By Remco Livain in Marketing — 06 Aug 2026

We cut the top-of-funnel and the marketplace 'grew.' It was harvesting demand we'd already built.

When a home & living brand trimmed brand spend, marketplace sales climbed. The obvious read was the wrong one — and the dashboard was quietly cheering it on.

A home & living brand I look after trimmed its brand and top-of-funnel spend for a stretch — the usual pressure to make the numbers leaner. And right on cue, sales through the marketplace channel climbed. The obvious reading wrote itself: the marketplace is our strong channel, the expensive brand stuff clearly wasn't pulling its weight, so shift the money to where it's converting. That reading is wrong, and it's the most expensive kind of wrong because it looks like insight.

The demand didn't vanish. It went looking for a checkout.

Here's what actually happened. Cutting the top of the funnel didn't switch off the demand that funnel had already created. People who'd seen the brand, remembered it, half-decided on it — they were still out there, still ready. When they went to buy, they went where it's easiest to buy: the marketplace. So the marketplace "grew." But it didn't create that demand. It collected it. It cashed a cheque a channel you just defunded had written.

The channel that looks like it's winning is often just standing at the exit, collecting demand someone else paid to create.

Building versus harvesting

There's a solid framework for this, and it predates the marketplace era. Les Binet and Peter Field's "The Long and the Short of It" splits marketing into two jobs that work on different clocks. Brand building creates future demand — slowly, cumulatively, by lodging you in people's memory. Sales activation harvests present demand — quickly, by converting the people already predisposed to buy. As System1's write-up of the research puts it, the long-term work is a marathon and the short-term work is a sprint, and you need both.

Marketplaces sit almost entirely on the harvest side. They're brilliant distribution — low-friction places to convert intent. What they don't do is manufacture the intent in the first place. So if you defund the thing that creates demand and then admire the thing that harvests it, you've built a machine that looks efficient precisely because it's living off stored-up demand it isn't replenishing.

The dashboard is cheering for the wrong channel

Why does the wrong read feel so right? Because the measurement rewards it. Last-click attribution hands all the credit to wherever the purchase finally happened, and that's the harvester, every time. The channel that built the memory two months ago gets nothing on the report; the marketplace that closed the sale gets the gold star. I've written about this blind spot before — how your brand search isn't the hero your dashboard thinks it is, and how an analytics dashboard lies by leaving things out. The revenue line can keep climbing for a while even as the engine that feeds it runs down — which is exactly the trap in the revenue climbed and I assumed we'd spent our way there. The number going up is not proof the strategy is working.

Furniture makes the trap sharper

This bites harder in home & living than almost anywhere, because the purchase is high-consideration and slow. People spend weeks choosing a sofa. The brand memory you build in week one is what gets "cashed" in week three — and increasingly it gets cashed on a marketplace, under a listing that looks interchangeable with three competitors. Harvest-only looks wonderfully efficient right up until the moment there's nothing left in the field to harvest, and then it looks like a brand that's suddenly, mysteriously, become a commodity.

The discipline is resisting the obvious read

Binet and Field are careful that their 60/40 split — roughly 60% brand building, 40% activation — "is not an iron rule." The right mix depends on your category, your maturity, how often people buy. Fine. But "spend only where it converts" isn't a mix at all — it's a plan to strip-mine the brand for a couple of good-looking quarters, after which penetration softens and pricing power quietly erodes. You feel clever right up until you don't.

To be clear, this is not marketplace-bashing. Marketplaces are excellent distribution and I'd put more product on them, not less. The point is attribution honesty. Before you conclude that a harvesting channel is your growth engine, ask what fed it. If the answer is the brand spend you're about to cut, you're not reading a success story. You're reading the first page of a decline, dressed up as a win.

Sources & further reading

External

IPA — Binet & Field, The Long and the Short of It (brand building creates future demand; activation harvests present demand). System1 — The Long & Short of It (the 60/40 rule and why it "is not an iron rule").