- 60/40 is a statistical average from 996 IPA campaigns: the mean, not a law. [1]
- The effective range is roughly 40-70% on brand, set by category and maturity. [2]
- Young brands skew to performance: practitioner frameworks put them nearer 30/70 early, moving toward 60/40 once established.
- Brand spend lowers future CAC. Cutting it defers a rising acquisition cost, it doesn't save money. [3]
- Set the split from maturity, runway and cycle length. Then let experiments move it.
There is no more misused number in marketing than 60/40. It gets quoted in pitch decks like scripture, defended in budget meetings like doctrine, and applied (with a straight face) to a three-year-old direct-to-consumer brand with eight months of runway. The number is good. The way it's used is often indefensible, and the gap between those two things has cost more brands more money than any bad channel ever did.
Start with what the number actually is. Binet and Field derived the 60% brand / 40% activation split from analyzing 996 IPA effectiveness campaigns, finding the ratio that maximized combined short- and long-term profit. [1] The finding dates to "The Long and the Short of It," published in November 2013 and still the most influential report in the IPA's effectiveness databank. [5] Read that sentence carefully: it is the average across nearly a thousand campaigns, mostly from established brands in broad-reach categories. It describes the center of a distribution. It was never a claim that every brand, at every stage, in every category, should land on the mean.
The rule is real. The range is wider than the rule
The follow-on research makes the point for us. Later effectiveness work finds an effective range of roughly 40-70% of budget on brand building rather than a single optimal ratio, with the right figure depending on category dynamics and brand maturity. [2] Practitioner guidance in 2026 is blunter still: there is no universal ratio, and generic rules of thumb are a starting point at best. [2][4]
None of this demotes brand building. The evidence that it works (that it compounds, that it does what performance cannot) is some of the sturdiest in the discipline. The correction is narrower and more useful: 60/40 is the right default for the kind of brand it was measured on, and the wrong law for a brand that looks nothing like that sample.
Why the split has to move with maturity
The variable the dogma ignores is time: specifically, whether the brand can afford to wait for brand building to pay back.
As a practitioner rule of thumb: 0-3 years: roughly 30% brand / 70% performance. Prove the model, build conversion volume, survive. 3-7 years: shift steadily toward balance as the brand and its runway stabilize. 7+ years, established: 60/40 or beyond, because now the long-term asset is worth compounding. Small, short-cycle e-commerce can sit performance-heavy longer than the textbook allows.
A startup that puts 60% into brand is often just funding awareness it will not be alive to monetize, which is why startup-stage guidance tends to put young brands nearer 30/70 in favor of performance. The performance skew early isn't a rejection of brand theory. It's an acknowledgment that the payback period of brand investment can exceed the survival horizon of a young company. You earn the right to a 60/40 split by first existing long enough for the long term to matter.
The half of the argument the performance crowd gets wrong
If the brand purists over-apply the rule, the performance purists make the opposite error, treating brand spend as a cost to cut when times tighten. It's the more expensive mistake, because it's invisible on the dashboard that reports it.
Brands that over-invest in short-term performance end up paying more per conversion over time, because brand strength directly determines how efficiently performance media works. [3] The pattern holds at scale: across Analytic Partners' ROI Genome, built from more than 750 brands in 45 countries, brands that cut brand investment see marketing return on investment fall across every metric, and upper-funnel tactics run 60% more effective over the long term than lower-funnel ones while giving up only 25% in the short term. [6] A weak brand makes every click do more lifting, and the CAC creeps up quarter after quarter while the spreadsheet insists everything is efficient. The useful reframe circulating among operators in 2026 is exact: brand spend is a deposit into lowering your future cost of acquisition. [3] Zero it out and you haven't saved the money. You've borrowed it from next year's performance budget, at interest.
Set the number from your situation, then let the data move it
The way out of the dogma isn't a better ratio. It's better questions. Before you defend or attack 60/40, answer three things. How established is the brand? Does it have an asset worth compounding yet? How much runway can absorb a payback measured in years, not weeks? And how long is the purchase cycle in your category: considered and slow, or impulsive and fast?
A safe, established brand in a considered, long-cycle category (the kind of buy that leans on channels like B2B LinkedIn advertising) earns something like 60/40. A cash-constrained three-year-old in short-cycle e-commerce earns closer to 30/70. Both are correct, and neither is a betrayal of the research, because the research was always a distribution, and these are just two different points on it.
Then stop treating the number as fixed. Run the incrementality tests, build the mix model, and let your evidence move the split: toward brand as you stabilize, toward performance when survival demands it. Quoting an average from a study you haven't read, at a brand it was never measured on, isn't discipline. It's the costume of it. The rule is one of the best things marketing knows about itself. Obeying it blindly is one of the worst.
Sources
- deepmarketing · The 60/40 rule: brand vs performance budget (Binet & Field, 996 campaigns)
- Funnel · How to balance brand and performance marketing
- Objective Platform · Brand or performance: where should you invest more?
