The 70/20/10 marketing budget rule splits your spend three ways: 70% into channels already proven to bring customers, 20% into winners you are scaling up, and 10% into experiments that might become tomorrow’s winners. It stops the two classic small-business mistakes — spending everything on one channel, and spreading money so thin that nothing works. Here is how each bucket works.

What Is the 70/20/10 Marketing Budget Rule?

The rule is a discipline, not a formula from a study: for every $10 of marketing spend, $7 goes to what already works, $2 to what is starting to work, and $1 to trying something new. The percentages matter less than the shape: most money on proven return, a real slice on growth, a small protected slice on learning. Without the split, businesses either never test anything new and slowly fall behind, or chase every new channel and starve the ones paying the bills.

The rule only works if you know which bucket a channel belongs in, and that takes measurement. A channel is “proven” when you can trace customers or enquiries back to it over at least a few months — not when it merely feels effective. Our guide to how much to spend on digital marketing each month covers how to set the total budget the rule then divides.

How Does the 70% Proven Bucket Work?

The 70% bucket funds the channels that already bring customers at a cost you can live with — usually local SEO and Google Business Profile work for service businesses, or search ads and email for shops. This is the money that keeps the lights on, so protect it: it gets spent first, every month, before anything experimental. If a proven channel stops performing, fix it or demote it — but do not quietly raid its budget for a new idea.

Proven does not mean frozen: the 70% bucket still needs optimisation — better landing pages, tighter ad targeting, fresh content for ranking pages. Our digital marketing cost guide breaks down what each channel typically costs, so you can check whether your proven bucket is priced fairly.

What Goes in the 20% Scaling Bucket?

The 20% bucket is for channels that have shown early promise and deserve more fuel — a social campaign that brought its first real enquiries, a content topic that started ranking, a referral offer customers actually use. The test for this bucket is evidence, not enthusiasm: a channel graduates here only after a small trial produced real leads, calls, or sales. Scaling means spending more on the same thing that worked, not trying five new variants at once.

If a scaling channel stalls for two or three months, demote it or cut it — the 20% is for growth, not for funding hope. When one proves itself over several months, it graduates into the 70% bucket and something new takes its place.

Why Spend 10% on Experiments at All?

The 10% bucket is insurance against change. Channels decay: ad costs rise, algorithms shift, and a platform your customers loved can fade in a year. A small, protected experiment budget means you are always testing the next channel while the current ones still pay — so when one declines, a replacement is already half-proven instead of starting from zero.

Keep experiments small and time-boxed. One new channel, one clear test, a fixed budget, and a date to judge it — usually 60 to 90 days. Our Google Ads cost guide for small businesses is a good example of how to size a test: a meaningful trial needs enough spend to learn from, and below that you are guessing. The same applies to social media management costs — test one platform properly before adding a second.

How to Apply the 70/20/10 Rule to a Real Monthly Budget

Take an illustrative $2,000 monthly budget: $1,400 to proven channels — say local SEO and search ads that already bring calls; $400 to scaling — perhaps the content series that started ranking last quarter; and $200 to one experiment — a new platform or referral offer — judged after 60 to 90 days. The numbers are an example, not a prescription: the split is the point, and your totals come from your own budget and margins.

Rebalance quarterly, not weekly. Channels need time to show results, and constant reshuffling destroys the data you need to judge them. At each quarterly review, promote what proved itself, demote what stalled, and pick one new experiment — the rule then runs itself.

When Should You Break the Rule?

Break it when you have nothing proven yet: a brand-new business has no 70% bucket, so it spends its first months testing — small trials across two or three channels until one earns its place. Break it too when a channel collapses — then the emergency is rebuilding the proven bucket, not protecting experiments. And break it for a genuine one-off, such as a website rebuild — but fund it separately and return to the split the next month.

Frequently Asked Questions

What is the 70/20/10 rule in marketing?

It splits a marketing budget into 70% proven channels, 20% scaling winners, and 10% experiments — most money on what works, a slice on growth, and a small slice on learning.

How much should a small business spend on marketing?

It depends on revenue, margins, and goals — our guide to how much to spend on digital marketing each month walks through setting the total before you split it.

Does the 70/20/10 rule work for a brand-new business?

Not at first — a new business has nothing proven, so it tests small trials across a few channels until one earns its place in the 70% bucket.

How often should I rebalance the split?

Quarterly. Channels need time to show results, and rebalancing more often destroys the data you need to judge them.