Brand vs Performance Marketing on a Startup Budget: Why 60/40 Breaks

Somebody in your company has quoted the 60/40 rule at you. Possibly a board member, possibly an agency, possibly the CEO after a conference. The claim is that you should be spending 60% of marketing budget on brand and 40% on performance, and that your current split is short-termist.

The rule is real, the research behind it is serious, and the number is also contested by people with equally serious credentials. More importantly, the ratio is close to meaningless at startup spend levels, for a reason that has nothing to do with whether brand building works.

Here is where the number came from, why the B2B version disagrees with itself, why it does not transfer to a small budget, and what to fund instead.

Where 60/40 actually comes from

Les Binet and Peter Field built the rule from the IPA effectiveness databank, a collection of case studies submitted to an advertising effectiveness awards scheme, and popularised it in their 2013 work The Long and the Short of It. The finding: campaigns delivering the largest long-term business effects tended to sit near a 60% brand building, 40% sales activation split, with brand building working slowly through memory and activation working quickly through response.

Two features of that origin matter for you.

It is a media-budget rule. The split describes how to divide advertising money between mass-reach brand work and targeted response work. It presumes you are buying media at a scale where share of voice is a meaningful quantity.

It is derived from award submissions. This is the crux of the main criticism, and it is not a fringe objection. Professor Byron Sharp of the Ehrenberg-Bass Institute has attacked the rule directly, saying in an August 2022 talk: “If you actually read Peter Field and Les Binet’s first report on this, they analysed a very weird data set, which is award submissions.” His broader point is that the number exists partly because the market wanted one: “where there is demand, there will always be supply. People wanted a number.” He also argues practitioners cannot cleanly separate “advertising” from “activation” in the first place, which makes the ratio arbitrary in application.

You do not have to accept Sharp’s conclusion to accept the useful part: a self-selected sample of award entries is not a random sample of businesses, and a rule derived from one should be held loosely.

The B2B number that disagrees with itself

When the rule gets quoted at a B2B company, it usually arrives as 46/54. That figure deserves a closer look, because its own sources do not tell the same story.

So the widely circulated B2B ratio traces back to preliminary work that the sponsoring publisher itself rounded to “closer to 50/50” and flagged as a small sample. That does not make it wrong. It makes it a directional finding wearing a benchmark’s clothing, and worth exactly the confidence its authors gave it.

One finding from the same B2B work is more useful than the ratio, and almost nobody quotes it: Binet and Field found no examples in their B2B sample of campaigns that worked by increasing brand loyalty. The campaigns that worked grew the customer base through reach. If you take one thing from the B2B research, take that, not the percentage.

Why the ratio does not transfer to a small budget

Here is the part that gets skipped, and it is not about whether brand building works. It is about what the rule is a rule about.

The mechanism underneath 60/40 is share of voice. Brand building works, in this model, by buying enough presence relative to competitors that memory accumulates faster than it decays. Excess share of voice above share of market is the input.

Now run the arithmetic on a startup. Take a company with a marketing budget in the low tens of thousands per month, which is the realistic band for most Series A teams once salaries are counted (our Series A budget benchmarks reconcile the published ranges). Allocate 46% of it to brand-building media in a category where the incumbents spend seven figures a quarter. Your resulting share of voice rounds to zero. You have not bought slow-accumulating memory. You have bought a rounding error, and you have taken those dollars out of the one horizon where they would have produced a measurable answer.

The rule is not wrong at that scale. It is inapplicable. A ratio designed to allocate media spend between two mechanisms only helps once both mechanisms are actually available to you. Below the threshold where paid reach registers, the brand half of the equation has to be bought with something other than media.

There is a second reason the ratio misleads early companies, and it is more fundamental. Brand building assumes you know what you want remembered. A company still adjusting its positioning every quarter, which describes most Series A startups honestly, would be paying to build memory of a message it is about to change.

What to fund instead

The goal of brand spend is that a buyer entering the market already knows who you are and what you stand for. There are cheaper ways to produce that at your scale, and none of them appear in a media plan.

A position specific enough to be repeated by someone else. The test is not whether your website sounds good. It is whether a customer can describe what you do, unprompted, in one sentence, to a peer. If they cannot, no amount of reach will help, because there is nothing to remember.

Distinctive assets used relentlessly. Consistency of look, name, phrase, and format is the mechanism by which memory attaches to a company rather than to a nice ad. Its cost is discipline, not money. Changing your visual identity annually is the most expensive free thing a startup can do.

Founder and team distribution. At Series A, the founder’s own audience is usually the cheapest reach available, and it is the only channel where credibility is not purchased. It has an obvious ceiling and it does not transfer well to a successor, which is why it is a bridge rather than a destination.

