When the Promoter Stops Bringing People

Promoter arrangements are among the last handshake deals left in hospitality.

A number per head, or a percentage of the table, or a flat night rate. Comped entry for their list, sometimes comped bottles. The terms live in a text thread and the evaluation lives in whether the room looked full.

That works right up until the night a promoter's crowd stops spending, which happens gradually and is invisible from the floor.

A full room and a profitable room are different things, and the promoter is paid on the first one.

The question the deal cannot answer

Take a promoter you pay. Ask what they brought you last month.

The answer you can usually get is headcount — the list, the scans, the door count attributed to them. The answer you usually cannot get is spend: what did the people on that list actually put on tabs.

Without the second number, the arrangement is being renewed on attendance. And attendance is exactly the metric a struggling promoter can still hit, by widening the list and dropping the quality of the invite.

A promoter under pressure does not bring fewer people. They bring cheaper people. That looks identical at the door and completely different at the register.

The number that matters

Revenue per attributed guest, tracked monthly, per promoter.

Total tab spend from guests attributed to that promoter, divided by the number of guests they brought. Compare it to your house average for the same night of the week.

Three readings, three conclusions:

Above house average — the promoter is bringing your better guests, and the fee is probably underpriced rather than over. Worth protecting.

At house average — you are paying for volume, which is a real service on a night you cannot fill. Fine, but price it as a filling service, not a quality one.

Below house average and falling — the list is being widened. This is the one to catch early, because it moves slowly and the room still looks fine.

Why the trend beats the level

A single month tells you very little. Promoter performance is seasonal, weather-dependent, and affected by whatever else is happening in the city that weekend.

Three months of direction tells you almost everything.

Revenue per attributed guest declining month over month while headcount holds steady is the clearest signal in the whole arrangement. It means the promoter is working harder to deliver the same number, and the way they are doing it is by lowering the bar on who gets on the list.

That is not necessarily bad faith. It is usually what happens when a promoter's core crowd ages out or moves to another room, and they backfill.

Either way you now know before the quarter, rather than after it.

Attribution without a wristband

The obvious objection: how do you know which guests came from which promoter.

Options, in rough order of accuracy:

Named list check-in. The promoter submits the list, the door checks names against it, the name goes on the guest record. Most accurate, requires the door to actually do it.

Unique link or code per promoter. Guests reserve through it, the attribution is automatic. Works if you take reservations; useless for walk-ups.

Table attribution only. Track only the tables the promoter books. Undercounts significantly — misses their walk-in crowd — but the number it does produce is clean, and a clean partial number beats a dirty complete one.

Night-level comparison. No per-guest attribution at all; compare nights the promoter worked against matched nights they did not. Crude, but it survives a busy door and it is better than nothing.

Most rooms should start with the fourth and move to the first for their two or three largest arrangements. Perfect attribution across every promoter is not worth the door friction it costs.

The comped bottle that is not in the deal

Promoter arrangements almost always accumulate side terms that were never written down.

A comped bottle for the promoter's own table. Entry for their staff. A round for the group that showed up early. Each one was a reasonable call on a specific night, and none of them are in the fee.

Add them up for a month and the effective cost per head is frequently well above the agreed rate — sometimes by a third.

The fix is not to stop doing it. It is to record it against the promoter, so the fee conversation happens with the real number. A promoter costing you the agreed rate plus 30% in unrecorded comps is not necessarily a bad deal — but you should be negotiating on the total, not on the part that got written in the text thread.

When to renegotiate, and how

The renegotiation is easier when you bring numbers, because the promoter usually has their own and they are usually about headcount.

The productive framing is not "your people spend less." It is: here is what a guest is worth to us on this night, here is what your guests are worth, here is the gap, what would it take to close it.

Frequently the answer is structural and fixable — the list is going out too late to catch dinner spend, or the entry time is pushing arrivals past the first hour, or the promoter is being measured on a metric that pulls against yours.

A promoter paid per head optimizes for heads. A promoter paid partly on the spend of their attributed guests optimizes for something closer to what you want. Even a small variable component changes behaviour, because it changes what they report to themselves.

The nights they do not work

The most under-used comparison in the whole arrangement.

Take the same night of week, the promoter working versus not working, matched for season. Compare total revenue, not headcount.

Some promoters produce a large lift. Some produce a headcount lift and a flat revenue line — the room is fuller and the register is not. That second case is not automatically a reason to end the deal; a full room has value in reputation and in what it does to the next weekend. But it should be priced as atmosphere, and atmosphere is cheaper than revenue.

Three numbers to hold

Revenue per attributed guest, monthly, per promoter, against house average for that night.

Effective cost per head — fee plus recorded comps plus entry value, divided by attributed guests. The number that should be in the negotiation.

Matched-night lift — the promoter's nights against comparable nights without them, measured in revenue.

Where the record has to sit

All three require guests, tabs, and attribution to live in the same place. When the list is in a messaging app, the door count is on a clicker, and the spend is in the POS, the join never happens and the deal gets renewed on how the room looked.

tasteck keeps guest records across visits with a source attached, so what did this promoter's crowd spend, over three months, against our average is a question with an answer rather than an argument.

The deal does not need to change. The evidence behind it does.

---

Read next

• The Guest List That Stopped Working
• RSA, Knock-Off and the Shift That Ends Early
• The Report You Open Every Monday