How our tests work
What’s in our IQ and personality tests, how we turn answers into scores, and what the results can and can’t tell you.
Updated 2026-09-30
The IQ test
The test has 30 puzzles and a 20-minute time limit. They cover five kinds of reasoning, six puzzles each, and get harder as you go:
| Area | What it asks you to do |
|---|---|
| Pattern recognition | Find the rule in a 3×3 grid of shapes and choose the missing piece |
| Numerical reasoning | Work out how a number sequence grows and give the next number |
| Verbal reasoning | Complete analogies and find the word that doesn’t belong |
| Logical deduction | Decide what must be true from a few plain statements |
| Spatial reasoning | Tell a rotated shape from its mirror image |
We write every puzzle ourselves. Pattern and shape puzzles are generated from explicit rules, and every question is checked automatically to confirm it has exactly one correct answer. Skipped questions count as wrong.
How we score it
Your number of correct answers is converted to the standard IQ scale, where 100 is average and the standard deviation is 15 (see the IQ scale explained). The conversion currently uses provisional norms: an estimate of how people typically perform, not a large, representative research sample. We will refine it as more people take the test, and we will update this page when we do. Scores are reported between 70 and 145.
Your free result shows a 10-point range, such as 110–119. The full report shows the exact estimate, your percentile and how you did in each of the five areas.
The personality test
The personality test uses the IPIP 50-item Big Five markers, a public-domain questionnaire developed by psychologist Lewis Goldberg. You rate 50 statements from “very inaccurate” to “very accurate”. Each trait has 10 statements, some worded in reverse to reduce bias. We add them up and show each trait on a 0–100 scale, where 50 is the midpoint of the scale rather than a population average. See the Big Five explained.
What the results can’t tell you
- They are estimates for self-reflection, not a clinical or diagnostic assessment.
- They shouldn’t be used for decisions about school, jobs or health.
- Online results aren’t accepted by high-IQ societies such as Mensa.
- A single sitting can be off by several points because of tiredness, stress or practice effects.
Changes
September 2026: first version of the relaunched tests, with 30 IQ puzzles and the IPIP 50-item Big Five.
Common questions
Is this a real IQ test?
It is a real test of reasoning, built on the same kinds of puzzles used in established IQ tests. It gives an estimate, not a clinical result. A formal IQ assessment is given one-to-one by a psychologist.
Where do the questions come from?
We write every puzzle ourselves. The pattern and shape puzzles are generated from rules, and every question is checked automatically to make sure it has exactly one correct answer.
How accurate is my score?
Treat it as a range of several points either side. Our scoring uses provisional norms, which we will refine as more people take the test, and any single sitting is affected by sleep, stress and familiarity with the puzzle types.
Try it yourself
Take the free IQ test