Sales Forecasting and Break-even

Two calculations sit under "sales" on this paper, and they are not interchangeable. Sales volume is simply the number of units sold in a period — a count, no currency sign. Sales revenue is what those units were sold for: sales volume × selling price, in whatever currency the extract uses. A firm can grow one without the other — a price cut can raise sales volume while cutting sales revenue, and a premium price rise can do the reverse — which is exactly why the spec (2.3.2.1a) lists them as two separate calculations rather than one.

Cost sits on the other side of the same ledger, split the same way this course's Economics papers split it, but without the marginal-cost machinery WEC13 builds on top of it — this spec item (2.3.2.1b) stops at fixed, variable, total and average. Fixed costs (TFC) don't change with output — rent, insurance, a salaried manager's pay, owed in full even if the business sells nothing all month. Variable costs (TVC) rise with output — raw materials, piece-rate labour, packaging — £0 at zero output, and roughly proportional to units made. Total cost is TC = TFC + TVC, and average cost is AC = TC ÷ output — the cost of one "typical" unit, useful for comparing against selling price to see whether a unit is even worth making.

Improving sales volume and sales revenue (2.3.2.1c) usually means pulling one of a small number of levers: cutting price to sell more units — which only raises revenue if demand is elastic enough that the extra volume outweighs the lower price per unit, a WBS11 demand-and-elasticity idea this paper is explicitly allowed to draw on; heavier promotion or a wider distribution network reaching more potential buyers; extending the product range so existing customers buy more often; or improving quality and reputation so the firm can raise price without losing volume. None of these levers is free — each has its own cost — which is exactly why "ways of improving sales" questions on this paper are typically marked as an Analyse (6-mark) question or higher, not a Define: the mark scheme wants the cost/trade-off reasoning behind the lever, not just the lever's name.

A sales forecast is a prediction of future sales volume or revenue, and its purpose (2.3.2.2a) is entirely practical: it feeds directly into how much inventory to order, how many staff to schedule, what cash-flow forecast to build, and what the business plan itself claims when raising finance. A forecast isn't a target or a hope — it's an input other decisions are built on top of, which is exactly why getting it wrong doesn't just embarrass the forecaster. It cascades into overstocking, understaffing, or a cash-flow forecast that turns out to be fiction.

Three factor categories move a forecast, named directly by the spec (2.3.2.2b): consumer trends — fashions, tastes and habits shifting, sometimes gradually (a slow move toward plant-based food) and sometimes suddenly; economic variables — inflation, interest rates, exchange rates and the wider business cycle changing how much disposable income and confidence customers actually have; and the actions of competitors — a rival's price cut, a new entrant, or a competitor's marketing campaign can move a forecast that had nothing wrong with its own internal logic. None of these three is under the forecasting firm's own control, which is precisely why forecasting is a genuinely different skill from calculating: the numbers on one side of the equation are simply unknowable in advance.

The difficulties of sales forecasting (2.3.2.2c) follow directly from that: forecasts are least reliable exactly where a business most needs them — for a genuinely new product with no sales history to extrapolate from (see Amara's Bakes in the level-exemplar below), in a market moving through sudden shocks, or wherever consumer trends are shifting faster than historical data can capture. Sales forecasting is not the same skill as market research — market research gathers information about what customers currently want; sales forecasting predicts what they will actually buy, at what volume, in the future — a confusion the examiner reports flag directly (see the trap below).

Every margin of safety calculated so far in this lesson has come from a single actual or budgeted sales figure set against a single break-even point — but a real business rarely sells exactly the same amount in every hour, every day or every month, and a margin of safety calculated as an average or a total across a wider period can tell a genuinely different story from the same figures broken down into their components. Pearson's own October 2023 paper (Q1(d), an 8-mark Discuss question, verified directly against the primary mark scheme) is the clean real-world illustration: Arditi Tours, a bus company, ran four scheduled journeys a day, each needing 17 passengers to break even. Its own extract data gave four actual passenger counts for one day — two comfortably above 17, two below it. Averaged across all four journeys, actual passengers came to 19 a journey — a small positive margin of safety of 2, exactly matching the mark scheme's own stated figure ("a margin of safety of two passengers because the actual average number of passengers... is 19"). Added up as a whole day's total against the whole day's combined break-even (4 × 17 = 68), the same data gives a surplus of 8 passengers — again matching the mark scheme exactly ("a margin of safety of eight passengers when spread over the whole day"). Both of those numbers are genuinely true, and neither is a mistake — but neither one shows that two of the four journeys, looked at individually, carried fewer passengers than their own break-even point and, in the mark scheme's own words, "would have made a loss" that specific day.

The general point survives the swap to any other business: averaging or aggregating a margin-of-safety figure is a genuine analytical choice, not a neutral default, and the choice changes which of two different questions gets answered. "Is this business safely above break-even overall, across the period as a whole?" and "is every part of this business — every time slot, every batch, every month — safely above break-even on its own?" are different questions, and a business can score comfortably on one while failing the other. A chronic loss on a specific recurring component (the same time slot, the same product line, the same month every year) matters even when the aggregate figure looks safe, because it's still revenue that isn't covering its own share of fixed costs — and if enough of a business's activity falls into that category for long enough, that's precisely the mechanism (2.3.3.3a, poor management of cash flow as an internal cause of business failure) behind a firm's fixed costs going unpaid, not a separate, unrelated risk.

Break-even analysis works cleanly because it makes real assumptions the spec expects you to be able to name and evaluate (2.3.2.3f), not because business costs and revenue are actually always this well-behaved. First: selling price and variable cost per unit are assumed constant at every output level — no bulk-buying discount pushing VC down once volume rises, no price cut needed to shift extra units — which is exactly why TC and TR are drawn as straight lines on the chart rather than curves.

Second: every unit made is assumed sold — there's no unsold inventory sitting between the production line and the revenue line, an assumption that gets shakier the further a forecast is from reliable (see above). Third, and easy to miss: fixed costs are only fixed within a relevant range of output, not forever — sell far enough past break-even and a firm may need a second shift, a bigger unit, or new machinery, at which point TFC jumps to a new, higher constant rather than staying flat across the whole chart.

Fourth: it's a snapshot, not a forecast — the moment price, wage rates or material costs change, the whole chart needs to be redrawn from the new numbers; a break-even chart doesn't update itself as conditions change. Fifth: it only works cleanly for a single product, or for a firm that can build a sensible weighted-average contribution figure across a genuinely mixed product range — a firm selling ten different products at ten different contributions doesn't have one break-even point, it has ten, unless the model is deliberately simplified to combine them.