The Aon (cut-e) Scales Numerical Test: Why It's Nothing Like SHL

Candidates who have prepared thoroughly for SHL-style numerical tests often walk into the Aon numerical reasoning test — still widely known by its original name, cut-e — and get caught completely off guard. Not because the maths is harder. Because almost every habit that serves you well on a conventional numerical test works against you on this one.

Cut-e was a German assessment company acquired by Aon, and its "scales" test family is used by a number of major employers, particularly in finance — Deutsche Bank is among the firms reported to use Aon assessments for graduate and early-careers screening, though providers can vary by role and region, so always confirm from your own invitation email. If that email mentions Aon, cut-e or "scales", this is the test family you are dealing with, and it deserves preparation on its own terms.

The Aon scales numerical format: 37 questions in 12 minutes

The full-length version of the Aon scales numerical test presents 37 tasks in 12 minutes. A shorter version runs 18 tasks in 6 minutes. Either way, that is roughly 20 seconds per question — compare that with SHL's Verify Interactive at nearly two minutes per question, and you begin to see why preparation for one transfers so poorly to the other.

Each task is a statement about a set of company data, and you must mark it True, False or Cannot Say. The data is not presented as a single neat chart. It is spread across multiple tabs — different tables, charts and documents for different aspects of a fictional business — and each statement tells you which area to look in. Part of the task is navigating to the right tab quickly and reading only what you need. Candidates report that clicking through tabs under a 12-minute clock is a skill in itself, and it is one that generic numerical practice never builds.

The test also comes in different industry variants — commonly described as consumer, finance and industrial versions — with data themed to the sector you are applying to. The underlying skills are the same, but the finance version's data (portfolios, balances, rates) rewards familiarity with financial tables.

How the Aon numerical reasoning test is scored: why guessing can cost you

Here is the single most important thing to know, and the reason standard test advice backfires: preparation providers consistently report that the scales tests penalise wrong answers — your score reflects correct answers with deductions for incorrect ones, and unanswered questions are not penalised.

On most numerical tests, the standard advice is sound: never leave a blank, because a guess has positive expected value. On a test with penalty scoring, that logic inverts. A blind guess on a three-option question is more likely to hurt you than help you. The better strategy, widely echoed by candidates and preparation providers alike, is to answer only when you are confident, skip cleanly when you are not, and never burn 90 seconds wrestling one stubborn question — at 20 seconds per question, one stubborn item costs you four or five answerable ones.

Nobody completes all 37 questions comfortably. The test is engineered so that almost everyone runs out of time, which means your result is determined by the trade-off you strike between speed and accuracy. Candidates who don't know about the penalty scoring strike that trade-off wrongly — they rush, guess, and quietly bleed points.

The verbal sibling: scales verbal

If your invitation includes verbal reasoning, expect the same design philosophy. The Aon scales verbal test presents up to 49 statements in 12 minutes — around 15 seconds each — again answered True, False or Cannot Say against information spread across tabs. As with the numerical test, you are not expected to finish; you are expected to be accurate at pace. And as with any True/False/Cannot Say format, the discipline is strict textualism: a statement is only True if the relevant tab fully supports it, only False if the tab at least partly contradicts it, and Cannot Say when the information simply is not there — regardless of what you happen to know about the real world.

How to prepare for a test built to feel unfinishable

Recalibrate your expectations first. If your practice standard comes from SHL-style tests, your internal clock is set to the wrong speed. Sit at least one full timed run of a scales-format practice test purely to feel the pace. Most candidates describe the first attempt as a shock; that shock is much cheaper in practice than in the real thing.

Train the navigation, not just the maths. The arithmetic itself — reading values, comparing figures, checking simple calculations — is not advanced. What needs training is the loop: read statement, identify the right tab, find the figure, decide, move on. That loop only gets faster through repetition on the actual format.

Practise skipping. This sounds trivial and is not. Under time pressure, walking away from a half-solved question feels like losing. On a penalty-scored test, it is often the correct play. Practising the discipline of a clean skip — no lingering, no guessing — is genuinely one of the highest-value things you can rehearse.

Know your baseline before you drill. The scales format punishes two different weaknesses — slow data location and shaky rapid arithmetic — and they need different work. A timed diagnostic taken before you start preparing tells you which one is actually costing you marks, rather than leaving you to guess.

The honest summary

The Aon (cut-e) scales numerical test is short, dense, penalty-scored and deliberately unfinishable — the near-opposite of the long-form, generously-timed numerical tests most candidates practise on. Treat it as its own discipline. Confirm the exact test family from your invitation, practise the real format under the real clock, and be strict about answering only what you can answer confidently.

Before you start preparing, find out where you actually stand. The Acumen Logic Benchmark is a free, one-shot diagnostic — 20 questions under real timed conditions, taken once, like the real thing. It shows you your percentile and exactly where your speed–accuracy trade-off sits before you commit hours to practice.