What does the 95-5 rule mean for a video budget?

It means a video budget aimed entirely at the in-market 5% is aimed at a twentieth of the audience it reaches. A workable split funds two distinct jobs: assets that register with future buyers and build memory, and assets that convert the buyers who are in market now. The two require different briefs, different placements and different success measures, and the commonest budgeting error is funding one and measuring it as if it were the other.

Is the 95-5 rule accurate for every B2B category?

The 5% figure is an average derived from purchase-cycle length, not a constant. A category where contracts renew annually has a larger in-market share at any moment than one where the buying cycle runs five years. You can estimate your own: if the average customer buys once every four years, roughly a quarter of the market is in a buying cycle in any given year, and a fraction of that is actively deciding this quarter.

Does the 95-5 rule mean lead generation does not work?

No. It means lead generation is fishing in 5% of the pond, which is a perfectly good thing to do and a poor thing to do exclusively. Assets aimed at the in-market 5% have the shortest payback and the clearest attribution, so they should be funded first. The rule is an argument against spending the entire budget there, not an argument against spending any of it there.

How do you measure video aimed at buyers who are out of market?

Not with conversion metrics, which is why this work gets cut. Use leading indicators that move on the same timescale as memory: branded search volume, direct traffic to named asset pages, share of buyers who mention you unprompted in first calls, and the proportion of new opportunities that arrive already knowing what you do. None is precise. All of them are more honest than attributing a five-year memory effect to a last-touch model.

What split between brand and demand video should a B2B team use?

There is no universal number, and any page that gives you one is guessing at your category. A more useful rule is directional: fund the in-market assets first because they pay back fastest and are easiest to defend, then check whether anything at all in the remaining budget is built to be remembered by someone who cannot buy for two years. In most B2B video libraries the honest answer to that second question is nothing, and that is the imbalance worth correcting.

Field guide

The 95-5 rule and your video budget

Ninety-five per cent of your audience cannot buy today.

By Paul Joseph · Updated 23 Aug 2026

The 95-5 rule holds that at any moment roughly 95% of business buyers are out of market and about 5% are actively buying. It comes from Professor John Dawes of the Ehrenberg-Bass Institute, published by the LinkedIn B2B Institute in 2021. For a video budget the consequence is direct: a library built entirely to convert demand is built for a twentieth of the people who will see it, and the other nineteen-twentieths are reached anyway — at full cost, with nothing designed for them.

Before rebalancing anything, it is worth knowing whether your existing assets can be traced to pipeline at all. The Dark-Funnel Video Audit scores that in two minutes, free — a split argument is academic if neither half is instrumented.

Where the number comes from

The rule is an arithmetic consequence of purchase-cycle length rather than a survey finding. If businesses in a category replace a system roughly every five years, then in any given quarter only a small fraction of them are in a buying process — and Dawes's work put that fraction at about 5% for the categories examined. Everyone else is a customer of somebody, mid-contract, or not thinking about the category at all.

That derivation is also the caveat. The 5% is an average, not a constant, and it moves with your category's cycle length. Annual renewals put far more of the market in play at any moment than a five-year capital purchase does. Estimate your own before you plan against the headline: divide one by your average years between purchases, and that is roughly the share of the market in a buying cycle in a given year. A fraction of that is deciding this quarter.

The precise figure matters less than the direction, which no plausible number changes. Even a generous reading — 15% in market rather than 5% — leaves most of your reach going to people who cannot act on it. The rule is not a target. It is a description of who is on the other side of the screen.

Two jobs, and why one budget cannot do both

The useful implication is not “spend more on brand”. It is that the two audiences need different assets, and that an asset built for one will underperform against the other's success measure no matter how well it is made.

How assets for in-market and out-of-market buyers differ
	The 5% in market	The 95% not yet
The job	Be chosen	Be remembered
Typical asset	Demo, customer story, product explainer, pricing walkthrough	Category film, founder point of view, documentary short, a show
Placement	A step in your funnel you already send traffic to	Wherever attention is, mostly outside your properties
Measured on	Conversion at that step, cycle time, pipeline touched	Branded search, unprompted mentions, direct traffic, reach against the right audience
Payback	Weeks to a quarter	Years, and never cleanly attributable

Read the bottom two rows together and the political problem is obvious. The right-hand column is measured on indicators that no dashboard treats as revenue, and pays back on a timescale longer than most marketing tenures. That is why it gets cut first in a difficult quarter, and why the cut is invisible for eighteen months and then shows up as rising cost per lead that nobody connects back to it.

A worked example: what the imbalance costs

Illustrative arithmetic, not a benchmark. Round numbers, so the mechanism stays visible.

A B2B company has 4,000 target accounts and a four-year replacement cycle, so roughly 1,000 accounts enter a buying process each year — about 250 in any quarter, or 6% of the base. That is their own 95-5 number, and it is close enough to the headline to use.

Their annual video budget is $120,000, and every asset in it — six customer stories, two product explainers, a demo — is built for a buyer already in an evaluation. The assets are good and the in-market work is effective: they win 40 of the 250 quarterly opportunities they are aware of.

The problem is the word “aware”. They are only invited into a shortlist by accounts that already know they exist. Of the 250 accounts entering a buying cycle each quarter, they are invited into 90. The other 160 run a process, draw up a shortlist from the three vendors that came to mind, and never make contact.

