Demand generation vs demand capture

Two fundamentally different jobs that get lumped together as “marketing”:

• Demand generation - making people want a thing they didn’t previously want, or surfacing a problem they didn’t know they had. The work happens before the buyer even knows there’s a category to shop in.
• Demand capture - getting in front of buyers who are already actively shopping for a solution. The work happens after the buyer has decided to buy something in the category.

Capture is easier to measure (clearer intent signals, shorter time to conversion) so most performance-led marketing teams over-invest in it. Capture also has natural limits. There are only so many people in-market at any time, and if you don’t generate new demand, capture revenue eventually plateaus.

How the channels split

• Generation channels: paid social to cold audiences, content marketing, PR, podcasts, sponsorships, brand campaigns, influencer content (the educational kind, not affiliate links).
• Capture channels: paid search on category keywords, brand search, comparison sites, retargeting, abandoned-cart email, affiliate links targeting buyer-intent searches.

Some channels do both. Paid social to lookalike audiences of existing customers is closer to capture. To cold interest-based audiences is closer to generation. The distinction is about the audience’s state more than the channel itself.

Why this matters for CRO

CRO almost always runs against capture-stage traffic. Hot traffic landing on a PDP, comparison-shoppers on a pricing page, abandoned carts coming back via email. That’s where conversion rates are highest and tests run fastest, so it’s where CRO budgets concentrate.

The problem is that capture-stage tests can’t grow the overall market for you. If demand generation is weak, your capture channels eventually run dry no matter how well-optimised the pages are. CRO that obsesses over the PDP without ever asking “are we generating enough demand to feed this funnel?” is rearranging deck chairs.

There’s a direct mapping to the brand-vs-performance split most marketing teams talk about. Brand work is mostly demand generation. Performance work is mostly demand capture. Two framings of the same axis.

Where CRO can help generation

A few specific spots:

• Top-of-funnel conversion - optimising email opt-in rates on blog content, ebook downloads, newsletter sign-ups. These are generation-stage conversions and they’re testable.
• Awareness-aware copy - testing copy variations on landing pages for cold traffic where the goal isn’t “buy now” but “learn enough to come back warmer”.
• Funnel architecture - testing whether a sequence (cold ad → educational article → retargeting → product page) outperforms a direct route (cold ad → product page).

Measuring generation

The usual reason generation gets underfunded isn’t that anyone disputes it works. It’s that it doesn’t produce a number in the platform, so it loses every budget argument to a channel that does.

The mistake inside that is holding generation to an in-platform ROAS target at all. Generation pays off downstream, in capture metrics, weeks or months later - a cold-audience campaign judged on its own last-click return will always look like a failure, and cutting it will look like a good decision right up until brand search starts declining.

It’s harder to measure, not impossible. Brand-search volume, direct-traffic growth and assisted conversions all move when generation is working, and they’re the proxies worth tracking. The clean answer, where you have the scale for it, is an incrementality test - turn the channel off in matched regions and watch what happens to the capture channels. That’s also the only version of this argument that survives contact with a sceptical finance team.