Complete guide to aptitude & psychometric tests (2026)

Psychometric tests are standardised assessments that measure how you think, not what you know. Employers use them because they predict job performance better than CVs or unstructured interviews. This guide covers every type you will encounter and how to prepare for each.

The five test families

Almost every graduate or experienced-hire pipeline uses one or more of these. The names change between providers (SHL calls it “Verify”, AON calls it “scales”), but the underlying skill being measured is the same.

1. Numerical reasoning — tables, charts, percentages, ratios. Tests whether you can extract insight from data under time pressure.
2. Verbal reasoning — passages followed by True / False / Cannot Say. Tests whether you can distinguish what is stated from what is implied.
3. Logical (abstract) reasoning — shape sequences and pattern matrices. Tests fluid intelligence and rule extraction.
4. Situational judgement (SJT) — workplace scenarios where you rank or rate responses. Tests behavioural alignment with the employer’s competency framework.
5. Personality / behavioural style — forced-choice or Likert questionnaires. No right answers, but extreme profiles raise flags.

How scoring works

Raw scores are converted to percentiles — your rank relative to a norm group. A 65th-percentile score means you outperformed 65% of the comparison population. Most employers set a cut-off between the 50th and 70th percentile.

Adaptive tests (Talent Q, some AON batteries) adjust difficulty in real time. Getting harder questions is a good sign — it means the algorithm thinks you can handle them.

The four-phase prep plan

1. Week 1 — Diagnose. Take the Forge diagnostic (30 questions, free). It tells you which of the three core skills is weakest.
2. Week 2 — Drill weaknesses. Spend 80% of your time on the skill that scored lowest. Use tagged question packs so every session targets one sub-skill.
3. Week 3 — Timed mocks. Switch to full-length, timed tests. The goal is pacing, not perfection.
4. Week 4 — Simulate test day. Do one mock at the same time of day as your real test, with the same setup (desk, calculator policy, browser).

Provider differences that matter

SHL gives fixed-length tests. Talent Q adapts difficulty per question. AON uses a battery of short game-like modules. Cappfinity focuses on strengths-based scenarios. Knowing your provider changes your strategy — an SHL numerical test rewards skipping hard questions and returning later, while a Talent Q test penalises you for slow answers because the algorithm interprets hesitation as low ability.