Quarterly opportunity flow with and without out-of-market video work
Per quarter	Today	With 25% of budget on memory
Accounts entering a buying cycle	250	250
Shortlists they are invited into	90	110
Win rate on shortlists	44%	44%
Deals won	40	48

Moving $30,000 of the $120,000 to work aimed at the 95% does not improve the conversion assets at all — the win rate is unchanged by construction. It changes how many contests they are invited into, which is the variable the in-market budget cannot touch however well it is spent. Twenty percentage points more shortlist entry is eight more deals a quarter, at an average deal size of $60,000.

Two honest caveats. The shortlist lift is an assumption, not a measurement — it is the thing you would be testing, and it would take two years to observe. And the reallocation costs eight deals' worth of conversion capacity in the short run if the in-market assets were already saturating demand, which is why the sequencing below matters more than the ratio.

What the arithmetic does establish is where the leverage sits. When conversion assets are working, adding a seventh customer story competes with six existing ones for the same 90 shortlists. Nothing in that budget can change the 160 contests you were never invited to.

How to split it without a leap of faith

A percentage split is still a percentage, and how to set a B2B video marketing budget makes the case against sizing from one at all. Published brand-versus-demand ratios exist and are widely quoted. They are averages across categories with wildly different cycle lengths, and adopting one wholesale is how a plan acquires a number nobody in the room can defend. A sequence is more useful than a ratio.

1. Fund the in-market assets first. They pay back fastest, attribute most cleanly, and every argument for the other half is easier to make from a position where the conversion work is visibly working. Sequencing matters: this is not the half to experiment with.
2. Ask what in the library is built to be remembered. Not repurposed — built. In most B2B video libraries the answer is nothing, and discovering that is more useful than any percentage.
3. Fund one out-of-market asset properly rather than four cheaply. Memory structures are built by distinctiveness and repetition. Four forgettable films produce four times nothing; one that is genuinely worth watching, in service for three years, has a chance.
4. Agree its indicators before it ships. Branded search, unprompted mentions in first calls, direct traffic to its page. Agreed in advance, they are a measurement. Chosen afterwards, they are an excuse, and everyone in the room knows the difference.
5. Instrument both halves the same way. Out-of-market does not mean untracked. The asset still needs a page you control, a next step and a contact record — see video attribution for B2B.

Step five is where most 95-5 arguments quietly collapse. The rule is regularly used to excuse work nobody intends to measure, which is the fastest way to lose the budget for it permanently. Long payback is not the same as no evidence, and treating them as equivalent hands the sceptics their case.

What this changes in a brief

The practical consequence lands one level down, in how the asset is commissioned. The 60-Second Brief opens with the funnel stage, and “the 95% who are not in a buying cycle” is a legitimate answer to that question — but only if the following four decisions are answered consistently with it.

Consistently means: the viewer action is proportionate to someone who cannot buy, so a subscription or a follow rather than a demo request. The trigger adds them to something they will still be receiving in two years. The contact record captures an account-level signal rather than a lead status the sales team will chase and be annoyed by. And the pipeline outcome is expressed as an indicator with a horizon attached, not as a quarterly conversion number the asset was never designed to move.

Get those four wrong and the asset fails on paper regardless of what it achieved — which is the most common way out-of-market video is defunded. It is not usually judged and found wanting. It is judged against the wrong measure and never given a second commission. The dark funnel is where most of the evidence for it ends up, and the CAC case sets out how the slow mechanism eventually shows up in the acquisition number.

The 95-5 rule, answered

What is the 95-5 rule in B2B marketing?: The 95-5 rule holds that at any given moment roughly 95% of business buyers are out of market and only about 5% are actively buying. It comes from Professor John Dawes of the Ehrenberg-Bass Institute, published by the LinkedIn B2B Institute in 2021. The implication is that most marketing reaches people who cannot act on it today, so the job of that marketing is to be remembered later rather than to convert now.
What does the 95-5 rule mean for a video budget?: It means a video budget aimed entirely at the in-market 5% is aimed at a twentieth of the audience it reaches. A workable split funds two distinct jobs: assets that register with future buyers and build memory, and assets that convert the buyers who are in market now. The two require different briefs, different placements and different success measures, and the commonest budgeting error is funding one and measuring it as if it were the other.
Is the 95-5 rule accurate for every B2B category?: The 5% figure is an average derived from purchase-cycle length, not a constant. A category where contracts renew annually has a larger in-market share at any moment than one where the buying cycle runs five years. You can estimate your own: if the average customer buys once every four years, roughly a quarter of the market is in a buying cycle in any given year, and a fraction of that is actively deciding this quarter.
Does the 95-5 rule mean lead generation does not work?: No. It means lead generation is fishing in 5% of the pond, which is a perfectly good thing to do and a poor thing to do exclusively. Assets aimed at the in-market 5% have the shortest payback and the clearest attribution, so they should be funded first. The rule is an argument against spending the entire budget there, not an argument against spending any of it there.
How do you measure video aimed at buyers who are out of market?: Not with conversion metrics, which is why this work gets cut. Use leading indicators that move on the same timescale as memory: branded search volume, direct traffic to named asset pages, share of buyers who mention you unprompted in first calls, and the proportion of new opportunities that arrive already knowing what you do. None is precise. All of them are more honest than attributing a five-year memory effect to a last-touch model.
What split between brand and demand video should a B2B team use?: There is no universal number, and any page that gives you one is guessing at your category. A more useful rule is directional: fund the in-market assets first because they pay back fastest and are easiest to defend, then check whether anything at all in the remaining budget is built to be remembered by someone who cannot buy for two years. In most B2B video libraries the honest answer to that second question is nothing, and that is the imbalance worth correcting.

Related: the Frame to Funnel Method is the framework this sits inside, the glossary defines the 95-5 rule alongside every other term the show uses, and a B2B video strategy that is not working covers the three failures that are not this one.