Wine industry data can tell several different stories with the same numbers. Falling volume and rising sales value might suggest a changing product mix, higher prices or currency effects. They don’t automatically demonstrate healthier demand or better producer profits. The useful work starts after the figures have been copied into the spreadsheet.
For D6, analysis should answer the prescribed research task, using evidence you understand and calculations another reader could follow. You don’t need to manufacture a dramatic trend. A careful explanation of what a dataset can and cannot establish is often more revealing than a confident headline built on incompatible measures.
Define the decision before choosing the dataset
Write down what the question requires you to compare or explain. Production, shipments, retail purchases and consumption are related activities, but they aren’t interchangeable. A harvest estimate won’t establish which packaging consumers prefer, and a national import total won’t show the profitability of one fictional winery’s export strategy.
Source evaluation begins with this match between measure and question. Record the dataset’s population, geography, period, units, definitions and publication version. Add its provisional or revised status where relevant. If the source doesn’t explain a crucial term, investigate before using the figure as the foundation of an argument.
The OIV statistics service provides different products, including preliminary estimates and consolidated statistical publications. Choose deliberately and preserve the version used. A newer release may revise an earlier period; silently joining different versions can create a movement that comes from the editing history rather than the wine industry.
Keep production separate from availability and consumption
Production concerns wine made under the dataset’s stated definition. A volume can be affected by area, yield, weather, harvest decisions and processing. Before offering an explanation, establish which factors the evidence actually addresses. A smaller harvest doesn’t, on its own, prove declining consumer interest.
Consumption can be estimated rather than directly observed. The OIV’s April 2023 sector report explains apparent consumption as a supply-based proxy, using production and trade with inventory adjustments, and discusses limitations where relevant components are uncertain. This is a methodological example from that report, not a statement of current market totals.
Read the notes for the particular series you use. Wine available in a market isn’t necessarily wine physically drunk during the period, and incomplete information about stocks or other uses can affect an estimate. Don’t translate an apparent-consumption figure into a precise claim about individual drinking behaviour without suitable evidence.
Put every quantity into comparable units
Check litres, hectolitres, thousands and millions before calculating anything. One hectolitre equals 100 litres. A table labelled millions of hectolitres uses a different scale again. Keep the original unit in a source column and show any conversion in a separate calculation so a later correction is manageable.
For an original fictional example, a shipment of 12,000 litres is 120 hectolitres or 16,000 bottles of 750 millilitres. Those bottle equivalents describe volume, not necessarily the number of physical bottles shipped: the shipment might be bulk wine. Name the conversion accurately rather than let an illustrative unit become an invented packaging fact.
Likewise, distinguish litres of wine from litres of pure alcohol and vineyard area from productive wine-grape area. They answer different questions. Numerical precision can’t repair a mismatched definition. Before comparing two tables, write a sentence explaining why their units and coverage genuinely permit the comparison.
Distinguish the size of a change from its rate
In another fictional example, annual sales increase from 100,000 to 120,000 units. The absolute increase is 20,000 units; the relative increase is 20%, calculated by dividing the change by the original value. Both descriptions are correct, but they communicate different things. Include the starting scale when it affects the interpretation.
If a market share rises from 20% to 25%, it increases by five percentage points. Relative to the original share, that is a 25% increase. Calling it simply a five percent increase gives the wrong relative change. Keep the wording consistent with the calculation, especially when comparing shares with growth rates.
Watch the denominator. A producer can gain share in a shrinking market while selling less wine. A small category can record a high growth rate yet add relatively little volume. Explain the numerator and denominator rather than assume a large percentage always indicates a large commercial opportunity.
Interpret value and volume together
Dividing total sales value by total volume gives an aggregate unit value when the measures have compatible coverage. In a fictional case, value of 240,000 currency units divided by 120,000 litres equals two currency units per litre. Keep currency, trade stage and period visible. An export unit value isn’t automatically a retail shelf price.
A rising unit value can reflect changes in prices, quality tiers, packaging or the share of different products. It doesn’t prove that the same wine commanded a higher price, or that consumers consciously traded up. To investigate those explanations, seek evidence that separates composition effects from changes within comparable products.
Profit requires cost information too. Higher nominal revenue can accompany higher costs, and foreign-currency movements can affect reported value. State whether the series is in current prices, adjusted prices or another specified basis. Don’t attempt an inflation adjustment with an unrelated index and then present the result as an authoritative real-terms measure.
Choose a meaningful period and baseline
Two endpoints can conceal considerable variation. Look at intervening years and explain why the period fits the task. A rebound after an unusually low year may look spectacular as a percentage without establishing a sustained expansion. If the series changes definition halfway through, the apparent trend may need a qualification or a different dataset.
