Consulting Aptitude Test 2026: Name Your Test Publisher

Why naming the aptitude test product decides everything else

Almost every page about this subject describes a category. Categories have no timing. Products do, and the publisher puts the timing in writing.

SHL sells its cognitive range under the Verify name, and each product ships with a one page fact sheet. We downloaded seven of them and read the specification lines. Here is what they say, with the wording taken from the sheets themselves.

The spread runs from 10 minutes to 36. The per item budget runs from 50 seconds, which both the calculation test and the verbal test land on exactly, to 108 seconds on the interactive numerical test. Two products carry the same G+ name, both at 36 minutes, one with 30 multiple choice items and one with 24 activity based items. A page that tells you how long your test lasts without naming the product is wrong a large share of the time.

One line appears on the deductive sheet and nowhere else in the seven: "Number of Sittings One plus Verification Test". Six of the seven simply say "One".

What the invitation email is for

Your invitation names the platform. That makes it the only current source on which product you are about to sit, and it outranks every guide including this one.

Bain writes the routing down. Its digital assessment page says that "you will receive an email after submitting the application detailing whether you are required to complete Sova or Test Gorilla", and the two arrive with different rules. TestGorilla is to be completed in one sitting. For Sova the firm says you "do not need to complete the Sova assessment in one sitting" as long as every section is done before the deadline. Acting on the wrong one of those two costs the application. The page is at bain.com, checked 29 September 2026.

Two more things from that page are worth carrying into the sitting. Pen, paper and a calculator are permitted for the assessment, and the calculator stops being permitted at the interviews that follow. And the deadline is absolute: "Applications will not be considered if the digital assessment is not completed by the deadline."

Some firms describe the content instead of the vendor. OC&C calls its stage an "Online Blended Assessment" that "Combines numerical and critical reasoning questions alongside behavioural and motivational questions", and adds that the link may arrive before the vacancy closes, "just to stagger the amount of people doing the test at one time". Its vacancy page is at careers.occstrategy.com, checked 29 September 2026.

A worked numerical item, with the arithmetic shown

The published description of the numerical family is narrow. SHL's candidate report says the numerical test measures the ability to make correct inferences from numerical data, and that "the emphasis in these tasks is on understanding and evaluating data rather than on computation". Items are built on a table you read once and query several times.

Ours below is written by us in that format. Publisher items are not reproduced anywhere on this site.

Which region earns the most revenue per consultant, and how much? Region A. Dividing 48,000,000 by 120 gives 400,000 per consultant. Region B gives 32,400,000 divided by 90, which is 360,000. Region C gives 21,000,000 divided by 70, which is 300,000.

If Region B reached Region A's revenue per consultant with the same headcount, how much revenue would it add? The gap per consultant is 400,000 minus 360,000, so 40,000. Multiply 40,000 by 90 consultants and you get 3,600,000, which is $3.6 million.

Two habits are visible in that answer. The division was set up before any number was typed, because the question asked for a ratio and the table gives two columns. And the second part reused the first result instead of recomputing the whole table, which is where the time saving on these items actually lives.

A worked deductive item, four slots and three rules

SHL describes its deductive practice test as measuring the ability "to draw logical conclusions based on information provided, identify strengths and weaknesses of arguments, and complete scenarios using incomplete information". The scheduling form below is ours.

Four candidates, W, X, Y and Z, occupy four consecutive 45 minute interview slots starting at 9:00. Y is not in the first slot. X is interviewed immediately after W. Z has the 10:30 slot.

Which slot does Y have? The 11:15 slot. Work it in the order the constraints allow. Four slots run at 9:00, 9:45, 10:30 and 11:15. Z takes the third. W and X must be adjacent, so the pair sits in slots one and two, two and three, or three and four. Z holds the third, which rules out the last two of those. W goes first, X second. Y takes what is left, the fourth slot, and the rule about the first slot is satisfied.

The move that solves it is picking the most constrained statement first. Z is fixed to one slot, and that single fact removes two of the three possible homes for the W and X pair. Starting from the negative statement about Y leaves all three open, so it does no work on the first pass.

Check the conclusion against every rule before committing. Deductive items have exactly one arrangement that satisfies all of them, and an arrangement that satisfies two of three is a wrong answer that feels finished.

A worked inductive item, described instead of drawn

Inductive items are the odd family on a written page. The G+ fact sheet says the inductive questions "are completely non-verbal and feature only shapes and figures", so a faithful example would be a picture. We describe the rule instead, which lets you check your reading of it.

Five cells in a row. Cell one holds an arrow pointing up with one dot beneath it. Cell two holds an arrow pointing right with two dots. Cell three holds an arrow pointing down with three dots. Cell four holds an arrow pointing left with one dot.

What does cell five hold? An arrow pointing up with two dots. Two independent rules run in parallel. The arrow turns 90 degrees clockwise each step. The dot count cycles through one, two, three and back to one, so cell four resets and cell five holds two.

The trap in this family is finding one rule and answering with it. A single rule explains the arrows perfectly and gives the right direction with the wrong dot count. Count the attributes that change before you predict anything: direction, quantity, size, shading, position. Each is a separate rule, and the item is only solved when every changing attribute has one.