A published point of view. Original opinion, honest teardowns, real numbers from your own operations. This is the horizon 3 work in our payback horizon framework, and it compounds the same way brand does, on a similar clock. It also happens to be the input to AI search visibility, which is increasingly where a category gets first described to a buyer.

Customer proof. Named customers, specific outcomes, and public evidence of a working thing do more to make a startup memorable than an equivalent spend on reach, because in an unfamiliar category buyers are looking for permission more than awareness.

None of that is free. It costs the scarcest resource a small team has, which is production capacity and consistency over quarters. That is a different budget line than media, and it should be argued for on those terms.

The one measurement that keeps this honest

Brand work at small scale fails in a specific way: it becomes unmeasurable, then unaccountable, then unkillable. Two instruments prevent it.

Branded search volume and direct traffic, tracked monthly, reviewed twice a year. Neither is precise. Both move before pipeline does, and both are the fingerprint of memory being built.

An open text “how did you hear about us?” field on your demo form and asked in the first call. It is self-reported, imperfect, and the only place a podcast mention, a conference talk, or a peer recommendation ever shows up. Read the actual sentences. Do not turn them into a pie chart.

What you must not do is run brand work through the same last-touch model as your performance spend. Last touch systematically credits the branded search that brand work created, which makes brand look worthless and branded search look extraordinary. We cover the mechanics in traced vs modeled attribution.

When to start funding brand the way the rule means

Our framework, not a benchmark. Three conditions, and we would want all three before moving real money into paid brand reach:

1. Your capture horizon is saturated. You are buying every high-intent term you can profitably win and the marginal cost per qualified conversation is rising. If capture still has headroom, that is where the next dollar goes. See demand generation vs demand capture.
2. Your positioning has been stable for four quarters. You are no longer paying to build memory of a message you are about to replace.
3. Your budget can buy noticeable share of voice in a definable segment. Not the whole category. A vertical, a geography, a persona. If the honest answer is that your spend disappears into the category’s noise floor, narrow the segment until it does not, or wait.

Until then, the answer to “what is our brand budget?” is not a percentage. It is a list of the unpaid mechanisms above, with names next to them and a cadence.

Frequently asked questions

What is the 60/40 rule in marketing?

It comes from Les Binet and Peter Field’s analysis of the IPA effectiveness databank, popularised in their 2013 work The Long and the Short of It: roughly 60% of budget to long-term brand building and 40% to short-term sales activation. It is a rule about dividing advertising media spend, and its underlying data is case studies submitted to an effectiveness awards scheme, which is the basis of the main criticism against it.

What is the brand vs activation split for B2B?

Sources disagree and rarely say so. The Drum reported Binet and Field’s B2B work as 46% brand and 54% activation, isolating B2B cases in the IPA databank. LinkedIn’s own page on the same research says the proportions are “much closer to 50/50” and calls it early data from a small sample not yet formally published. Treat 46:54 as directional, not as a benchmark.

Should an early-stage startup spend on brand marketing?

Not in the form the rule implies. The 60/40 mechanism runs on share of voice, and at startup spend levels a brand slice of a small budget buys a share of voice indistinguishable from zero. Build the same memory through unpaid mechanisms instead: a repeatable position, consistent distinctive assets, founder distribution, a published point of view, and customer proof. Move to paid reach when capture is saturated, positioning has been stable for a year, and you can buy noticeable presence in a definable segment.

How do you measure brand marketing at an early-stage company?

With leading indicators, not attribution. Branded search volume and direct traffic monthly, plus an open text “how did you hear about us?” field read qualitatively. Never put brand work through the same last-touch model as performance spend, because last touch credits the branded search that brand work produced and will tell you to defund the thing that is working.

The bottom line

The 60/40 rule is a serious finding from a self-selected dataset, and its B2B extension is reported as 46:54 by coverage of the research and as “closer to 50/50, small sample” by the publisher that sponsored it. Byron Sharp’s objection to the whole method is worth reading before you quote either.

None of that is the reason to ignore the ratio at a Series A company. The reason is structural: it is a rule for allocating media at a scale where share of voice exists, and below that scale the brand half has to be bought with consistency, position, distribution, and proof rather than with impressions. Fund those deliberately, measure them with branded search and an open text question, and revisit the ratio when your capture channels are saturated and your positioning has stopped moving.

The trade-off you are actually making is not brand versus performance. It is memory versus response, and at your size, memory is bought with discipline rather than with budget.

See it live: Marqeable’s campaigns draft the content and social work that builds memory without a media budget, AI website chat answers the buyers who arrive because of it, and attribution ties revenue back to the exact message.

Marqeable runs your campaigns, answers every visitor, text, and email in seconds, and turns them into booked jobs and meetings - even at 9pm on a Saturday. We’re in private beta with a small early cohort. Get early access