By Haru · Updated · 2026-09-07
Quiz Night Ideas — Pub Quiz Rounds That Work
A quiz night rarely fails for lack of questions — it fails because the questions are all the same difficulty. Six rounds of ten, with two easy and two hard questions inside every round, will beat a hundred clever questions arranged badly, because a table that answers nothing in round one stops playing long before the night ends. This page is about the arrangement rather than the trivia: how many questions you actually need, and how to spread difficulty so both the confident table and the quiet one stay in it. It also covers which round formats a phone cannot beat, when to read the answers out, and where to get ready-made rounds without writing sixty questions yourself.
Six rounds of ten is the shape to start from
Sixty questions in six rounds fills about two hours once you include scoring breaks, and that is the shape to copy until you have a reason to change it. The arithmetic is simple: a question takes roughly forty seconds to read, answer and let tables argue about, so ten questions run eight to ten minutes, and the break between rounds — swapping answer sheets, reading answers, updating the board — takes another five. Six of those cycles land near two hours, which is as long as a room stays sharp.
The number of rounds matters more than the number of questions. Fewer, longer rounds punish a table that starts badly, because a team sitting on a bad score has nothing to reset against. More, shorter rounds give the room a fresh start every ten minutes and a scoreboard that keeps moving. If your night is only an hour, cut to four rounds of ten rather than six rounds of six — the break is what makes it feel like a quiz rather than a test.
Mix difficulty inside every round, not across them
The most common mistake in a homemade quiz is the hard round — one round built out of the questions the host found interesting. It reliably kills the night, because difficulty stacked in one place produces a round where half the room scores two and stops caring. Spread it instead: in a round of ten, aim for two questions almost everyone gets, six that a decent table gets after arguing, and two that separate the winners.
The two easy ones are not filler. They are what keeps a table that knows nothing about the topic still writing something down, and a table still writing is a table still playing. The two hard ones do the opposite job — they decide the round, and they are the reason the strong table cannot coast. A useful test while writing: read your ten questions and mark each one with the fraction of the room you expect to get it. If nothing is above eighty percent and nothing is below twenty, the round has no shape and every table will finish on six.
Pick round formats a phone cannot answer in five seconds
Straight factual questions are the easiest to write and the easiest to cheat, so build at least half your rounds out of formats where looking something up does not finish the job. Four that work anywhere:
- Ordering. Put four things in sequence — oldest to newest, smallest to largest, nearest to furthest. A search gives you one fact at a time, but the answer needs all four in the right order.
- Connections. Five answers that share something the sixth line asks for. Even a table that gets every answer still has to see the link.
- Which came first. Two familiar things, one date each. It reads as trivia and plays as an argument, and it is nearly impossible to look up as a single query.
- Pictures and sounds. A logo with the wordmark removed, a close crop, a five-second clip. These are slow to search and fast to answer if you actually know.
A phone-free rule is easier to announce than to enforce, and enforcing it makes you the police for two hours. Writing rounds that reward knowing over searching does the same work and costs you nothing on the night.
Read the answers out at the end of each round
Read the answers after each round, not in one block at the end of the night — the ten minutes a table just spent arguing is exactly when they want to know, and a whole night of held-back answers turns the finish into an admin session nobody enjoys. Reading them per round also lets tables swap and mark each other's sheets, which removes the host's biggest bottleneck and makes the scoring visibly fair.
Two details are worth deciding before you start. Say in advance how strict spelling is — the usual answer is that a recognizable attempt counts, and saying so early stops three arguments later. And keep the running total visible; a scoreboard nobody can see removes the only thing that makes round four matter to a table sitting in fifth place. If you want the ending to stay live, hold back only the final round's scores, not the answers.
Where to get the questions without writing sixty
Writing sixty original questions takes most hosts three to four hours, which is why borrowing whole rounds is normal rather than lazy. A ready-made themed quiz is a round: play it, take the questions that fit your room, and drop the rest. The advantage over a list of "100 trivia questions" is that a real quiz has already been arranged by difficulty, so you inherit the shape as well as the content.
