How to Write the Perfect Pub Quiz: A Host's Craft Guide

The promise we made in our last post on running a pub quiz was that we'd come back and talk about the questions themselves. Here we are.

Writing a good quiz isn't about clever questions. It's about the right questions, in the right order, at a difficulty the room can handle.

This post is for two readers. One of you has been booked to host a quiz for the first time and you're sitting at a kitchen table with a blank document open. The other has been running quizzes for a while and wants to be better at the writing bit specifically. The principles are the same for both, but the difficulty section in particular will matter more if you're just starting out.

What you're actually doing when you write a question

A pub quiz question isn't just a test of knowledge. The best ones get a table arguing: the team that knows the answer feels clever, the team that doesn't gets to narrow it down and make a confident guess. Both are good. A question that produces neither (silence, instant agreement, or six bored faces) is a bad question, however interesting the fact behind it.

So you're not really writing for difficulty. You're writing for that moment of a table talking it out. A hard question nobody can have an opinion on is worse than an easy one the whole pub has a go at.

The question types, and what each is for

A good quiz uses several question types. Not for the sake of variety, but because different types do different jobs. Here are the main ones you'll use.

Straight factual questions

Most questions you write will be these. One question, one answer, a fact you either know or don't.

Good: "Which English king was crowned in 1066, after the Battle of Hastings?"

Why it works: Specific, single answer, well-known event, dateable. Teams who don't know it can still narrow it down (English king, 1066, Norman Conquest). The answer is William the Conqueror, which most teams will get, and even the team that says "Harold" can argue about it for a minute first.

Bad: "Which country has the biggest population?" Why it fails: Today? In 1900? In 1500? Even teams confident in the current answer (India, as of 2023) aren't sure whether you're asking historically or now. The question feels like a trap, and trap questions make a table go quiet. Specificity is everything; pin the question down to a single, unambiguous answer.

Closest-wins questions

Used for facts where the exact answer is unreachable, but a confident guess is. Reward the team whose guess is closest, rather than demanding the right answer.

Good: "How many steps are there to the top of the Eiffel Tower? Closest team wins the point."

Why it works: No team will know it. Every team has a number in their head within thirty seconds. The wide spread of answers means closest-wins actually picks a winner. (It's 1,665, by the way.)

Bad: "What year was the Battle of Trafalgar?"

Why it fails: This isn't a closest-wins question; it's a factual one with a date. Asking "closest wins" on something half the room actually knows turns it into a question of "which team has the closest schoolroom memory," which is irritating rather than engaging.

A rule of thumb for closest-wins: only use them for quantities nobody could reasonably know exactly. Lengths, weights, populations, statistics, prices over time. Not dates of well-known events.

Multiple choice

The lifeline. A factual question with three or four options, one of them right. Use these to make a hard question approachable: a team that would never know the answer cold can still reason between four options, which keeps them talking instead of writing nothing.

Good: "The first FIFA World Cup was held in 1930. Which country hosted and won it? A) Italy B) France C) Uruguay D) Brazil." (Answer: Uruguay.)

Why it works: Nobody is shut out. Even a team with no idea can argue their way to a fifty-fifty, and the team that knows it still gets the point. Multiple choice is the best tool you have for asking about something genuinely obscure without killing the round.

Bad: Multiple choice on something easy, where three of the four options are obviously silly.

Why it fails: If the answer is obvious, the options are just padding that slows the round down. Save the format for questions that actually need a leg-up.

Picture rounds

The visual round, and a welcome change of pace. Usually six to ten images on a printed sheet, marked at the half-time interval. Pictures can be celebrities (cropped, blurred, or as children), logos with the text removed, locations, food, anything visual.

Good: Six famous people photographed as toddlers.

Why it works: Recognisable enough that every team gets a few, hard enough that nobody gets all six. Teams chat throughout the round, which keeps the energy up.

Bad: Obscure historical figures nobody recognises even once you've given the answer.

Why it fails: The round dies, the table goes quiet, and you've wasted ten minutes of a ninety-minute quiz.

Picture rounds work because they let teams talk among themselves instead of waiting on you. They're also the easiest round to mark during the half-time break, which is why they belong in the middle.

