Staffing and Organisational Design

The spec lists staff as an asset and staff as a cost as two approaches, and it's tempting to read that as two opposing opinions about whether staff are 'good' or 'bad' for a business. They aren't opinions — they're two different ways of doing the same investment calculation, and which one a firm applies is genuinely derivable from how it treats the RETURN side of spending on people.

Any spend a firm makes can be evaluated one of two ways: as a pure expense (money out, nothing counted coming back in against it) or as an investment (money out, with an expected future return weighed against it). Treating staff as a cost applies the first lens to wages, training and headcount — minimise the spend, because nothing offsetting is being counted. Treating staff as an asset applies the second lens — training spend, higher wages or job security are expected to generate a return (higher productivity, lower recruitment and retraining costs from reduced turnover, deeper firm-specific skill) that exceeds what was spent, the same logic a firm applies when deciding whether to buy a new machine.

This is why the two views make opposite predictions about the same real decisions. A firm that sees staff as a cost minimises headcount, keeps training spend to the legal minimum, and leans on flexible/temporary labour to avoid paying for hours it doesn't strictly need. A firm that sees staff as an asset invests in training even when the immediate output gain is hard to measure, offers job security to protect the retention of skills it has already paid to build, and treats a resignation as a genuine loss of capital, not just a vacancy to refill. Neither view is simply 'wrong' — which one fits a given business depends on how replaceable the skill actually is: a role any new hire can do competently within a day suits the cost lens; a role that takes years of firm-specific learning to do well suits the asset lens, because that's precisely where training spend has time to earn its return before the person could plausibly leave.

The spec names five flexible-workforce approaches — multi-skilling, part-time and temporary staff, zero-hours contracts, flexible hours and home working, and outsourcing — and the single highest-value thing to get right about all five is that they are not interchangeable synonyms for 'flexible staffing.' Each one solves a genuinely different mismatch between labour supply and the pattern of demand a business actually faces, and the exam has confirmed, more than once, that candidates who blur them into one generic category lose marks for exactly that reason (see the trap-taxonomy below).

Multi-skilling trains one worker to do several jobs, so the SAME headcount can be redeployed across tasks as demand for each task shifts through the day or the week — it solves a skill-coverage problem, not an hours problem, and it's the only one of the five that doesn't change a single person's contract at all. Part-time and temporary contracts solve two different problems that share the word 'flexible': part-time fixes a REDUCED but still guaranteed number of weekly hours, suited to demand that is regular but below full-time level (e.g. a role only needed three days a week, every week); temporary fixes a limited DURATION with an end date, suited to demand that is real but time-bound (a seasonal peak, covering a colleague's parental leave, a defined project).

A zero-hours contract guarantees no hours at all — the worker is offered, and paid for, only the specific shifts the business actually needs, which is the tool that fits demand too unpredictable to commit to any fixed weekly pattern in advance. Flexible hours and home working change WHERE or WHEN contracted hours are worked, not how many hours exist or how long the contract lasts — this solves a different problem again: matching a role to an employee's own constraints (commuting, caring responsibilities, personal productivity patterns) without changing the underlying headcount or hours commitment at all. Outsourcing doesn't touch a business's own employment contracts at all — it removes the function from direct employment entirely, paying an external specialist a contracted fee instead, trading the fixed cost and management responsibility of employing people in-house for a variable cost to someone else's workforce.

Correctly NAMING which of the five tools applies to a scenario, the mechanism above, only answers a Define- or Explain-level question. Discuss, Assess and Evaluate questions ask a different thing entirely — what does USING this tool actually do to the business — and two independently verified mark schemes (Jun 2024 Q2d, Market Kurly, on part-time employees; Jan 2023 Q3, Toyota, on flexible working generally) confirm the same recurring set of costs, benefits and risks across the tools, not just the naming test.

On the cost side, the saving is real and direct, and it isn't only about paying for fewer hours: 'a major benefit of using part-time employees is lower costs. Often part-time employees receive fewer, or no benefits compared to permanent staff' (Jun 2024 Q2d) — the saving includes the non-wage costs (pension contributions, paid leave, sick pay) that come bundled with a full-time contract, not just the wage itself. Home working adds a second, unrelated cost saving: 'employees working from home can reduce costs... as it can save on the cost of office overheads' (Jan 2023 Q3), plus a productivity gain from cutting the employee's own commute.

On the benefit side, flexible arrangements can genuinely raise motivation and widen who is willing to work for the business at all, not just cut cost: 'using part-time employees may attract employees who have other commitments and do not wish to work on a full-time basis. This may improve the work life balance of employees and result in motivated and productive staff' (Jun 2024 Q2d). The same logic extends to choosing hours rather than just their number: 'allowing the employees to choose working hours is aimed at balancing personal needs and job requirements. This may help [the firm] retain its best employees, which may reduce recruitment costs' (Jan 2023 Q3) — a RETENTION benefit, distinct from the attraction benefit.