- Rocket Agency · What percentage of marketing budget should go to brand building?
- IPA · Les Binet & Peter Field: effectiveness researchthe industry body's hub for the 60:40 work; dates "The Long and the Short of It" to November 2013
- Analytic Partners · Brand marketing drives sales, ROI and even performance campaignsROI Genome analysis of 750+ brands in 45 countries; vendor research, directionally consistent with the IPA work
Frequently asked questions
Where does the 60/40 rule actually come from?
From Les Binet and Peter Field's analysis of 996 IPA effectiveness campaigns, which found that a roughly 60% brand / 40% activation split maximized combined short- and long-term profit. Crucially, it's described as a statistical average across many established brands in broad-reach categories: the mean of a distribution, not a prescription for any single brand. Treating it as a fixed law misreads what the research says about itself.
Is 60/40 wrong, then?
No. It's a strong default for the brands it was derived from: established, broad-reach, not fighting for survival. The error is universal application. Later work finds an effective range of roughly 40-70% on brand depending on category and maturity, and practitioner guidance increasingly says there's no single magic ratio. Use 60/40 as a planning anchor you then adjust, not a number you obey.
What split should a young or small brand use?
Skew to performance early. Startup-stage frameworks suggest around 30% brand / 70% performance for brands aged roughly 0-3 years, moving toward 60-70% brand only once established (7+ years). Small businesses often start near 70/30 weighted toward whichever type fits the model. Short-cycle e-commerce leans performance. The logic is runway: brand building pays back over years you may not have yet.
Why does brand spend make performance cheaper?
Because brand strength changes how efficiently performance media converts. Brands that over-invest in short-term activation tend to pay more per conversion over time, as weak brand demand makes every click work harder. Framed usefully: brand spend is a deposit that lowers your future cost of acquisition. Cut it entirely and you don't save money. You defer a rising CAC into next year.
How do I actually set the split?
Start from three questions, not a ratio. How established is the brand? How much cash runway can absorb a slow payback? How long is the purchase cycle in your category? A safe, established brand in a considered category earns something like 60/40. A cash-constrained three-year-old in short-cycle e-commerce earns closer to 30/70. Then let incrementality tests and mix modeling move the number, rather than defending the textbook figure.
What is the most common mistake marketers make with the 60/40 rule?
Treating a population average as a prescription for one brand. The 60% brand / 40% activation figure is the mean across 996 IPA campaigns, mostly established brands in broad-reach categories, so applying it unchanged to a young, cash-constrained, or short-cycle brand ignores everything that made the average what it is. A mean describes a distribution, not any single member of it.
When should a brand shift more budget toward brand building?
As it stabilizes and earns a long-term asset worth compounding. The practitioner rule of thumb moves brands from roughly 30/70 in the first 0-3 years toward balance across 3-7 years, reaching 60/40 or beyond once established at seven or more years. The trigger is having enough runway to wait out a payback measured in years rather than weeks.
Does the 60/40 rule apply to short-cycle e-commerce?
Less than the textbook implies. Short-cycle e-commerce, where purchases are fast and impulsive rather than considered, can sit performance-heavy for longer than the rule allows. A cash-constrained brand in that category earns something closer to 30/70, because the long payback of brand building is harder to justify when both the runway and the purchase cycle are short.
Does brand building actually compound, or is that just theory?
The evidence that brand building works, that it compounds and does what performance cannot, is some of the sturdiest in the discipline. That's exactly why the correction here is narrow. 60/40 is the right default for the established, broad-reach brands it was measured on, not a rejection of brand investment. The problem is applying that mean to a brand the sample never contained, not the value of brand building itself.
Should I use incrementality tests or mix modeling to set the split?
Use them to move the split, not to set it cold. Start from three questions, brand maturity, cash runway, and purchase-cycle length, then let incrementality tests and marketing mix modeling push the number toward brand as you stabilize or toward performance when survival demands it. The point is to let your own evidence override the textbook figure rather than defend a population average at a brand it was never measured on.
Can the right split ever fall outside the 40-70% range?
Yes, at the edges. The roughly 40-70% effective range describes the established, broad-reach brands the research covered. A cash-constrained brand in short-cycle e-commerce can sit more performance-heavy than 40% brand for longer than the textbook allows, which is why startup-stage frameworks put young brands nearer 30/70. The range is a strong default for the middle of the market, not a fence every brand has to stay inside.
Is 60/40 saying brand matters more than performance?
No. The 60% figure is a budget split that maximized combined short- and long-term profit across 996 campaigns, not a ranking of which type matters more. Read as importance rather than allocation, it gets misused in both directions: brand purists treat it as a mandate, and performance purists treat brand as the first line to cut. Both misread an average as an argument.
Mikołaj Salecki
Writes about media, tech, and AI business for people who actually run digital. Former agency lead. Skeptic of frameworks that read better than they perform.
Get the next issue by email.
One letter, once a week. Sharp coverage of media, tech, and AI business. No filler.