One trick worth stealing: hand the picture sheet out at the very start, with the answer sheets, rather than when the round comes up. It's colourful, it gives early arrivals something to chew on before the quiz has even started, and it means teams can pick away at it in the gaps all night. It's how we run it at Quizzer, and it turns the dead ten minutes before kick-off into part of the game.

Music rounds

A universal favourite, and the hardest to get right. The standard format is ten short clips (ten to twenty seconds each); teams write down song title and artist, sometimes a point for each.

Good: A mix of decades, mostly recognisable, with one or two obscure clips to separate the teams who really know their music.

Why it works: Everyone gets something, the music-nerds get to shine on the hard ones, and the variety stops any one table feeling shut out.

Bad: Ten songs from the same decade you happen to like, or ten intros-only so even people who know the song can't place it, or ten tracks in a genre half the pub doesn't listen to.

Why it fails: If a whole table can't get a foothold in the round, they stop trying, and a dead music round drags down the back half of the quiz.

Specific rule: avoid Shazam-friendly clips. The chorus is the giveaway, and the easiest part to cheat on. Take the intro, an instrumental break, or a less-famous lyric. The best music questions are about songs people know, framed in a way they have to work to recognise.

Themed rounds

A round where every answer fits a category. "All answers are types of pasta." "All answers are countries with an island in their name." "All answers are anagrams of each other."

Good: The category is announced before the round starts.

Why it works: It gives lower-scoring teams a way in. Even if you don't know the answer, knowing it's a type of pasta narrows the field and keeps you guessing.

Bad: The category is the round's punchline, revealed only at the end.

Why it fails: It's cute for the writer and frustrating for players, who could have answered three more questions if they'd known the theme from the start.

The exception: a meta-round where teams have to spot the connection themselves, and there's a bonus point for naming it at the end. That works because the connection-spotting is itself the puzzle.

Difficulty: getting the spread right

This is where most quizzes go wrong. The temptation is to write questions you'd find satisfying yourself. The discipline is to write questions for the room you're actually walking into.

The rule of thumb: every round should have at least two easy questions (any team can get), at least two hard ones (top teams can get), and the rest in the middle. Not because difficulty needs to be neatly distributed, but because every team needs to leave each round with at least one point. A team that scores zero on round three has stopped enjoying themselves by round four.

Avoid the two extremes.

Too easy: questions where the answer is in the question. Nobody has to think, nobody talks, the round ends in seventy seconds and the energy drops.

Too hard: questions where only one team in twenty will know the answer, and even then will second-guess themselves. Teams stop writing and the mood deflates. If you find yourself thinking "I doubt anyone will get this but..." you're writing for yourself, not the players.

A good test: read each question and ask, "what does the average team in the pub do when they hear this?" If the answer is "look at each other and start talking," it's a good question. If the answer is "look blank and move on," it's the wrong difficulty.

What a good round actually looks like

Theory is easy to nod along to and hard to apply, so here's a real round of ours, annotated with what each question is doing and why it sits where it does. It's a five-question themed round on football, the kind you might drop in mid-quiz.

1. Which English football club's home ground is Anfield? (Easy. Most teams have it, and you open with a gimme so nobody starts the round demoralised. Answer: Liverpool.)
2. (Picture) Name this distinctive German stadium, lit red on home matchdays, home of Bayern Munich. (Easy-medium. A single picture question dropped into a text round changes the texture and gets the table pointing at the sheet. Answer: the Allianz Arena.)
3. England hosted the UEFA European Championship in which year, immortalised in the Lightning Seeds song "Three Lions"? (Medium. The song is the way in: a team that doesn't know the football might know the music. Answer: 1996.)
4. Which Italian club, founded in 1899, shares the San Siro with city rivals Inter Milan? (Medium-hard. The founding date and the San Siro detail give teams something to reason from. Answer: AC Milan.)
5. (Multiple choice) The first FIFA World Cup was held in 1930. Which country hosted and won it? A) Italy B) France C) Uruguay D) Brazil. (Hard on its own, which is why it's multiple choice: nobody gets shut out, and even a guessing team has a fair shot. Answer: Uruguay.)

