Demand Analysis

Demand analysis is the systematic study of how much of a good or service customers are willing and able to buy at a given price, income level, and set of market conditions, and of why that willingness changes over time. It combines economic theory (the law of demand, elasticity, the determinants of demand) with applied market research (surveys, historical sales and transaction data, econometric modelling) to answer one practical question: how much of this offering will the market absorb, at what price, and under what conditions will that absorption grow or shrink.

In scope: quantifying and explaining current and near-term demand for a specific, clearly bounded product or service category, in a specific geography and customer segment; measuring how that demand responds to price, income, and the price of substitutes and complements (elasticity); and identifying which underlying drivers (demographics, regulation, technology adoption, seasonality) shift the whole demand curve rather than simply move a point along it.

Out of scope, handled by adjacent methods on this site: sizing the total addressable market in absolute currency terms (see TAM, SAM and SOM); assessing how many competitors can profitably serve that demand and on what terms (see Supply Analysis and Porter's Five Forces); projecting demand multiple years forward under named scenarios (see Market Forecast and Scenario Analysis); and compounding a historical growth rate into a single headline percentage (see CAGR). Demand analysis supplies the evidence and assumptions those methods consume; it does not replace them.

• Before setting or changing a price, to estimate how a percentage price change will move unit volume and total revenue (price elasticity of demand).
• When deciding whether to enter a new segment, geography, or channel, to confirm real buyer willingness exists before commissioning a supply-side build.
• When explaining a sales change after the fact: was it a movement along the demand curve (a price change) or a shift of the curve itself (a change in income, substitutes, regulation, or season)?
• As the evidence base feeding a market-sizing exercise (TAM, SAM and SOM), a forecast, or a scenario model: demand analysis supplies the input assumptions those later steps compound forward.
• Whenever a stated preference from a survey or interview needs to be checked against a revealed preference from actual purchase or usage data before either is trusted on its own.

1. Define the good or service and the boundary of the market precisely: what counts as a substitute, and which geography and customer segment are in scope. An imprecise boundary invalidates every number that follows.
2. Gather revealed-demand data: historical unit sales, transaction or point-of-sale data, search volume, or usage logs for the defined category, over as long a time window as is available.
3. Gather stated-demand data: surveys, structured interviews, willingness-to-pay studies, or conjoint analysis, run against a sample representative of the target segment.
4. Identify which determinants of demand were at work in the observed period (price, consumer income, the price of substitutes and complements, tastes and preferences, consumer expectations, and the number of buyers), and separate which ones shifted the curve versus moved a point along it.
5. Calculate price elasticity of demand from at least one clean price change, a natural experiment, an A/B price test, or a historical price move, using the midpoint (arc) formula so the result does not depend on the direction of the change.
6. Segment the analysis: elasticity, income sensitivity, and the shape of the demand curve itself typically differ by customer segment, channel, and geography, so an aggregate number can hide an important underlying split.
7. Reconcile revealed and stated demand. Where the two disagree materially, treat that gap itself as a finding worth explaining before the analysis is used downstream.
8. Document every assumption and its source, including the price range over which elasticity was measured, so the analysis can be re-run when a new price point, competitor entry, or regulatory change invalidates it.

A subscription analytics tool is priced at $49/month and sells 4,000 seats/month. The company raises the price to $59/month; after demand stabilizes over one full billing cycle, monthly seats settle at 3,400.

Step 1, percentage change in quantity (midpoint): (3,400 - 4,000) / ((4,000 + 3,400) / 2) = -600 / 3,700 = -16.2%

Step 2, percentage change in price (midpoint): (59 - 49) / ((49 + 59) / 2) = 10 / 54 = +18.5%

Step 3, elasticity: Ed = -16.2% / 18.5% = -0.87, so |Ed| = 0.87.

Because |Ed| is less than 1, demand is inelastic across this price range: an 18.5% price increase produced a smaller, 16.2%, drop in volume. The revenue arithmetic confirms it: old revenue = 4,000 x $49 = $196,000/month; new revenue = 3,400 x $59 = $200,600/month, a roughly 2.3% revenue gain despite losing 600 subscribers, because the price increase outweighed the volume loss. Had elasticity been greater than 1 in absolute value, the same price rise would have reduced total revenue instead. These figures are illustrative arithmetic built to demonstrate the calculation, not a reported market statistic for any real company.

Not yet available.

• Alfred Marshall, Principles of Economics (1890), the original formulation of the law of demand and the demand curve.
• Corporate Finance Institute, 'Law of Demand' and 'Price Elasticity of Demand' explainers.
• Wikipedia, 'Price elasticity of demand', for the point- and midpoint-elasticity formulas.