Written Exercise: Assessment Centre Case Study Practice

The marker reads the first paragraph properly and skims the rest

That is not cynicism about assessors; it is what happens when one person marks 8 scripts of 800 words each against a competency grid, having already run 2 other exercises that day. The opening paragraph is read closely, the structure is scanned, and the body is read for whether it delivers what the opening promised. A script whose argument only becomes clear on page 2 has already lost the reader it needed.

Which produces the single most useful rule in the whole exercise, and the one most candidates violate: put the recommendation in the first 3 sentences. Not the background. Not the context. Not a restatement of the brief. The answer.

This feels wrong to anyone trained in academic writing, where you build evidence toward a conclusion and the conclusion is the reward for reading. Business writing inverts it, and assessment centre written exercises are business writing being marked by people who do it for a living. Candidates who lead with context are not being penalised for style — they are failing the "communicates clearly in writing" indicator, which in most frameworks explicitly references getting to the point.

What written exercises measure

Written communication, analytical reasoning, judgement under incomplete information, prioritisation, and — depending on the employer — commercial or organisational awareness. It is the only exercise of the day where your reasoning is captured verbatim and can be re-read, argued about in the wash-up meeting, and quoted back to you.

Indicators tend to include:

• States a clear recommendation and sticks to it
• Uses evidence from the case rather than general assertion
• Distinguishes fact from inference from assumption
• Identifies risk and proposes mitigation
• Structures the document so a reader can navigate it
• Writes in the register appropriate to the named audience
• Finishes — a complete short answer beats a truncated long one

The register point is worth dwelling on because it is cheap marks. If the brief says "write a briefing note to the Director of Operations," then the document has a subject line, a recommendation, and short labelled sections. If it says "write an email to the client," it is shorter, warmer and has no headings. Candidates who write the same essay regardless of the stated audience miss an indicator that required no additional thinking.

The "finishes" indicator deserves the same attention, because it is the one candidates are most surprised by. Markers are not scoring effort, and an unfinished document is evidence of a planning failure rather than of ambition — on several published frameworks, completing the task within the constraints given is a named behaviour in its own right. A note that stops mid-sentence at the word limit tells the assessor that the candidate could not size the task, which is precisely the judgement the exercise exists to test.

How it differs from an in-tray exercise

This confusion costs candidates real marks, so it is worth being precise. An in-tray or e-tray exercise gives you many small competing items and scores how you triage: what you do first, what you delegate, what you defer, and why. A written exercise gives you one substantial case and scores how deeply you analyse it.

In-tray rewards breadth and speed. Written rewards depth and structure. A candidate who brings in-tray habits to a written exercise produces 8 shallow observations where 3 developed ones were wanted, and the score reflects it.

The reverse error also exists and is rarer but costlier. A candidate who brings written-exercise habits to a 45 minute in-tray will analyse item 1 beautifully and never reach items 9 through 18, which on an e-tray platform that timestamps every action is unmistakable in the data. If the day contains both — and at many graduate centres it does — treat them as genuinely different tasks requiring different gears, and check which one you are in before you start writing.

One practical tell: count the pages. A pack of 6 to 8 pages with a single question at the end is a written exercise. A screen with an inbox and 15 unread items is an in-tray, whatever the invitation called it.

Common formats

Business case analysis. A pack of 4 to 10 pages — a scenario, some financials, 2 or 3 options — and a request for a recommendation. The default in commercial hiring.

Policy or public-sector case. Competing stakeholder interests, a statutory or budgetary constraint, and usually an explicit requirement to consider impact on service users. Scored against published behaviours.

Data-heavy analytical exercise. Tables, a chart or two, and questions that require you to actually compute something. Arithmetic errors are noticed here in a way they are not elsewhere.

Summary or briefing exercise. A long document to be condensed to 1 page for a named senior reader. Deceptively hard — it scores judgement about what to omit, and omission is invisible work.

Report with a defined structure. The brief specifies sections. Follow them exactly; deviating is scored as a failure to follow instruction, not as creativity.

Time allowances run from 45 minutes to 2 hours, most commonly 60 minutes or 90 minutes. Word limits, where given, are usually 500 words to 1,200 words.

A worked case extract

Here is a realistic extract from a 90 minute business case exercise — the kind of material you would actually be handed, in the density you would actually get.