The aptitude test settings that belong to the employer

Several numbers you will read about this stage are not properties of the test. They are options the hiring firm picked, and the publisher documents them for its clients.

TestGorilla's client guidance says the clock is not negotiable for the candidate: "The time limit of each test is fixed and cannot be changed." The firm can grant extra time for every test in the assessment, and doing so degrades its own comparison, because percentile scores "are calculated based on candidates who took tests with the standard time limits". That page is at support.testgorilla.com, read on 30 September 2026 through a text proxy after a direct request returned 403.

Three more settings from the same page matter to you. The window starts at the invitation, and the example given is a seven day deadline running "from the moment they are invited". The published floor is three days, since "Deadlines have a minimum of 3 days provided to ensure candidates have sufficient time to complete the assessment". And a candidate already in the test when the deadline passes can finish.

Accommodations are a candidate action there as well. The publisher says it lets candidates apply for them, naming non fluent speakers of the assessment language and conditions affecting concentration or memory. Sova takes the other route and sends you to the employer. Its hub says that if adjustments are needed, "we recommend that you contact the organisation you are applying to".

Three ways the score leaves the table

Sitting the wrong product in your head. Candidates budget for the test they read about instead of the one they were sent. Twelve items in 10 minutes and 30 items in 36 minutes want opposite habits. Divide allowed time by maximum items the moment you know the product, and rehearse at that pace.

Treating a skip as free. TestGorilla tells candidates the opposite in its own words: "Skipped questions count as wrong answers. If you don't know the answer, guess." Under an adaptive design a blank is also a lost measurement, because the next item is chosen from your last answer.

Losing the session to the machine. SHL's help pages list conditions that end a sitting, including switching away from the test window. One article states that a test "will get locked permanently if a candidate is found switching through windows while taking the Test". Close everything else, plug the laptop in, and do not open a second tab to check a formula.

The fourth failure mode is the calendar, and it is the cheapest to remove. Track the invitation date rather than the application deadline, because the two are different dates and only one of them expires your test.

Five aptitude test facts nobody publishes

The pass mark. SHL's candidate report shows a percentile against a named comparison group and a letter band from A to E. It never says which band continues. The employer sets that, and no firm we have read publishes it. Any page quoting the score you need at a named firm invented it.

Which firm uses which publisher. SHL and HireVue do not publish client lists. Bain is the exception that proves the shape of the problem, because it names two possible vendors for the same stage and says the email decides. Even there, the criterion that routes you to one or the other is not published.

What your sitting contains. Both Sova and TestGorilla assemble an assessment from a catalog the employer chooses from. Sova puts it plainly on its candidate hub: "You will see a variety of activities selected by your potential employer for you to complete within the assessment hub." So the honest answer to what the test contains is that the composition varies by employer, and the invitation is the document that resolves it.

Whether a second sitting is possible. SHL's candidate report says results "remain valid for about 12 to 18 months", which makes a retake inside the same recruiting cycle unlikely to be offered. The publisher does not state it as a rule, and no firm publishes a retake policy, so the honest answer is that the first sitting is the one that counts. Our SHL guide carries the validity line and what it does and does not imply.

Whether monitoring is switched on. The publishers document webcam snapshots, browser focus tracking and window switching rules. They also say the client can turn some of it off. Assume it is on and behave accordingly, and treat any page that states it as a fact with suspicion.

What to do between the invitation and the sitting

1. Read the invitation and write down the product name and the window. Two dates go in your calendar: the day the invitation arrived and the last day it is valid.
2. Find the fact sheet for that product and compute your per item budget. Allowed time divided by maximum items gives a pace you can rehearse against a kitchen timer.
3. Do one timed set in each family the assessment names, then write down which family lost you items and which failure mode caused each loss.
4. Repeat only on the failing family. Two or three sessions remove the novelty of the format, and after that a raw percentage measures more than it improves.
5. Set up the room before the day. Charged laptop, wired connection if you have one, pen, paper, calculator if the firm allows it, and every other window closed.

The order matters more than the volume. A candidate who spends four hours on verbal items and loses the test on deductive pacing did the work and skipped the diagnosis.

Where to practice, and what our drills will not do

Start with the publishers, because their material is free and it is theirs. SHL publishes example questions and full length practice assessments for candidates, and four of them return a report. Sova publishes practice tests covering situational judgment, video questions and three ability families. HireVue hosts a public practice interview that is deleted automatically after 30 days. Our assessment practice page lists all four, with what each one does not reproduce.

We have no aptitude drill, and saying otherwise would be easy and false. Our free drills score case work, and how an answer is assessed sets out the weights they use, read from the code that does the scoring. Our assessment simulators cover firm built formats rather than publisher tests, and none of the six is live yet, which that page says itself.

For the publisher specific mechanics, SHL style assessments covers the adaptive behavior and the personality section. For the two families that carry most of the weight, numerical reasoning covers where the time goes on exhibit questions and deductive reasoning covers the guess versus skip decision in detail.