None of this is one-sided, and the same two mark schemes name the recurring risks just as explicitly. Staff who spend less time in the business know it less well: 'part-time employees may have less knowledge and familiarity with the company. This can affect employee performance and impact the efficiency' (Jun 2024 Q2d). A more flexible workforce can also turn over faster and lose commitment: 'recruitment costs may be higher if a business has a more flexible workforce as the turnover of employees may be high,' and staff who specifically want full-time security may respond to a flexible contract with reduced commitment, 'leading to [the firm] losing workers to competitors and facing disruption to customer service' (both Jun 2024 Q2d). Home working carries a distinct risk that has nothing to do with hours or commitment: 'home working could result in lower productivity if staff abuse the trust and freedom given to them' (Jan 2023 Q3) — a monitoring problem specific to work the employer can't directly observe — plus a coordination cost every flexible arrangement adds on top: 'there may be additional administrative work involved in setting up and running flexible working arrangements' (Jan 2023 Q3).

One final, genuinely generalisable point both mark schemes converge on independently: none of this suits every job equally. 'It is likely that... certain flexible working arrangements may be difficult to achieve. Factory employees will generally need to work specific hours to meet production requirements' (Jan 2023 Q3) — flexible hours and home working assume a job that CAN be done at a time or place of the employee's own choosing, which fits office and knowledge work far more readily than a fixed-shift production line, a warehouse pick-and-pack schedule, or a shop floor that has to be staffed exactly when customers are in the building. The same mark scheme states the conclusion directly: 'flexible working may be more appropriate for employees working in offices rather than those working in car factories.'

Dismissal and redundancy both end with an employee leaving, which is exactly why they get confused — but they answer a different question, and the test for which one applies is simple: does the ROLE still exist, for someone else, the day after the person leaves? Dismissal is termination for a reason tied to the individual — misconduct, poor performance, an inability to do the job — and the role itself carries on, ready for whoever fills it next. Redundancy is termination because the ROLE itself is no longer needed by the business at all, for any occupant, for a structural or business reason (the function is automated, the department is closed, demand for that work has permanently fallen) — nothing about the specific person's conduct is the cause. Getting the dismissal side wrong carries a real legal and reputational cost, not just a definitional one: a real mark scheme (Oct 2021 Q2c, a bank restructuring by redundancy) credits the observation that a poorly-handled process risks 'unfair dismissal cases,' which 'can be costly and may impact on the bank's profitability' — treating every job loss as a simple redundancy when it's actually contested as an unfair dismissal is a genuine financial and legal exposure, not just a vocabulary slip.

Individual and collective bargaining describe who is on the employee's side of a pay negotiation, and the mechanism that separates them is bargaining power, not just headcount. A single employee negotiating alone has one real fallback if talks fail: look for another job elsewhere — and the employer's fallback is just as easy: hire someone else for the same role. Neither side loses much by the other walking away, which caps how much a lone employee can realistically extract. Once a group of employees bargains as one unit (commonly through a trade union), the employer's fallback disappears — it can no longer simply replace one dissenting worker without effectively confronting, and potentially losing, the whole group at once. That change in what the employer stands to lose from a breakdown, not the number of people in the room, is what raises collective bargaining power above the sum of individual negotiations.

That's the mechanism from the EMPLOYEE side — but a real exam question can just as easily ask the reverse: why would the EMPLOYER actively prefer to negotiate collectively? A verified mark scheme (Oct 2021 Q2c: 'Analyse two possible benefits for Deutsche Bank of using collective bargaining to negotiate redundancy payments with its employees') credits a genuinely different mechanism for this side of the same transaction. First, speed at scale: negotiating one agreement with employee representatives is a single conversation, however hard-fought, while negotiating individually with a workforce spread across multiple countries is thousands of separate conversations — the mark scheme credits that this 'is likely to have made the process quicker than using an individual approach, allowing [the firm] to focus on the restructuring plan' rather than being consumed by negotiation logistics. Second, legal-risk reduction: a single, consistently-applied collective agreement is harder to challenge case-by-case than thousands of individual settlements, which is exactly the mechanism behind the 'unfair dismissal' cost named above — collective bargaining was credited as being used specifically 'to protect its reputation and limit the number of unfair dismissal cases.' Neither of these is the employee-bargaining-power mechanism above — they're reasons a rational EMPLOYER volunteers to negotiate collectively even though it strengthens the other side's hand, which only makes sense once the employer's own administrative and legal savings are counted too.

Filling a vacancy has two genuinely different sourcing routes, and the choice between them is a real trade-off, not a default. Internal recruitment promotes or transfers an existing employee into the role — the business already has direct evidence of that person's performance, they already know the firm's systems and culture, and the process is faster and cheaper because there's no need to advertise externally or pay agency fees. Its real cost is what it doesn't solve: the role the internal candidate vacates still needs filling, so internal recruitment often only delays, rather than avoids, an external recruitment cost — and it narrows the pool of applicants to people the business already employs, which can mean missing a stronger candidate, or a genuinely new perspective, available outside the firm. External recruitment opens the vacancy to the whole labour market — a larger pool, potentially new skills and ideas the existing workforce doesn't have — at the cost of the advertising, agency and selection expense of reaching and assessing candidates the business has no track record with at all.

