What a screen-out actually costs

A study with a 12% incidence rate turns away seven people out of eight. Those seven each read the invitation, decided it was worth their evening, answered three or four screening questions honestly, and were told no. The buyer sees one line on an invoice. The panel sees seven people deciding whether this was worth doing again.

Whether the screen-out is paid decides which of those two views wins, and the decision compounds faster than almost anything else in panel economics.

Who actually bears it

There are three parties and the cost has to land on at least one of them.

1. The participant bears it when screen-outs are unpaid. Four minutes, no reward, no explanation. This is the default across most of the industry, and it is the cheapest option right up until the moment it is not.
2. The supplier bears it when screen-outs are paid out of margin. Predictable, small per event, and it turns an unbounded annoyance into a line item that can be budgeted.
3. The buyer bears it when it is priced in — a slightly higher cost per complete on low-incidence work, which is where screen-outs actually come from.

We put it on the second and third. Every screen-out pays, the amount is set per study rather than being a constant in the code, and low-incidence work is quoted knowing that more people will be turned away.

What happens when nobody pays it

The mechanism is not moral, it is behavioural, and it is fast.

A participant who has been screened out unpaid four times in a week learns something specific: the screener is the obstacle. Not the survey — the screener. The next time a question asks whether they manage a budget over $50,000, the honest answer and the profitable answer are different, and they have just been given four lessons in which one the system rewards.

The panel has now trained its own respondents to defeat its own screening. Every subsequent quality control exists to catch behaviour the incentive structure created. This is the single most expensive thing a panel can do to itself, and it does not show up anywhere until the data starts failing validation against known quantities.

An unpaid screen-out is not a saving. It is a small payment to teach somebody that lying to you is the profitable move.

Reading the rate as a measurement

Across a single study the screen-out rate is noise. Across a hundred it is one of the few clean measurements of how well the targeting upstream matches the panel underneath.

Observed rate	Usually means	What to change
Far below the estimate	The screener is letting people through it should stop, or the quota cells are broader than the brief	Check the screener logic before celebrating. A low screen-out rate on a genuinely rare audience is a warning, not a result.
Close to the estimate	Feasibility modelling and panel reality agree	Nothing. Record it — this is the calibration data that makes the next estimate better.
Well above, one market only	Profile coverage is thin in that market, so people are being asked to qualify on data nobody collected	Profile before fielding, not during.
Well above, everywhere	The brief describes an audience the panel does not contain	Say so before the study goes live. A supplier who fields it anyway is selling you three weeks and a shortfall.

The last row is the one worth arguing about. Telling a buyer their audience is not reachable costs a sale today. Fielding it anyway costs the relationship in a month, and the buyer still does not have their answer.

The arithmetic, since it is small

Take a B2B study, 12% incidence, 300 completes needed, a complete paying $4.00 and a screen-out paying $0.15.

• Completes required: 300 → $1,200 to participants
• Entries required at 12% incidence: 2,500 → 2,200 screen-outs
• Screen-out cost: 2,200 × $0.15 = $330

That is 27% on top of the reward budget — a real number, not a rounding error, and the honest reason most of the industry does not pay it. It is also roughly a quarter of what it costs to replace the respondents you lose by not paying it, and replacement cost is the comparison nobody runs because it arrives on a different budget line eighteen months later.

What a participant is told

Being screened out is the most common thing that happens to a panel member, and it is almost always presented as a dead end. Ours says which study, what it paid, that the payment has already landed, and that being screened out carries no penalty of any kind — because the most frequent question that follows is whether it counted against them.

It does not, and it cannot: screen-outs are not quality events, they do not touch the quality score, and a person who is screened out of nine studies in a row has done nothing wrong at all. They have simply been asked nine questions that were not about them.