BRIEFING PACK — EXTRACT (page 3 of 7)

Thornbury Council: Leisure Services Review

The Council operates 4 leisure centres. The service is required to reduce its net annual subsidy from £2.4 million to £1.6 million within 2 financial years. Current position by site:

| Site | Annual visits | Net subsidy | Subsidy per visit | Building condition | |---|---|---|---|---| | Eastgate | 410,000 | £310,000 | £0.76 | Good — refurbished 3 years ago | | Marsh Lane | 96,000 | £840,000 | £8.75 | Poor — roof replacement due, estimated £1.1 million | | Kingsford | 285,000 | £520,000 | £1.82 | Fair | | Whitton | 71,000 | £730,000 | £10.28 | Fair |

Additional information: - Whitton serves the 2 most deprived wards in the district. 62 percent of its users are on concessionary rates. It is the only site with a hydrotherapy pool, used by 340 people a week under NHS referral. - Marsh Lane is within 3 miles of Kingsford, which has spare capacity of approximately 40 percent at off-peak times. - Closing any site requires a 12 week statutory consultation. - A leisure trust operator has made an unsolicited approach offering to run all 4 sites under a 10 year contract, projecting a subsidy reduction of 45 percent by year 3, with no capital contribution.

Your task: Write a briefing note of no more than 1,000 words to the Director of Place, recommending how the £800,000 reduction should be achieved.

Read the numbers before reading on. The obvious answer — close the 2 expensive sites — collapses on contact with the additional information. Whitton is the worst performer on subsidy per visit at £10.28 and is also the site with the strongest social case and a clinical service attached. Marsh Lane is nearly as expensive at £8.75, has a £1.1 million capital liability attached, and sits 3 miles from a site with 40 percent spare off-peak capacity. Those two facts are not decorative; they are the exercise.

This is how well-designed cases work. The data points to one answer and the context points to another, and the exercise is scoring whether you noticed the tension and resolved it explicitly rather than picking a side and hoping.

The worked recommendation: an actual opening

Here is the opening of a strong response — written out, not described.

BRIEFING NOTE

To: Director of Place Subject: Achieving the £800,000 leisure subsidy reduction — recommendation

Recommendation. I recommend closing Marsh Lane and transferring its users to Kingsford, which delivers £840,000 of the required £800,000 reduction and avoids a £1.1 million roof liability, while protecting the Whitton hydrotherapy service. This should be paired with a 12 week statutory consultation beginning this quarter, a funded transport offer for displaced Marsh Lane users, and a separate commercial review of Whitton's operating model rather than its closure. I do not recommend the trust proposal at this stage; its 45 percent projection is unverified and a 10 year contract would remove the Council's ability to act on a subsidy target it has only 2 years to meet.

Why Marsh Lane rather than Whitton. On subsidy per visit, Whitton is the weaker site at £10.28 against Marsh Lane's £8.75, and a purely financial reading would close it first. Three factors reverse that. First, Whitton's hydrotherapy pool serves 340 people a week under NHS referral and has no identified alternative provision in the district, so closure transfers cost to a health partner rather than removing it. Second, 62 percent of Whitton users are on concessionary rates in the 2 most deprived wards, which makes closure the option with the highest equality impact and the highest consultation risk. Third, Marsh Lane carries a £1.1 million capital liability that no scenario in this pack funds — the roof has to be paid for or the site has to close, and the decision is better taken now than in 18 months.

What makes this deliverable. Kingsford is 3 miles away with approximately 40 percent spare off-peak capacity, which is sufficient to absorb a material share of Marsh Lane's 96,000 annual visits if off-peak programming is rescheduled. I have assumed a retention rate of 60 percent of transferred users; at that level the combined Kingsford position still improves. This assumption is the main sensitivity in the recommendation and I would want it tested before consultation opens.

Stop and notice what that does inside roughly 350 words.

The recommendation is in sentence 1, with the number attached to it. A marker who reads no further knows the answer and the amount.

It names what it is not recommending, and why. Declining the trust proposal explicitly is worth more than ignoring it — an unaddressed option in the pack reads as an option not noticed.

It confronts the inconvenient number directly. "A purely financial reading would close it first" concedes the strongest argument against the recommendation in the candidate's own words, in the second paragraph. Assessors score this as judgement; candidates who quietly avoid Whitton's £10.28 are assumed not to have seen it.