Our own themed quizzes are built to be used this way — fifty questions each, free, no account. For a general opening round the mythology quiz and the art quiz both start gently and get sharp. The ocean quiz and the dinosaur quiz work as the family-friendly round that keeps younger tables in it. For a sports round the baseball quiz is deeper than most pub sets, and the cooking quiz makes an unusually good closer because almost everyone has an opinion about food. Take ten from one, ten from another, and you have written two rounds in five minutes.
If you would rather sharpen your own question-writing, our guide to getting better at trivia covers how quiz answers are actually remembered, which is the same skill from the other side of the table.
Teams, scoring, and the tie-break you will actually need
Four to six players per team is the range where everyone still talks; past that, two or three people answer and the rest watch. If you have a big room, more teams beats bigger teams — eight teams of four is a better night than four teams of eight, even though it is the same number of people.
Scoring can stay at one point per question all night, and the temptation to weight hard questions double is worth resisting, because it hands the win to whoever guessed the two obscure ones rather than to the table that was steady for sixty. Keep a joker or a wildcard if you want a comeback mechanic — one round a team can nominate to double — and announce it at the start so it feels like strategy rather than a rule you invented in round five.
Write the tie-break before the night starts, because you will need it more often than you expect. The standard is a numeric estimate nobody knows — a distance, a height, a year — with the closest answer winning. Numeric tie-breaks settle cleanly, which is exactly what you want at eleven o'clock.
Five ways a quiz night goes wrong
Most bad quiz nights fail for one of the same five reasons, and all five are decided before anyone arrives.
- Every question aimed at the host's own knowledge. If four rounds are music and film, the tables who came for a quiz rather than for your taste are out by round two.
- A round with no easy questions. Covered above, and it is the single most common one.
- Questions that need one exact word. "Name the third character" produces arguments; "which of these three" produces answers. Multiple choice is not a lesser question — it is a faster one.
- Too long. Two hours is the ceiling for most rooms. Eight rounds looks generous on paper and empties tables in practice.
- Answers that turn out to be wrong. One disputed answer costs more goodwill than a whole weak round. Check anything you half-remember, and if a question depends on a record or a current title, check that it is still true this year.
None of these is about the questions being clever enough. They are about the night being playable, and that is entirely in the arrangement.
Frequently asked questions
Are online quiz questions free, or do I need an account?
Plenty are free, but the ones that ask for an email usually do it to send you something later, not because the questions cost anything. Every quiz on this site is free to play with no sign-up, no email and no app, and the questions are visible on the page, so you can read a whole set before deciding whether to use it as a round. If a site wants an account before it shows you a single question, you are paying with your inbox for something widely available without it.
How many questions do I need for a two-hour quiz night?
Sixty is the standard answer — six rounds of ten — and it fills roughly two hours once you include the breaks for marking and reading out answers. Write two or three spares in case a question turns out to be broken or already known to the room. If your night is an hour, use forty in four rounds rather than shortening every round, because the break between rounds is what keeps the room fresh.
Can I just use an online quiz as one of my rounds?
Yes, and it is the fastest way to build a night. A themed fifty-question quiz gives you five rounds' worth of material already arranged by difficulty, so you can lift ten questions and keep the shape. The one thing to check before using any set is the answers: play it through yourself first, so a wrong or out-of-date answer surfaces in your kitchen rather than in front of thirty people.
What do I do about people looking up answers on their phones?
Design around it rather than policing it. Ordering questions, connection rounds, picture rounds and "which came first" all survive a phone because the search only returns one piece of the answer. If you do want a phones-down rule, announce it as part of the format at the start and give an amnesty round where phones are allowed — it is far easier than watching thirty people for two hours.
How hard should the questions be?
Aim for the average table to finish each round on six out of ten. That comes from mixing inside the round — two questions nearly everyone gets, six that reward a decent table, two that decide the winner — rather than from picking a single difficulty level. If your last quiz had rounds where most teams scored two or nine, the problem was the spread, not the topic.