Direct answer
The best-evidenced starting point for B2B is roughly 46% brand, 54% activation — Binet and Field's IPA Databank analysis for the LinkedIn B2B Institute. Most companies run nowhere near it: Duke's CMO Survey found marketers' ideal split 50/50 and their actual 31/69 toward short-term. The wrong split doesn't fail loudly. It shows up as lead generation getting dearer every year because nobody knew you before the ad. Set it from sales cycle and buyer count, not from what reports fastest.
The belief in most mid-market boardrooms is that brand is what large companies do with money they can afford to waste, and lead generation is what serious businesses do with money they need back. It is a sensible-sounding belief. It is also the reason the lead-generation line keeps getting more expensive, and the reason nobody can explain why.
What the evidence says
The most cited work on this is Les Binet and Peter Field's analysis of the IPA Databank — roughly a thousand effectiveness case studies — which put the optimal split for consumer brands near 60% brand and 40% activation. For the LinkedIn B2B Institute they re-ran the analysis on B2B cases and published "The 5 Principles of Growth in B2B Marketing": B2B businesses grew best at around 46% brand and 54% activation. Brand investment took longer to pay back and paid back for longer, and the memories it built made the activation that followed significantly more effective. Companies whose share of voice exceeded their share of market tended to grow.
Underneath that sits a simpler finding from Professor John Dawes at the Ehrenberg-Bass Institute, published through the same B2B Institute in 2021: at any given time, only around 5% of the buyers in a B2B category are in the market to buy. The other 95% are not looking. Lead generation, by definition, reaches the 5%. Brand is the only thing addressing the 95% who will be in the market next year, when your ad will be one of five they see.
And here is what companies actually do. Duke University's CMO Survey, Fall 2024, 260 marketing leaders: their ideal split was 50% long-term brand building and 50% short-term performance. Their actual split was 31.2% to 68.8%. The March 2026 edition of the same survey found that when pressure from the CEO, board or CFO rises, 70.6% of marketers shift further toward short-term impact. The gap between what they believe and what they fund is not ignorance. It is reporting.
Lead generation is the cost of being chosen. Brand is the reason you were on the list.
Why the split drifts to lead generation
Because a lead-generation campaign produces a number by Friday, and a brand campaign produces an argument about attribution. In a company closing thirty or forty deals a year, that asymmetry is decisive: the lead-gen dashboard shows cost per lead, the brand line shows nothing measurable for months, and the budget meeting rewards the line that can defend itself. I've set out elsewhere why, at low deal volumes, the arithmetic for proving any channel's effect runs to years — see how to measure marketing when you close few deals. The point here is what that does to the split: it hands the decision to whichever side reports fastest, which is not a strategy. It is an accident with a spreadsheet.
The drift then compounds. Cut brand, and next year's in-market buyers have never heard of you, so every lead has to be bought cold. Cost per lead rises. The rise is read as a lead-gen problem and answered with more lead-gen budget, taken from the brand line. The company ends up at 20/80 explaining to the board why leads cost more every year — the pattern behind why marketing reports look better than the results.
| Brand | Activation (lead generation) | |
|---|---|---|
| Who it reaches | The ~95% not in the market yet (Ehrenberg-Bass / B2B Institute) | The ~5% in the market now |
| When it pays | Slowly, then for a long time (Binet and Field) | Quickly, then it stops when the spend stops |
| How it's measured | Badly, by default — share of voice, unprompted inbound, branded search | Easily — cost per lead, by Friday |
| What happens if you cut it | Nothing this quarter; cost per lead rises every year after | Pipeline drops next month |
| Who argues for it in the budget meeting | Usually nobody | The agency, the platform, the dashboard |
The cost of the wrong split
It is not paid once. It is paid in installments, the same way most upstream mistakes are: a cost per lead that rises a little every year, a sales team that opens every conversation from zero, a pipeline that never contains anyone who came looking for you by name. None of those appears as "brand underinvestment" in any report. They appear as marketing getting less efficient, which is then solved by buying more of the thing that made it less efficient.
From the book: an executive programme was losing the perception battle to a competitor the market saw as the superior option. Rankings and salary statistics — the rational, activation-style arguments — barely moved it. The campaign that worked showed a man in a dark suit on the edge of a bed holding a baby, under a line about knowing that students have a life to take care of. It was a brand message. It also produced the highest click-through and conversion rates the programme had ever seen. The split is a false choice once the position is decided: the same execution did both jobs, because it said something the competitor could not.
How to set your own split
Start from three facts about your business, not from a benchmark. How long is the sales cycle? The longer it is, the larger the share of buyers who are out of market at any moment, and the more of your budget has to reach them before they start looking. How many buyers are there? Thirty deals a year from a pool of two thousand companies is a brand problem with an activation tail; three thousand small transactions is closer to the reverse. And what is your share of voice against your share of market? If competitors are louder than their size, the brand line is where you're losing, whatever the lead-gen dashboard says.
Then decide the split, write it down, and protect it for twelve months. Measure brand with the measures it has — unprompted inbound, branded search, the number of first meetings where the buyer already knew what you stood for — and stop asking it to justify itself on cost per lead. None of this works, of course, unless there is a position to build memory of; a brand budget behind a message any competitor could sign is just slower waste. That decision comes first, and it is what Reports, Not Revenue is about.
- Write down your current split. If nobody knows it, that is your first finding.
- Compare it with 46/54. A gap is not a failure; an unexplained gap is.
- Check the trend in cost per lead over three years. A steady rise with a shrinking brand line is the drift, in numbers.
- Set the split from sales-cycle length, buyer count and share of voice — then protect it for a year.
- Confirm there is a decided position to build memory of. If not, fix that before funding either line.
If your split was set by whichever line reports fastest, and nobody can say what the brand budget is supposed to make buyers remember, that is what a Commercial Immersion Diagnostic settles first: €2,900 fixed, no retainer, and if it doesn't surface something concrete and actionable, you don't pay for it.
Cut brand and nothing happens this quarter. Then every lead you buy for the next five years is a stranger.
Book an introduction call See what's included