It labels its assumption as an assumption. "I have assumed a retention rate of 60 percent … this is the main sensitivity" hits the distinguish-fact-from-inference indicator, which almost nobody hits, and it costs 1 sentence.

It uses the pack's own figures. £840,000, £1.1 million, 340 people, 62 percent, 3 miles, 40 percent. Every load-bearing claim is traceable to page 3. Nothing is asserted from general knowledge about leisure centres.

It is the right register. Subject line, bold labelled sections, no preamble, no throat-clearing about the importance of leisure provision.

Compare that with the opening most candidates write: "Thornbury Council faces a significant challenge in reducing its leisure subsidy. This note will consider the options available and make a recommendation." Sixty words that contain no information and have already burned the paragraph the marker reads most closely.

Planning: the split that works

For a 90 minute exercise with a 1,000 word limit, a reliable allocation is:

• 20 minutes reading and annotating. Not reading passively — marking every number you might use and every constraint.
• 15 minutes planning. Decide the recommendation, write the section headings, and write your opening sentence in full.
• 45 minutes writing.
• 10 minutes checking. Arithmetic, the word limit, and whether the document answers the question asked.

Thirty-five minutes of not-writing out of 90 feels extravagant when the clock is running, and it is the difference between a structured note and a stream of observations. Candidates who start writing at minute 5 produce documents that change their mind halfway through, which is visible and is scored.

Scale it proportionally: for a 60 minute exercise, roughly 12 minutes reading, 10 minutes planning, 30 minutes writing, 8 minutes checking.

And one hard rule: finish. An incomplete document that stops mid-argument is scored far below a shorter complete one. If you are at minute 75 of 90 and 400 words from the end, cut a section rather than run out of time — a briefing note with 3 developed sections and a proper close beats 5 sections with no ending.

How it varies by sector and level

Consulting and finance exercises are arithmetic-heavy, shorter, and scored substantially on whether your numbers are right. Check them.

Public sector exercises look like the Thornbury case above: statutory constraints, equality impact, consultation requirements, and scoring against published behaviours. Use the framework's vocabulary, and name affected groups explicitly.

Graduate schemes use simpler packs of 4 to 6 pages and weight structure and clarity over depth of analysis.

Management and senior hiring provide genuine strategic ambiguity with no clean answer, longer time allowances of up to 2 hours, and often require the written output to be presented afterwards — in which case write something you can defend out loud.

Technical and analytical roles may require a defined output format — a model, a memo with a specified appendix — where following the format is itself a scored instruction.

Common mistakes

Describing the case back. The most common failure by a wide margin. Summarising page 3 is not analysis; the marker wrote page 3.

Burying the recommendation. Covered at length above.

Ignoring an option in the pack. Every option provided is there to be addressed, including the ones you reject. The trust proposal in the worked case is a trap for exactly this.

Asserting without evidence. "This would improve customer satisfaction" with nothing behind it scores nothing. "This would absorb 96,000 annual visits into 40 percent spare off-peak capacity" scores.

Hedging to the point of having no recommendation. A note that presents 3 options evenly and declines to choose fails the exercise's stated task, however elegant the analysis.

Running out of time. Structural, avoidable, heavily penalised.

Unchecked arithmetic. In a data-heavy case, a wrong subtotal undermines every conclusion that rests on it, and markers check.

Wrong register. An essay where a briefing note was requested.

How to prepare

1. Do 3 timed runs. At least 1 at 90 minutes with a real word limit. Untimed practice teaches you nothing about the part that actually fails.
2. Practise the conclusion-first opening in isolation. Take any decision you have made recently and write a 3 sentence opening: recommendation, headline reason, what you are not recommending. Do it 10 times.
3. Build a reusable skeleton. Recommendation → why, against the strongest alternative → what makes it deliverable and what you have assumed → risks and mitigation → next steps with dates. It fits nearly every brief.
4. Practise writing the sentence that concedes. "A purely financial reading would suggest X" is a high-scoring construction and it needs to be available to you under time pressure.
5. Practise cutting. Write 1,400 words and cut to 1,000. The cutting is where judgement is visible.
6. Do the arithmetic by hand once. If the sector is numerate, get used to computing a subsidy per visit or a payback period without a spreadsheet.
7. Learn the employer's framework vocabulary and use its actual words in your headings where they fit.

Frequently asked questions

TestSolve is independent and not affiliated with any test provider or employer named on this page. All product names and trademarks belong to their respective owners.