Demand generation

What is demand generation?

Demand generation is the whole job of making a market want what you sell, then being there when it decides to buy. It splits into two activities that share a name and share almost nothing else.

Demand capture goes after people who are already looking. Paid search on category terms, comparison pages, review site presence, retargeting, inbound demo requests. The intent already exists. You are competing for it.

Demand creation goes after people who are not looking yet. Category education, original research, events, podcasts, out-of-home advertising, anything that puts you in someone's head before they have a problem to solve. You are not competing for intent. You are manufacturing the memory that gets recalled when intent shows up later.

Here is the number that reframes the entire discipline. Professor John Dawes at the Ehrenberg-Bass Institute, in research for the LinkedIn B2B Institute, found that companies change providers for things like software, banking, and telecoms roughly once every five years. That means about 20% of a market is in-market in a given year, and only around 5% in a given quarter. The other 95% are not buying. They already have a vendor, they are under contract, or the problem is not urgent yet.

So when a team says demand generation is not working, what they usually mean is that they put the entire budget into capture, competed for the same 5% every competitor is bidding on, watched cost per opportunity climb every quarter, and concluded the channel was broken. The channel was not broken. The addressable pool was 5% the size they thought it was.

The uncomfortable corollary, and Ehrenberg-Bass is direct about this: you cannot persuade someone into the market. A buyer who signed a three-year contract last month is not going to be argued out of it by a better nurture sequence. They move themselves in-market when their circumstances change. Your only real job with the 95% is to be the name they remember when that happens. As Jenni Romaniuk puts it, you cannot push buyers down a funnel, but you can catch them as they fall.

How to run demand generation without screwing it up

• Label every line item creation or capture. Do this before anything else. Put your actual spend into two columns and look at the ratio. Most teams discover they have 90% or more in capture and have been calling the whole thing demand gen for years. You cannot fix a balance you have never looked at.
• Stop measuring creation with capture metrics. If you judge a brand campaign on last-touch attributed pipeline, it will lose, every time, by design. That is not evidence it failed. That is evidence you used the wrong instrument. Creation gets measured on branded search volume, direct traffic, share of voice, and whether people can name you unprompted.
• Run a geo holdout when you can. Buy one metro, deliberately skip a comparable one, compare branded search and inbound over the following two quarters. It is the only creation measurement that survives a CFO conversation, and it needs setting up before launch, not after.
• Own the category question, not just your product name. Capture spend on your own brand terms is often just paying a toll on demand you already created. The valuable real estate is the question your buyer asks before they know vendors exist.
• Be specific enough to be remembered. Generic is the actual enemy. If your creative could have your competitor's logo dropped on it without anyone noticing, you are buying impressions and building nothing. Segment ran billboards in San Francisco that said "Good morning, LA!" People who had no idea what Segment sold posted photos of it. The confusion was the distribution.
• Give it four quarters and say so out loud. Agree the timeline with your CEO before you spend, because the graph does not move in month two. A creation program killed at month five was killed before its first cohort could possibly have entered the market.
• Do not abandon capture. The 95:5 argument is not permission to stop harvesting demand. That 5% is real, in-market, and buying from somebody this quarter. Run both. Just stop pretending one is the other.

When is demand generation a bad idea?

Your total market is 200 companies. Broad reach is wasteful when you could get every buyer in a room. Go run account-based marketing instead and spend the money on 40 people who matter rather than a category nobody outside your list cares about.

You do not have product-market fit. Generating demand for something that does not retain is an expensive way to accelerate churn and burn your name in a market you will need later. Fix the product first.

Sales cannot handle the volume. Creating demand you cannot service produces slow follow-up, bad experiences, and a pipeline of annoyed people. This is a real failure mode at seed stage: marketing hits the number, sales has two reps, and half the demand rots in a queue.

You need bookings in ninety days. Then you need capture, and you should say that plainly rather than dressing it up. Buy the in-market 5%, accept the cost per opportunity, and be honest that you are harvesting rather than building.

Nobody will tolerate unattributable spend. If every dollar must trace to a closed-won deal, creation will lose every budget fight it enters. Have that argument up front, not in month four when the campaign is already running.

Examples of demand generation in the wild

Creation, high-budget. Billboards on the 101, airport takeovers, and transit buys across an entire metro. Nobody scans a QR code at 65 mph, and that is fine, because the job is being the name someone recalls in eight months. See out-of-home advertising for what that costs and how to measure it.

Creation, no budget. Chili Piper built short Spotify playlists for prospects, screenshotted them, and sent the image. Costs nothing but attention. Juicebox reprinted Guess Who with the faces of top recruiting voices, mailed it to those people, and watched them post it to their own audiences for free. Creation does not require a media budget. It requires being worth talking about.

Capture, done properly. Category comparison pages, review site presence, and paid search on the problem rather than the product. This is where the in-market 5% actually is, and there is nothing wrong with meeting them there. Just know the ceiling.

The handoff. Ergo walked 500 tubs of branded pre-workout into 50-plus SF startups by hand, and about half booked a demo with no sequence behind it. Creation and capture in the same afternoon, which is what happens when the creative act is itself the call to action.

Sources

1. Ehrenberg-Bass Institute for Marketing Science. "95% of B2B buyers are not in the market for your products." 2025. https://marketingscience.info/news-and-insights/ehrenberg-bass-95-of-b2b-buyers-are-not-in-the-market-for-your-products
2. Ehrenberg-Bass Institute for Marketing Science. "The 95:5 rule is the new 60:40 rule." 2025. https://marketingscience.info/news-and-insights/the-955-rule-is-the-new-6040-rule
3. Dawes, John. "Advertising effectiveness and the 95-5 rule," research for the LinkedIn B2B Institute. 2021.
4. uglyGTM. "8 Unconventional Account Based Marketing Real Examples." 2026. https://www.uglygtm.com/blog/account-based-marketing-examples