Marketplaces are becoming media owners.
Buy the moment, not the audience.

Marketplace media is usually sold to you as an audience. Here is the platform, here are the millions of people on it, here is what we know about them. It is a familiar pitch, because it is the pitch every channel has made since the first ad network.

It is also the least interesting thing about what is happening to marketplaces.

An audience is the one thing in media that is never scarce. There is always another platform with millions of people on it and a deck explaining why they are the right ones. What is scarce, and what marketplaces have quietly ended up owning, is a moment.

The moment is the product

Physical retail worked this out a long time ago. The most valuable space in a shop is not the entrance. It is the till, and it is not because of the footfall.

A placement attached to a completed transaction borrows something a feed cannot. Not context, which is what native formats are for. Not attention, which everyone is selling. It borrows a decision that has already been made.

The person has just done the thing every media plan is ultimately asking for. They have parted with money. For a few seconds they sit in a state most advertising never gets near: the small satisfaction on the other side of a decision, before the tab closes and the day resumes.

No amount of audience data buys you that window. Only the platform that processed the payment knows it is open.

Why a marketplace can sell it and a feed cannot

For most of the last decade, retail media meant paying to appear higher in a retailer's search results, intercepting a decision already in progress. Resale and peer-to-peer marketplaces have search as well, but it is not the whole of the behaviour. People also arrive to browse, to list, to sell, and to buy things they were not looking for. Sponsored listings are built for the moment when someone already knows what they want, and that is only part of what happens here, which is why the formats emerging on these platforms look so little like one.

There is a second difference, with no real equivalent elsewhere in media. On these platforms the user is both the supply and the demand. They are built so that selling and buying feed each other: money earned listing something can sit as a balance and be spent again without ever quite feeling like spending.

Which makes it less useful to picture a browsing audience, and more useful to picture one with money already inside the system, buying to a rhythm that has nothing to do with your category or your promotional calendar.

Where this goes wrong

Two ways, and both are avoidable.

Judging it purely on the last click. Measurable channels get held to a standard of proof that unmeasurable ones escape. It is the mistake the industry has made with out-of-home for thirty years, running in reverse. A format with a claim mechanic attached will report a cost per acquisition, and a number that arrives that cleanly has a way of becoming the only one anyone looks at. Whatever it says, it describes the response it can see, not the whole value of having been in that moment. Judge it against what it displaced, not against its own reporting.

Assuming your category is welcome. Marketplace advertising can come with category restrictions, and they are often broader than an advertiser expects. A marketplace is not a neutral pipe. It is a business with a proposition to defend, and an advertiser sitting close to that proposition, or competing with it outright, is the one most likely to be turned away. When a plan falls over on this, it tends to fall over late, after the brief is written and the thinking is done.

Ask first. It costs nothing and it saves a fortnight.

Whether it earns a place on the plan

A short test, and it is not the one about audience fit. Do you have something a person could act on within ten seconds of finishing something else? A trial, a first-order incentive, a sign-up, a sample, a reason to download.

If yes, the moment has earned a line on the plan, and it should be judged on what that moment is worth, not on how its reach compares with a platform fifty times the size. A small amount of inventory at the right instant is not a worse version of a big audience buy. It is a different job, and should be briefed, bought and measured as one.

If your product needs a considered, three-week decision, this is not your channel, and no audience number will rescue it. Put the money where consideration gets built. That is not a weakness of the channel, it is what it is for. The mistake would be buying it as a general-purpose awareness channel with good targeting, because that describes about forty other things, most of them cheaper.

Marketplaces are not trying to be another feed. Their advantage is that they know something no feed knows: that the person just bought something, and is briefly willing to do it again.