, the plan is probably asking marketing to fix a product or pricing problem. If it is far below, the target is probably a wish. Either way the conversation with the finance director is about the assumptions, which is the conversation you want to be having.