Consulting Assessment Center 2026: Who Is Rating You

What a consulting assessment center is

The publisher that builds and sells these days defines it in one sentence. SHL writes that "During an assessment, a group of candidates takes part in a range of exercises, which are observed by assessors. The assessors rate the performance of the candidates against the competencies required for the job." Three things are named there: a group, a sequence, and a standard that belongs to the employer.

The same page says the day runs in either of two places. "Assessment Center Exercises can be delivered face-to-face or using a virtual assessment center platform, during which the exercises are completed digitally." A virtual day is the same instrument on a screen, and the screen adds rules of its own.

Two consequences matter more than anything else on this page, and the publisher spells out both. The day is cumulative: "If there is more than one exercise, you will have other opportunities to show what you can do. If you feel you have performed poorly on one exercise don't give up. Your performance across all of them will be taken into account." One weak hour does not settle the morning.

And trying to reverse engineer the standard is wasted effort. "Don't make assumptions about how you should respond. If you try to guess what the assessors are looking for you may be wrong." That advice sits on shl.com/shldirect, checked 7 October 2026, alongside a line about eye contact, facial expression and gestures in team exercises and role plays.

Who is rating you, and against what

The standard is a competency model, and the publisher sells that too. SHL's Universal Competency Framework has three tiers: "8 general competency factors at the highest level. 20 competency dimensions further explaining the general factors. 96 skills at the most granular level." It defines a competency "in terms of observable behaviors that are instrumental in the delivery of desired results", and it dates itself, with an original release in 2001 and major updates in 2019.

Then comes the line that explains the whole design of an assessment day. The publisher's guidance on building a framework says to "Support each skill with positive and negative behavioral indicators to give concrete targets", and to "Include only skills that can be observed, measured, and assessed consistently." A rating is an observer matching what you did against a written indicator. An exercise exists to produce something observable.

Which is why one day holds several exercises. Here are the eight factors, named in the publisher's guide, with where a day could plausibly observe each one. The right hand column is our reading of the framework rather than anything SHL publishes.

One firm publishes its own list. KPMG sets out seven "Global Behaviours" in full, each written in the first person, which is the form a behavioral indicator takes: "Makes an Impact: I listen to others and present information clearly and concisely. I communicate ideas appropriately and adapt my style and approach to my audience and situation." The other six are Seizes Business Opportunities, Demonstrates Curiosity and Innovation, Drives Collaboration and Inclusion, Strives for Continual Improvement, Delivers Quality and Exercises Professional Judgment. That page is at kpmgcareers.co.uk.

The six exercises the publisher sells separately

SHL published its own exercise taxonomy in April 2025. The names below are the publisher's, and the descriptions condense its text. They are sold as separate products, which is why the composition of a day varies so much between employers.

The taxonomy is at shl.com/resources, dated 8 April 2025. Its product page adds a seventh item that the article leaves out, promising "simulations, group exercises, client meetings, and more", so a day can also contain a simulated client conversation.

Each of those exercises has its own published evidence, and this page does not repeat it. Written case interviews counts the eleven firms that name one. Consulting case presentations collects the published audiences and clocks for a readout. Group case interviews covers the room with other candidates in it. Work simulation assessments handles the inbox.

Which firms publish a consulting assessment center

The day is real, and it is far from universal. Here is what five firms publish about it, each from the firm that owns the claim. Note how little of it concerns exercises.

KPMG is the most specific. Its graduate process ends with "Launch Pad, an in-person, half day event" that "combines assessed and non-assessed activities", held on seven published dates from 17 November 2026 in London to 5 May 2027 in Leeds. The firm commits to a decision: "If successful you will have your job offer within 2 working days." It also commits to feedback on request if you are turned down. The page is at kpmgcareers.co.uk.

PwC UK publishes the day in a position worth noticing. "The penultimate stage of our recruitment process is a virtual immersive assessment centre", and the stage after it is a business assessment that for strategy applicants "will also consist of a series of case study interviews with a senior Strategy& leader". At that firm the case interviews come after the center rather than inside it.

Capgemini Invent publishes a consulting day with no case in it, describing "a half-day engaging event consisting of a group exercise and a 1:1 strengths based interview", plus ice breakers and a presentation about the firm. IBM publishes the day as a branch rather than a step, offering either "an interview (phone, video or in person) or an assessment center visit (group exercise and in-person interview)". And OC&C's London associate consultant posting lists six steps that end in two rounds of case interviews, with no center anywhere in the sequence, checked 7 October 2026.

The platform rules of a virtual consulting assessment center

If the day runs on SHL's platform, the material arrives inside a simulated workplace. The publisher tells candidates the exercises "include an immersive email inbox and document hub", and that "depending on the exercise type, you may be required to create and present a document with your recommendations". So the artifact you hand in may be built in their software rather than yours.

Then comes the rule that changes how you read: "Preparation activities are timed and can only be accessed once."

There is no second pass over the pack. That single sentence is the argument for fixing your structure from the question before opening anything, because the reading you do is the only reading you get. The same page tells candidates to "manage your own time" and to "enter and exit all activities at the correct time", so nobody herds you between rooms. It is at support.shl.com, checked 7 October 2026.

Sova, a second publisher, runs its virtual days with one difference worth planning for. Every part of the day, "including the individual and group exercises", happens online for all candidates and observers, on Zoom, Microsoft Teams or Whereby. It publishes a bandwidth floor of 600kbps up and down, recommends 1.5 Mbps, and advises against a phone because "the screen may be too small to read important materials effectively". Its hub is at sovaassessment.com.

A worked written case, and its sensitivity

Here is an analysis exercise written by us, in the format the publisher describes. A regional distributor has three depots and asks whether to close one.