An index can help compare relative movements. The House of Commons Library’s guide to time-series indices explains the use of a common base. For a fictional sales series starting at 80 units and ending at 100, setting the start to 100 gives an endpoint index of 125.
That index shows a 25% increase from the base, not the absolute size of sales. Two differently sized markets can both reach 125 after growing by the same proportion. Explain what the index makes comparable and what it hides. A change from index 120 to 125 is not a five percent relative increase; it is about 4.17%.
Look beneath the average
An industry total can conceal different regional, channel and price-segment movements. Check whether the argument concerns the whole market or a particular part. A national average may be poor evidence for a small producer selling mostly through cellar-door visits. Break down the data where suitable evidence exists rather than speculate about unseen segments.
For aggregate unit values, divide total value by total volume. Don’t take a simple average of country unit values when their volumes differ and call it the overall unit value. The weighting changes the answer. Explain the method so the reader can tell whether the calculation describes an average country or an average litre.
Treat missing observations honestly. A blank cell is not automatically zero, and an unavailable estimate isn’t evidence that no activity occurred. Keep missing values distinct, check notes and avoid filling gaps with convenient assumptions. If the comparison cannot be made consistently, narrow it or explain why it remains uncertain.
Separate patterns from explanations
Ideas and arguments require more than showing that two lines move together. A rise in online wine sales alongside a marketing campaign doesn’t establish that the campaign caused the rise. Distribution changes, wider shopping habits, promotions or altered measurement could also matter. Identify plausible explanations and seek evidence capable of distinguishing them.
Use cautious causal language where the design cannot establish causation. A pattern may be consistent with an explanation without proving it. Case evidence can help explain a mechanism, but one business’s experience doesn’t automatically account for an entire national series. Combine the measures with relevant context and keep the inference within its limits.
Informed conclusions should distinguish the observed result, your interpretation and the remaining uncertainty. For example, a fictional dataset may establish falling export volume and rising unit value while leaving the causes unresolved. That supports a precise finding and a focused research gap, rather than a sweeping assertion about every producer’s strategy.
Make calculations readable and reproducible
Keep raw data separate from formulas and retain source details, units and versions. Check a sample calculation manually. State rounding conventions and avoid displaying more decimal places than the evidence warrants. If a formula uses a zero starting value, ordinary percentage growth is undefined; don’t substitute a dramatic percentage.
Use a small table or chart when it clarifies a comparison, with a descriptive title, units, period, sources and explanatory notes. A chart should support reasoning already explained in the text. Follow the current brief for presentation and word-count treatment rather than assume that moving an argument into a figure removes it from the assignment’s requirements.
Before submitting, recalculate every number that appears in the conclusion from the retained source table. Check that an updated chart hasn’t left an older figure in the prose. This simple reconciliation catches a surprisingly ordinary research problem: each component may have been correct at a different stage, while the final document no longer agrees with itself.
Check your working process against building a research argument and evaluating research sources as linked checks: identify the evidence each claim needs, preserve source limitations and confirm that the final judgement still answers the authorised task.
Original Diploma practice
These original exercises use invented figures and situations. They are independent study practice, not official WSET questions, marks or assessed research findings.
Exercise 1: calculate the change
Fictional sales rise from 80,000 to 100,000 litres. Give the absolute change, relative change and final index when the first year equals 100.
Indicative points: difference; denominator; percentage; index; units; meaning.
Outline answer: The increase is 20,000 litres, or 25% of the starting value. The final index is 125. The index describes relative movement and doesn’t retain the original market size. Keep the calculation separate from any explanation of why sales changed.
Exercise 2: distinguish shares
A fictional category’s share rises from 10% to 15%. Explain both forms of change and one missing piece of context.
Indicative points: percentage points; relative change; denominator; market size; volume; interpretation.
Outline answer: The increase is five percentage points and 50% relative to the original share. The size and movement of the total market are needed to interpret the category’s absolute sales. Gaining share doesn’t by itself establish rising volume.
Exercise 3: interpret unit value
A fictional exporter reports rising value and falling volume. A draft calls this proof of greater profitability. Evaluate that conclusion.
Indicative points: compatibility; division; composition; currency; costs; limits.
Outline answer: Check the coverage and calculate compatible aggregate unit values. Changes can reflect prices, mix or currency. Profitability requires suitable cost evidence, so the reported movements alone don’t prove that margins improved.
Exercise 4: handle a gap
One year is missing from a fictional consumption series and another is provisional. How should you present the trend?
Indicative points: missingness; zero; revision; version; chart; conclusion.
Outline answer: Preserve the missing observation as missing, identify the provisional estimate and retain the source version. Don’t invent a value or draw an unqualified continuous trend through the gap. Explain how the uncertainty affects the comparison and any recommendation.