Notice the shape: an easy opener, a picture question to lift the energy, a couple of medium questions with reasoning routes built in, and a genuinely hard closer rescued by multiple choice so no team finishes blank. That's what you're aiming for in every round.

Round structure: pace and shape

The order of rounds matters as much as the questions.

Open with something accessible. General knowledge or current affairs. The first round sets the room's confidence level. If you open with your hardest round, half the teams have written themselves off by the time the picture sheets come out.

Use the music round to lift the energy in the second half. Round four of five is the sweet spot. It picks the pub back up after the inevitable mid-quiz lull, and leaves you free to close on something stronger.

End with a strong round. Either a themed round with a payoff, a connection round with a final reveal, or your hardest general knowledge round so the winning team feels they earned it. Don't end with a round that's lower-energy than the one before. The last thing players remember is the last round; make it count.

Always have a tiebreaker ready. A single closest-wins question on something nobody can know exactly. "How many steps are there in the Empire State Building?" works. "What year was the Magna Carta signed?" doesn't, because someone will know it.

Where to get material

The honest answer: read more than the average person, take notes when something interesting catches your eye, and write questions regularly.

Practical sources:

• The Guardian quiz, the Times quiz, the FT lunchtime quiz. Read them weekly, even if you don't use the questions, to keep a feel for what counts as easy and what counts as hard.
• Quiz books. Old ones in particular: facts from the 1990s feel obscure now, which makes them satisfying.
• The QI books and the No Such Thing as a Fish podcast. Both are basically pre-formed quiz questions.
• Wikipedia "On This Day" and the random article button. Pick a topic, dig in for ten minutes, find one question worth asking.
• The week's news and cultural moments. New cinema releases, a chart-topping album, the Oscars, Eurovision, a big sporting final. A question (or whole round) about something that happened in the last fortnight rewards the teams who actually pay attention to the world, and it stops the quiz feeling like it was written in 2009. One or two current-affairs questions per quiz keeps it feeling alive.

A word on recycling. It's fine to reuse a great question after six months or so, particularly if you've changed venues or the team mix has rotated. It's not fine to reuse questions across consecutive weeks at the same venue. Regulars notice, and once they notice they'll lose trust in the quiz.

How long to spend

A new host should expect to spend four to six hours writing a quiz; an experienced one, two to three. The difference isn't speed, it's habit. Experienced hosts keep a running list of question ideas through the week and assemble them into rounds on Sunday, rather than starting from a blank page on Saturday night. If you're spending less than two hours on the questions, the quiz will feel undercooked and the pub will know. Over eight hours and you're second-guessing yourself, not writing. Pick a question, commit to it, move on.

Test your quiz before it goes live

This is the single most-skipped step, and it's the one that separates working quizzes from disasters.

You don't need a focus group. Read the quiz aloud to yourself the night before. Time each round. Anything that takes you more than twenty seconds to say is probably too long-winded. Anything you have to read twice to understand will be misheard in a noisy pub.

Better: rope in one or two friends or your partner and run it on them. They don't need to be quiz enthusiasts. You're not testing whether they get the answers; you're testing whether the questions make sense, whether the spread feels right, whether the music round flows. They'll tell you which question is unclear, which is too hard, which one they wish was longer. Then rewrite the bad ones the next morning while their feedback is still fresh.

The questions that survive a test run and still feel sharp are the ones you can take to the pub with confidence.

When you don't want to write your own

Writing a quiz every week takes real time. Some hosts love it (it's the reason they got into it in the first place); some find it the bit they put off until Sunday night and resent. If you fall into the second camp, or if writing isn't your strength, Quizzer's host network handles question-writing as part of the service. Get in touch if you'd like to see what that looks like.

For everyone else: write your own. The questions you write will get better. The voice of the quiz, which is what regulars come back for, only develops if you're the one writing.

A well-written quiz has three things in common. Every team has something they can answer. Every round has a reason to be there. And the questions sound considered when read aloud. Most quizzes you'll attend in your life won't manage all three. The ones that do are the ones people drag their friends to.