Contribution first. North gives 12.0 times 0.22, which is 2.64. South gives 9.0 times 0.18, which is 1.62. East gives 6.0 times 0.25, which is 1.50.

Now subtract fixed cost. North makes 2.64 minus 2.40, so 0.24. South makes 1.62 minus 1.80, so a loss of 0.18. East makes 1.50 minus 1.50, so zero. Group profit today is 0.24 minus 0.18 plus 0, which is 0.06, or $60,000.

The pack adds one paragraph on page four: 40% of South's revenue can be served from East at East's margin, and South's fixed cost exits with the depot. That paragraph is the exercise. Retained revenue is 0.40 times 9.0, which is 3.6. At East's margin, 3.6 times 0.25 adds 0.90 of contribution, so East rises to 1.50 plus 0.90 minus 1.50, which is 0.90. Group profit becomes 0.24 plus 0.90, which is 1.14, or $1.14 million.

The recommendation. Close South and serve its retained customers from East, worth about $1.08 million of additional annual profit against today's 0.06.

The risk, quantified. The whole answer hangs on the 40% retention. At 20%, retained revenue is 0.20 times 9.0, which is 1.8, and 1.8 times 0.25 adds 0.45. Group profit lands at 0.24 plus 0.45, which is 0.69. The decision survives, and that sentence is worth more than the first number, because it is the one an observer can see you tested.

The report skeleton, and the ninety minute split

Write the report in the order a reader wants it, which reverses the order you discovered it in.

1. The answer, in one sentence with its number. Close South, about $1.08 million.
2. The two or three facts that force it. South's negative contribution after fixed cost, East's higher margin, the retention paragraph.
3. The arithmetic, laid out so it can be checked. One line per step. An observer rating analytical thinking is looking for a trail.
4. The risk and what would change the answer. The retention threshold, and where it breaks.
5. The next action and who owns it. Concrete, dated, and small enough to be real.

For the clock, work backward from the deadline rather than forward from page one. On a ninety minute exercise, a split that holds is 10 minutes on the question and a hypothesis, 30 minutes mining only what the hypothesis needs, 30 minutes writing, 10 minutes on arithmetic checks, and 10 minutes spare. Keep the spare block. It absorbs the exhibit you misread at minute 40.

The same discipline applies to a slide. Title each one with its message rather than its subject. "Depot economics" tells a reader nothing. "Closing South adds about $1.1m, and the answer holds at half the retention" carries the finding and the risk in one line.

Where a strong candidate loses the day

Reading the pack front to back. The pack holds more material than the question needs, and that surplus is part of the test. Decide what would change your answer, then go and find only that. Reading everything consumes the mining block and eats into the writing block, which is the trade that produces an unfinished report.

Writing an analysis with no decision in it. A report that lists considerations hands the decision back to the observer. Pick one, put a number on it, and say what would change your mind.

Spending the first minute of a group exercise restating the brief. That minute is the cheapest one to claim and restating decides nothing. A worked opener: "We have 25 minutes and three decisions. I suggest five minutes to agree the criterion, then we take the decisions in order of cost. Does anyone want a different order?" It puts a clock on the room, proposes a criterion ahead of any opinion, and hands the floor back. Later, the highest value move is naming what has been settled, because nobody in the room is assigned to keep that ledger.

Running out of energy. SHL warns that virtual assessments "can feel particularly exhausting" and tells candidates to use the breaks. Observers compare your morning with your afternoon, and Adapting and Coping is one of the eight factors, so consistency across a long day is itself a thing being measured.

What nobody publishes about the day

No publisher and no firm publishes a rubric or a pass mark for these exercises. SHL locates the standard inside the employer's competency model, and KPMG's seven behaviors are the only such list we have found in the open, with no weights and no threshold attached to them.

Four more gaps are worth naming so you do not go hunting. The first is the room: how many candidates sit in a group exercise, how many observers watch, and whether the same observers follow you from one exercise to the next. The second is the split between assessed and non-assessed activities, which nobody publishes, including the firm that uses the phrase. The third is the length of a full day, where the only two figures in the public record are two half days, at KPMG and at Capgemini Invent. The fourth is whether the document you build inside a publisher's platform is ever read by a person, while that publisher advertises "automated reports and dashboards" as part of what it delivers.

You can rehearse the interface once for free. SHL opens its virtual assessment center platform to candidates and states the limits itself: "the exercise is not scored, there are no feedback reports and there is no live interaction involved". Familiarity is the whole benefit, and the publisher says so.

For the habits that carry into the exercises, our free drills score structure, exhibits and synthesis, and how an answer is assessed publishes the weights they use, read from the code that does the scoring. Synthesis is where a written case is won, and Answer first carries 35 of 100 in our own scoring. Our assessment simulators cover firm built formats, and none of the six is live yet, which that page states itself. For what comes before the day, aptitude tests in consulting has the product level timing and assessment practice lists what each publisher gives away.

Two corrections to the day described here

7 October 2026. Two corrections, both from reading the sources again rather than from a reader.

The earlier version said that no firm publishes the competency model its assessors rate against. KPMG does, and we had not looked for it. Its seven "Global Behaviours" are now a section of this page. A student facing version of the same list names nine areas and dates itself "For 2019 programmes", so the two are quoted with their own labels and not merged into one history.

One correction on this page was itself wrong, and reversing it is worth recording. On 7 October 2026 we read OC&C's London vacancy page, found no office by office table and no written exercise, and removed the 90 minute Rotterdam written case from the summary. The source this page cites is OC&C's graduate recruitment brochure, a separate document from that vacancy page. It was still online the same day and still prints "1 x 90 minute written case study interview with Associate Consultants" under Rotterdam, so the figure is restored. A fact stops being true when the document that carried it stops carrying it, and we had checked the wrong document.