Logic & Patterns
Fifteen puzzles that get harder, and an explanation for every one you miss.
This is a set of fifteen reasoning puzzles — number sequences, word relationships, and a few short arguments where the trick is noticing what does not actually follow. They start easy on purpose and end somewhere most people do not reach.
What makes it worth doing is the part after the score. Every question you get wrong comes back with the rule spelled out, because a number on its own tells you nothing you can use. Knowing you scored nine is trivia. Knowing that you keep missing the sequences where two operations alternate is a thing you can go and fix.
There is no time limit. These measure whether you find the pattern, not how fast you panic — and every question has exactly one defensible answer, so if two options both seem to work, one of them is a trap rather than a second correct reading.
How it works
- Fifteen multiple-choice questions, four options each, one right answer.
- No timer. Take as long as you want on any question.
- Your score is out of 15, with a short description of what that range typically looks like.
- Every question you miss is explained afterwards — what the pattern was and why the option you chose does not hold.
Common questions
- Is this an IQ test?
- No, and it deliberately does not report an IQ number. An IQ score means something specific: a result from a standardised battery, administered under controlled conditions, scored against a large representative sample. Fifteen multiple-choice questions taken on a phone is none of those things. Any website that hands you an IQ figure after a short quiz is inventing it, and the figure is usually flattering because a flattered visitor shares the result. We would rather tell you that you got eleven out of fifteen, which is true and checkable.
- What do these questions actually measure?
- Three overlapping things: whether you spot a rule from a few examples, whether you can hold a relationship steady while swapping its terms, and whether you notice when a conclusion sounds right but is not actually forced by what came before it. That last one is the least intuitive and the most useful away from puzzles. None of these is general intelligence, but they are not nothing either — they are the specific habits that show up when you have to work something out from incomplete information.
- I think one of the sequences has two valid answers.
- If that is true, it is a bug and we want to fix it. Every sequence here was checked so that exactly one option is consistent with all the given terms, and the distractors were chosen to be wrong in an instructive way rather than arbitrarily. The usual culprit is a rule that fits the first two terms and then breaks on the third, which is why the explanation for each answer names the rule and the term where the alternative fails. If you have a reading that holds for every term, write to us.
- Does a low score mean I am bad at reasoning?
- It means these fifteen puzzles did not go well, which is a much smaller statement. Performance on this format improves quickly with exposure — not because you get cleverer, but because the same half-dozen structures keep reappearing, and once you have seen alternating operations or a look-and-say sequence you tend to recognise them forever. That is also the honest reason to be sceptical of a very high score: it partly measures how many of these you have seen before.
- Can I retake it?
- As often as you like, though the questions do not change, so a second attempt mostly measures what you remember from the review. Nothing is stored between attempts — no account, no history, no comparison to a previous run — because there is no server here to keep it on.
This is not an IQ test and it does not produce an IQ score. A real intelligence assessment is administered under controlled conditions and validated against a large standardisation sample; fifteen questions on a website is none of those things. What this measures is how you did on these fifteen puzzles, which is a genuinely interesting thing to know and a much smaller claim.