Aptitude tests
Logical and inductive reasoning test practice
Work out the rule from the examples, then apply it.
Inductive reasoning tests show you a few examples, such as shapes, grids or number sequences, and ask you to work out the underlying rule and apply it to a new case. Different publishers call them inductive, abstract, logical or diagrammatic reasoning. They all ask the same thing: spot the rule, then predict.
Our generator builds each puzzle from a stated set of rules, so every question has one defensible answer and the explanation can name the rule. You can practise the published timings, an untimed mode, or learn mode. Source badges below show which details are published.
Practise this format
Choose a format
- Inductive grids (Aon-style)Two grids share a rule. Pick the two of four grids that follow it. 6 minutes: do as many as you can. A skipped grid counts as wrong.
- Swift-style aptitude: verbal, numerical, diagrammatic (Saville-style)Three timed sections of 8 questions and 6 minutes each: 24 questions in 18 minutes. Shows your ability and your pace.
- Numerical and diagrammatic sections (Saville-style)Two timed sections of 8 questions and 6 minutes each. Shows how fast you were as well as how many you got right.
- Inductive reasoning, multiple choice (SHL-style)Up to 15 pattern questions in 18 minutes. Pick the figure that completes the series or the matrix.
- Interactive inductive reasoning (SHL-style)Up to 15 pattern questions in 18 minutes. Some you answer by picking, some by building the missing figure yourself.
- Logical reasoning (Talent Q-style)Twelve shape-pattern questions in 15 minutes. The difficulty follows your answers.
Need extra time or a different format? Every timed practice has 1.25x and 1.5x options. See reasonable adjustments.
Independent practice. Not affiliated with or endorsed by Aon, Saville, SHL and Korn Ferry Talent Q.
What we know about this format
Each point carries a badge showing where it comes from. Where only candidates have reported something, the badge says so.
What do the confidence badges mean?
- Vendor-published Stated in the test publisher's own document or practice site.
- Academic From peer-reviewed research.
- Official employer Stated on the employer's own page.
- Near-official Near-official: an employer-run practice portal, or a vendor or employer-quoted press release naming the employer.
- Candidate-reported Reported by candidates. Treat as likely but unconfirmed.
- Third-party From prep sites, press or forums. Not confirmed by the employer or publisher.
- Vendor-published
Interactive inductive test in the format we call SHL-style: up to 15 items in 18 minutes, using shape and letter or number sequences.
Source · Publisher fact sheet, 2022.
- Vendor-published
Aon-style 'ix' format: two grids share a rule, and you pick the 2 of 4 grids that follow it. The rule is only ever about the number and position of objects. 6 minutes, as many as possible, and skipped items count as wrong.
Source · One client's candidate guide.
- Vendor-published
Saville-style Swift Analysis: 24 items in 18 minutes across verbal, numerical and diagrammatic sections of 8 items and 6 minutes each.
Source · Publisher practice page. The technical summary is now offline, so section counts were read from a search excerpt.
- Vendor-published
The general-ability combined test is 30 items in 36 minutes (10 numerical, 10 deductive, 10 inductive), or 24 items in 36 minutes in the interactive version.
Source · Publisher fact sheets.
- Academic
Retake studies find an average gain of about a quarter of a standard deviation on cognitive tests, levelling off after a few sittings.
Inductive grids (Aon-style)
- Vendor-published
6 minutes, as many as possible; two grids share a rule and you pick 2 of 4 grids that follow it
Source · One employer's set-up.
- Vendor-published
The rule is always about the number and position of objects, never NOT or OR
- Vendor-published
Skipped questions count as wrong
- Third-party
30 questions are queued so you are unlikely to run out in 6 minutes
Source · The real number of items is not published; the set is built to be longer than anyone can finish.
Swift-style aptitude: verbal, numerical, diagrammatic (Saville-style)
- Vendor-published
24 questions in 18 minutes: verbal, numerical and diagrammatic sections of 8 questions and 6 minutes each
- Vendor-published
Separately timed sections built from large item banks
Source · Document is no longer online; read through a search excerpt.
- Candidate-reported
Groups of questions share one stimulus
Source · Candidate-reported. Not confirmed in the provider's published documents.
- Vendor-published
Reports show an ability score (sten, 1 to 10, against a comparison group) and a pace score (your time on completed items against the average time on the same items)
Source · Kvanta's sten is against Kvanta players and its pace line uses our own reference times.
- Third-party
Calculator hint shown on the numerical section
Source · Calculator policy is not officially documented. Check your invitation.
Inductive reasoning, multiple choice (SHL-style)
- Vendor-published
Up to 15 questions in 18 minutes; shape and letter or number sequences
Source · This is the multiple-choice version of the interactive format described in the fact sheet.
- Vendor-published
Verify G+ questions are multiple choice, or drag and drop in the interactive version
- Vendor-published
The test is adaptive
Source · Kvanta uses a simple difficulty staircase over five levels.
- Vendor-published
Results are reported as a percentile against a named comparison group; Kvanta shows an estimated band against Kvanta players instead
Source · Kvanta never shows an SHL percentile or grade. Its band is a rough guide against our own users and appears only once enough people have taken the test.
Interactive inductive reasoning (SHL-style)
- Vendor-published
Interactive versions use drag and drop and partial scoring
Source · Kvanta builds the answer by tapping cells rather than dragging, with the same idea.
Questions people ask
What is inductive reasoning?
It is reasoning from examples to a rule. You are shown a sequence or set of patterns, work out what changes from step to step, and choose what comes next or which option fits.
What kinds of rules come up?
Typical rules concern counting objects, position, rotation, reflection, shading, size and simple arithmetic on sequences. The Aon-style format described in one client's guide uses only number and position of objects.
Is inductive the same as abstract or diagrammatic reasoning?
Publishers use the names differently, but the task is the same family: infer a rule from visual or symbolic examples. Diagrammatic tests usually use flow-chart-like or box-and-arrow diagrams.
How should I approach a puzzle I cannot see at once?
Look at one feature at a time, such as count, then position, then shading, and note what changes between frames. If two rules fit, look for the one that explains every frame. In learn mode you can take as long as you like and read the rule afterwards.
Do skipped questions lose marks?
In the Aon-style guide for one client, skipped items count as wrong, so a guess costs nothing extra there. Other publishers do not document it.
Can practice really help with this?
Retake research shows gains on reasoning tests, mostly early. Seeing common rule types before test day is the main benefit, along with a calmer pace.
Are these the real puzzles?
No. Our puzzles are generated by our own rules, and we do not copy or reconstruct any publisher's items.
Where this appears in firm processes
- PwC UK Official employer
- EY UK Official employer
- KPMG UK Official employer
- UBS Official employer
- Standard Chartered Official employer
- Schroders Official employer
See all firm pages. Check your own invitation, because processes change each cycle.