Creators became the distribution layer

There is a slow-moving squeeze in this industry that most quarterly decks describe as a cost problem. It is really a trust problem wearing a cost problem's clothes.

Paid acquisition gets more expensive every year. Brand campaigns convert worse than they used to. Meanwhile a host with forty thousand followers can move more first-time users in a Sunday show than a six-figure media buy, because the audience believes them and does not believe an ad.

None of this is new. What has changed is that the tooling gap finally became embarrassing.

The spreadsheet tells you everything

Go and look at how creator-run pools actually operate right now. Most of them run on a shared spreadsheet, a group chat and a person manually chasing entries on a Saturday morning.

That is the strongest demand signal in this entire category. People are doing tedious administrative work by hand, every week, unpaid, because the format is worth it to them. Nobody does that for an idea that does not work.

The industry response has largely been to treat these people as an acquisition channel — sign them, pay a CPA, send traffic. Which misreads what they built. They did not build a traffic source. They built a recurring competition with standings, in-jokes, a leaderboard and a reason to show up next week.

Why flat CPA is the wrong instrument

Cost-per-acquisition pays for an introduction and nothing that follows it.

If a host brings in a hundred people and forty of them are still playing six months later because the host keeps running the format, the host captures none of that. The operator does. Which means the best hosts — the ones whose audiences actually stay — are systematically underpaid relative to the ones who churn people fastest.

Any incentive structure that pays worst for the outcome you want most is going to get replaced. The only question is by whom.

What the tooling actually needs to do

Four things, in roughly this order of importance.

A builder that needs no phone call. Fixtures, format, entry, prize, duration, live in minutes. If launching a pool requires a partnerships conversation, the model does not scale past twenty hosts.

A branded surface. The pool carries the host's identity, not the platform's. The platform is the rail. Getting this backwards is the single most common mistake, and it is usually made for reasons that have nothing to do with the product.

A visible split. Revenue share on sustained activity, in a dashboard the host can screenshot. Ambiguity about money is what kills these relationships, every time.

Standings that persist. Leaderboards that survive between events. This is what turns a one-off promotion into a league, and it is the difference between an audience that enters and an audience that returns.

The two questions that decide the model

Who is liable when a host oversells it? In practice, and increasingly in law, the platform. Which means compliance rules have to live inside the builder — as constraints on what a host can create — rather than in a PDF sent afterwards and never read.

Who owns the league? If standings and audience data are portable, a host can leave and take the whole thing with them. If they are not portable, the best hosts will never fully commit. There is no clever answer here, only a decision, and it is better made deliberately in month one than discovered in month fourteen.

What this looks like when it works

The shift is from buying audiences to renting infrastructure to people who already have them. Marketing spend becomes a product line. The acquisition team becomes a platform team.

That is a genuinely uncomfortable reorganisation, which is roughly why it remains available.