Demand Generation or Lead Generation: Which Should You Fund?

Demand Generation or Lead Generation: Which Should You Fund?

Both, in a ratio most B2B teams get backwards. The available buyer research points the same direction: the decision that matters is made long before anyone fills in your form, and most budgets are still concentrated on the form. That is the argument, and the data behind it is unusually clear.

We build the sites and content engines this argument gets fought over, so we watch it play out on real budgets. The version we hear most often is a marketing lead defending pipeline numbers to a board that only counts leads.

Here is what the research supports, where each side is genuinely right, and how we would split the money.

What Is the Actual Difference?

Lead generation captures existing intent. Demand generation creates the preference that intent later expresses. One is a net. The other is the reason fish swim towards it. They are usually presented as rivals and they are actually sequential.

The confusion is understandable, because both produce activity you can see in a dashboard. A gated report produces a list. A podcast appearance produces nothing you can attribute. That asymmetry, not the underlying effectiveness, is why the budget skews.

Here is the honest split of what each one optimises.

	Lead generation	Demand generation
Optimises for	Contact details now	Preference later
Feedback speed	Days	Quarters
Measurable	Directly	Indirectly
Audience	People ready to buy	Everyone who will be
Fails by	Capturing people who never buy	Never being counted

How Many of Your Buyers Are Even in the Market?

Far fewer than your campaign plan assumes. John Dawes of the Ehrenberg-Bass Institute set out the 95:5 rule, arguing that up to 95 percent of people or firms are not in the market for many goods and services at any one time. He published it through LinkedIn in 2021 and it was co-published in Marketing Week.

The reasoning is simple arithmetic rather than theory. Dawes notes that corporations change major service providers around once every five years on average, which puts roughly 20 percent in market in any given year and a much smaller share in any given quarter.

Dawes is careful about the number and we should be too. He states plainly that it is not meant to be a precise rule, and that it varies by category. His own example: if the average time between purchases is two years, roughly 50 percent of buyers are in market annually and about 13 percent quarterly. Work out your own category rather than borrowing the headline.

When Do Buyers Actually Talk to You?

Later than you would like, though the gap is narrowing. The 6sense B2B Buyer Experience Report for 2025 found that the point of first contact with a seller sits at 61 percent of the way through the buying journey, moved forward from 69 percent the year before.

That report is a substantial sample. It surveyed nearly 4,000 respondents, with a further 766 in a companion study, weighted towards senior people: 49 percent at VP level or higher, and roughly half of respondents acting as ultimate decision makers. Median purchase costs ran between 200,000 and 400,000 dollars.

The cycle is compressing too. 6sense reports the average buying cycle shortened from 11.3 months in 2024 to 10.1 months in 2025, and describes the earlier first contact as arriving roughly six to seven weeks sooner. Buyers are moving faster, but they are still doing most of the journey before you hear from them.

Does the Shortlist Form Before You Know They Exist?

Yes, and this is the finding that should settle the budget argument. 6sense reports that ninety five percent of the time, the winning vendor is already on the day one shortlist, and that four out of five deals are still won by the vendor the buyer preferred before engaging any seller.

Read that carefully. It is not saying that sales does not matter. It is saying that by the time a lead exists, the outcome is mostly decided, and it was decided by whether you were known and liked before the process started. That is demand generation's entire job.

The buying group makes this harder. 6sense found groups averaging more than ten members on deals averaging 250,000 dollars, and buyers evaluating 5.1 vendors against 4.5 previously. You are not building preference in one person's head. You are building it across a committee, most of whom you will never meet.

So Is Lead Generation Pointless?

No, and the same report shows why. 6sense found buyers averaged 16 interactions per person with the winning vendor, statistically similar to previous years even as first contact moved earlier. Those interactions have to happen somewhere, and a lot of them are the assets lead generation produces.

The validation phase after first contact is real work. Buyers need the security documentation, the reference call, the pricing conversation and the implementation detail. A team with no lead capture and no sales process cannot serve any of that, and the day one favourite loses anyway.

The right conclusion is not to defund lead generation. It is to stop expecting it to create demand. A form converts intent that already exists. Point it at people who are ready, make it painless, and stop measuring your brand work by how many forms it filled.

What Does Lead Gen Get Wrong Most Often?

It gates the wrong things. Putting your best explanation of the problem behind a form means the 95 percent who are not ready never read it, which is precisely the audience you needed to reach. You have traded the memory you were trying to build for a contact record you will not use.

It also optimises the metric rather than the outcome. A team measured on MQL volume will find volume, and the quality drops quietly until sales stops working the list. We wrote about the specific version of this trap in our piece on gated content in B2B SaaS.

The third failure is treating a downloaded report as buying intent. Given the 6sense finding that most of the journey happens before contact, someone who downloads a guide is usually early, and calling them within an hour trains them to avoid your forms.

What Does Demand Gen Get Wrong Most Often?

It hides behind unmeasurability. Because brand work is hard to attribute, it becomes a convenient home for spending that nobody has to defend, and quality slips. Dawes' argument is that advertising builds memory links, not that any activity labelled brand automatically does.

It also drifts off the buyer. Demand generation that entertains a general audience while your actual buying committee never sees it is expensive reach with no commercial claim on the budget. Reach has to be reach among people who will one day buy.

And it often skips positioning. You cannot build memory for a claim you have not made clearly. If your own team cannot state what you do and who it is for in a sentence, more reach just distributes the confusion. That is the prerequisite we cover in our guide to B2B SaaS positioning.

How Should You Split the Budget?

Start from the market maths rather than from last year's plan. If most of your category is out of market at any moment, and the winner is usually the day one favourite, then the majority of spend belongs to being known by people who are not ready yet.

We will not hand you a fixed ratio, because Dawes' own caveat about category variation applies directly to this decision. A category with a two year replacement cycle has a very different in market share than one with a five year cycle, and the split should follow.

What we will say is that the capture side has a natural ceiling. Once your forms are frictionless and your comparison pages exist, further investment there converts the same finite pool slightly better. The demand side has no such ceiling, and it is what makes the pool bigger next year.

What Would We Do First?

Work out your own in market percentage. Take your average customer's replacement cycle, invert it, and you have a defensible estimate of how many of your target accounts could buy this year. That single number reframes every budget conversation, and it is more persuasive than any framework.

Then audit what is gated. Move anything that explains the problem in front of the form, and keep behind it only what a serious evaluator wants: the detailed comparison, the security pack, the implementation guide. Explanation is demand generation. Evaluation material is lead generation.

Finally, make sure both are pointed at the right people. Neither motion survives a vague target, which is why we always start with the buyer definition, as covered in our guide to defining your ICP.

If you want help working out which side of this your budget is stuck on, we are happy to talk it through. Find us at phoenix.studio.