Starting a quiz night at your pub

The case for a quiz night is simple arithmetic. A midweek evening in most pubs is a handful of regulars nursing one drink each. A working quiz night puts eight to twelve tables in the room for two hours, and those tables drink through a break you have deliberately built into the middle of it.

The second-order effect is bigger and slower: a quiz night makes a Tuesday into a fixture. People plan around it, arrange to meet at it, and turn up in weeks when they would not otherwise have come out at all.

The case against is that it is work every single week, forever, and that it takes a couple of months to know whether it is going to take. Both of those are real and neither is usually the thing that stops people. What stops people is the week-to-week effort of writing and marking, which is the part worth designing away at the start.

Tuesday and Wednesday, almost always. They are the deadest nights in most pubs, which means the quiz is adding trade rather than moving it, and they are far enough from the weekend that people will come out.

Monday is harder than it looks, people are tired and skint. Thursday works in a town with students or a young crowd and works badly in a commuter village. Friday and Saturday are a mistake unless the pub is genuinely quiet then, because a quiz on a busy night annoys everyone who came in for something else.

Pick one and do not move it. A quiz that moves loses its regulars, and its regulars are the whole asset.

Three things: somebody's time, a prize, and whatever you use to run it.

The time is the real cost and the one people underestimate. Writing a quiz from scratch is two to three hours a week, every week, and it is the reason most quiz nights end. Using ready-made packs changes that to about twenty minutes of reading them through, which is the difference between a sustainable night and one that dies in March.

The prize should be small and should mostly come back to the bar. A bar tab is the standard for a reason: it costs you the wholesale price rather than the retail price, and it gets spent in the building.

The running cost of software is the smallest number in the list. What matters about it is whether it removes the two hours, not what it charges.

Six to eight weeks. Be ready for that, because the first three will be discouraging.

A typical shape: week one draws a crowd out of curiosity, week two is half that, weeks three and four are the low point, and somewhere around week five or six it starts to build on the people who came back. The teams that come to week six are the ones you actually have.

The mistake is cancelling in week three. The second mistake is changing the format in week three, which resets the clock on the people who were starting to get used to it.

Not takings on the night, which are too noisy and affected by things that have nothing to do with the quiz. Watch these instead.

Some rooms do not take to it, and running a failing quiz night out of stubbornness costs you a night's trade every week.

If after ten or twelve weeks you are still getting three tables and none of them are the same three, the format is not the problem and neither is the software. It is usually the room, the night, or the fact that another pub four hundred yards away already owns quiz night in that town.

Before giving up, try moving the night once. Not the format, the night. It is the cheapest experiment available and it is right more often than you would think.