There’s one test every marketing activity in your company gets judged by. Can you draw a line from it to a deal.
It’s a fair test. It’s also a test that one entire half of the job cannot pass, not because that half is failing but because it is not the kind of thing that leaves a line.
And it’s the half that determines whether the other half has anything to work with.
Here’s the distinction the industry buried under a job title. Demand creation makes somebody want a category they weren’t shopping for. Demand capture harvests people who are already shopping. Almost everything sold as demand generation is capture: brand search, high-intent keywords, review sites, outbound into active cycles, gated content aimed at people who typed the problem into Google this week.
Capture is real work and it’s necessary. It’s just not generation, and calling it that has let a lot of companies believe they’re building demand when they’re queueing for it.
The numbers that make this urgent rather than academic
The Ehrenberg-Bass research everyone half-quotes says roughly 5% of B2B buyers are in market at any given moment. The other 95% are not buying anything from anyone today.
Now Forrester’s buyer work, which is where it gets uncomfortable. Around 92% of buyers enter the formal purchasing process with at least one vendor already in mind. About 41% have a single vendor in mind. Forrester puts the pre-contact portion of the journey at 70% to 80%.
And the room is crowded: an average of around 13 internal stakeholders and nine external participants influencing a decision.
Sit with the 41% for a second. Four buyers in ten have effectively decided before the process that you can measure has begun. Your beautifully optimized capture program is competing for second place in most of those deals and for the small remainder who genuinely arrive with an open mind.
The competitive outcome was decided during the period you weren’t measuring, by activity you probably defunded.
Why the smart thing keeps happening
The doom loop is not caused by stupidity. It’s caused by a reasonable process applied consistently.
Capture produces attributable pipeline this quarter. Creation produces an effect you can feel and not prove. So at budget time, the line with a number next to it survives and the line without one gets trimmed. Nobody in that meeting is behaving badly.
Do it four quarters running and you’ve concentrated the entire budget on the 5% who were already in market, in an auction where every competitor did exactly the same thing. Costs rise, because everyone is bidding on the same small pool. Rising costs get read as an efficiency problem, which triggers more optimization of capture, which means more budget moves to capture.
That’s the trap, and the mechanism is that measurability, rather than effectiveness, is doing the budgeting.
I wrote this morning about two tools reporting different numbers, which is an attribution problem with a technical answer. This is the other kind. No configuration fixes it, because the thing you want to see never touched a tracked surface.
Being fair to the other side, because capture deserves it
The creation argument gets overstated by people who’d rather not be measured, so let me put the counter-case properly.
If you cannot capture the 5%, creation spend is charity. A company that builds category awareness and then loses the deal to somebody with better brand search coverage has paid to educate the market on a competitor’s behalf. Capture comes first in the order of operations for a reason, and a startup with no capture motion should build one before it writes a single thought leadership piece.
Creation also has a genuine failure mode, and it’s not a small one. Unmeasurable budget attracts work that avoids being judged. Conference sponsorships nobody can defend, a podcast with forty listeners, a brand campaign that exists because somebody wanted to make it. “You can’t attribute it” becomes a shield.
So the position isn’t creation over capture. It’s that you need both, that they need different evidence standards, and that applying the capture standard to creation work reliably kills it before it can work.
What to measure instead of pretending
“Don’t measure it” is not an acceptable answer to a CFO, and it shouldn’t be.
You measure creation with leading indicators that are not attribution. Five that actually work:
Ask every new opportunity two questions. Who else are you evaluating, and how did you first hear about us. Log the answers. Thirty deals in, you have something no analytics tool can give you, and it’s the only direct evidence of whether you’re on shortlists.
Branded search volume over time. People searching your name were created somewhere. It’s a crude measure and it moves slowly, which is exactly why it’s a decent signal.
Direct traffic to pages that aren’t your homepage. Somebody arriving straight at a product page learned about it somewhere you can’t see.
Win rate on competitive deals, tracked separately from overall win rate. Creation work should show up here before it shows up anywhere else, because it changes how you’re perceived when compared.
Buying group coverage rather than lead count. With 13 people involved, a single MQL from one of them tells you very little. How many people from that account have engaged is a better question, and it’s the shift the analyst firms have been pointing at for two years.
None of those close the loop. All of them move before revenue does, which is what a leading indicator is for.
Where I come out
Split the budget deliberately and label the two halves honestly. Then hold them to different standards and write that down, because the write-down is what stops the quarterly re-litigation.
Capture gets judged on cost per opportunity and win rate, monthly. Creation gets judged on the five indicators above, over three or four quarters, and gets protected from the monthly review that would otherwise kill it.
And if the argument in your own company is stuck, go and ask ten recent customers how they first heard of you. The answers are usually a person, a conversation or something somebody read, and almost never the last click that got the credit. That’s not an attribution failure. That’s what the pre-contact part of the journey looks like when you finally ask about it.
The real risk isn’t spending on things you can’t measure. It’s that your entire marketing strategy has quietly become whatever the reporting could see.
My notes on lead generation cover the capture side in more detail. If you’d like to think through your own split, drop me a line on email, WhatsApp or LinkedIn and we can have a quick chat. I’m contracted full time so this isn’t a pitch.