Whichever route is used, the process itself has real, additive costs the spec names directly: advertising the role (job boards, trade press, social media), any recruitment agency fee (often set as a percentage of the role's annual salary — see the worked calculation below for exactly how that adds up), the staff time spent shortlisting and interviewing, and the cost of the selection method itself (a full assessment day for a senior role is a materially different cost from a single interview for an entry-level one). None of this stops once someone accepts the job — training is the final, and often largest, cost component, and the spec names three distinct types worth being able to name separately, because the exam has confirmed (in a different but structurally similar case — test marketing vs product trials, in the market-research topic) that adjacent, similar-sounding terms get marked wrong when used interchangeably.

Induction training happens before or right at the start of employment, and covers what every new starter needs regardless of role — health and safety, company policies, where things are and who's who — it is about the FIRM, not the specific job. On-the-job training happens while the employee actually does the real work, typically shadowing or being coached by an existing employee at their normal workstation — cheap relative to sending someone away, and immediately relevant, but it also means a trainer's own productivity and the training itself are both happening at once, and any mistakes the trainee makes happen on real output. Off-the-job training happens away from the employee's normal workstation — a course, a conference, an external qualification — which costs more directly (course fees, travel, and the output lost while the employee isn't at their post at all) but delivers depth and transferable qualification that on-the-job coaching usually can't match.

Neither method's benefit is unconditional, and a real 20-mark essay confirms two distinct risks worth naming explicitly rather than treating either method as automatically effective. A verified mark scheme (Oct 2021 Q3, Japanese businesses training staff in AI skills) credits that on-the-job training's own success 'depends on the skills of existing staff and their expertise' — a firm whose current employees don't yet have the knowledge the trainee needs to acquire cannot manufacture that expertise just by pairing people up; on-the-job training transfers whatever skill already exists in the business, and no more. Off-the-job training avoids that ceiling by importing expertise from outside, but the same mark scheme credits a distinct cost that has nothing to do with money: sending graduates away for a year of external training means 'a time lag in benefitting from its training programme,' which can leave a business 'further behind' competitors during the gap before the trained employee actually returns to work. A firm choosing between the two isn't just weighing direct cost against depth of learning — it's weighing whether the skill it needs already exists somewhere inside the business against how much a DELAY in acquiring it from outside actually costs while competitors don't wait. The same real essay also credits offering competitive financial incentives (higher salaries) as a genuine recruitment tool for attracting scarce, in-demand skills — but one that trades directly against the cost side: 'businesses... paying higher wages will have increased costs and potentially lower profit margins,' which may 'reduce the ability to invest' elsewhere. Attracting talent with pay and training existing staff from within are two different answers to the identical 'we don't have this skill yet' problem, each with its own cost structure.

The spec lists tall, flat and matrix structures as three types to know, and the easy way to learn them is as three unconnected shapes to memorise from a diagram. The mechanism-first way is to notice that tall and flat aren't really two different categories at all — they're two ends of a single, calculable relationship between span of control (how many people one manager directly supervises) and hierarchy (how many layers of management exist between the top and the bottom of the business). Once you fix the total number of people who need managing, the span of control at every level determines exactly how many management layers are needed to reach all of them — not roughly, exactly, as the worked derivation below shows with real numbers.

Chain of command is simply the specific path that authority (and information travelling the other way) has to pass through, level by level, from the most senior manager down to the most junior employee — it's the concrete, named-people version of the abstract hierarchy-levels count. Centralisation and decentralisation answer a genuinely different question from span of control and hierarchy height: not how many layers exist, but WHO makes the decisions — reserved for senior management at the top (centralised) or delegated down to managers and staff closer to where the decision actually needs making (decentralised). The three ideas often move together in a real business — a flatter structure frequently goes hand in hand with more decentralised decision-making, because a manager with a wide span of control physically cannot closely direct everything every one of their many direct reports does — but they are three separate dimensions of design, not three words for the same thing, and treating them as interchangeable is a real, confirmed source of lost marks (see the trap-taxonomy below).

A matrix structure breaks the single-chain-of-command assumption underneath tall and flat altogether: an employee reports to two lines of authority at once — typically a functional manager (e.g. head of marketing) AND a project or product manager — rather than one line manager up a single hierarchy. It solves a different problem from span of control (how many people to supervise) or centralisation (who decides): it lets a business organise around both a permanent function and a temporary or cross-cutting project simultaneously, at the cost of every matrixed employee having two bosses who can, in principle, give them conflicting priorities. The upside that cost buys is real and confirmed directly in a verified mark scheme (Jun 2023 Q3, on a business moving AWAY from a matrix structure): 'one advantage of having a matrix organisation is that it encourages the sharing of ideas and good practice' between people from different disciplines who would never otherwise work together at all — a genuine benefit a business gives up, not just a reporting-line inconvenience it escapes, when it flattens out of a matrix into a single-chain structure.