Demand Analysis and Forecasting

Learning Objectives

• Define demand and identify the key factors that cause it to change.
• Explain the law of demand and recognize conditions where it may not hold strictly.
• Calculate and interpret price elasticity of demand, and apply it to pricing decisions.
• Compare qualitative and quantitative demand forecasting methods and select the appropriate one.
• Decompose a time series into trend, seasonality, cyclical, and irregular components.
• Evaluate forecast accuracy using error measures such as mean absolute percentage error.
• Apply demand analysis and forecasting to production, inventory, hiring, and marketing decisions.

Quick Answer

Demand analysis studies what drives how much customers are willing and able to buy, while demand forecasting estimates those quantities in the future so that managers can plan production, inventory, staffing, and pricing. A demand forecast is not a guess — it is a structured estimate built from historical data, market knowledge, and assumptions about income, competitor actions, and external conditions. Price elasticity of demand is the most critical analytical tool: it tells a manager whether raising price will increase or reduce total revenue. Good forecasting methods range from expert judgment and consumer surveys to regression analysis and time-series decomposition. Every forecast should be evaluated against actual outcomes.

What Is Demand?

Demand is the quantity of a product or service that buyers are willing and able to purchase at different prices during a period.

Demand depends on:

• Price of the product
• Consumer income
• Prices of substitutes and complements
• Consumer preferences
• Advertising and brand strength
• Population and demographics
• Seasonality
• Credit availability
• Government policy
• Expectations about future prices or income

Demand Curve and Law of Demand

The law of demand states that, other things equal, quantity demanded usually falls when price rises and rises when price falls.

This is a general relationship, not a mechanical certainty. Luxury goods, brand effects, emergencies, and expectations can complicate the response.

Elasticity and Managerial Decisions

Price elasticity of demand measures how responsive quantity demanded is to a price change.

Price elasticity of demand = % change in quantity demanded / % change in price

Demand Type	Meaning	Pricing Implication
Elastic demand	Quantity changes strongly when price changes	Price increases may reduce revenue
Inelastic demand	Quantity changes weakly when price changes	Price increases may raise revenue
Unit elastic demand	Revenue remains roughly unchanged	Price change has balanced effect

Elasticity is crucial for pricing. A manager should not raise price only because costs increased; they must consider how customers will respond.

Real-world examples:

• Petrol/gasoline demand is relatively inelastic in the short run — US drivers and Indian commuters still fill up even at higher prices because they need to travel.
• Restaurant meals tend to be more elastic — consumers switch to cooking at home or cheaper options when prices rise.

Types of Demand Forecasting

Method	Uses	Strength	Limitation
Expert opinion	New products, limited data	Fast and experience-based	Subjective
Consumer surveys	Understanding preferences	Direct customer input	Responses may not match behavior
Sales force estimates	Local market forecasting	Uses field knowledge	Can be biased
Time series analysis	Stable historical data	Uses past patterns	Weak when market changes suddenly
Moving averages	Smoothing fluctuations	Simple and useful	Slow to detect turning points
Regression analysis	Estimating demand drivers	Quantifies relationships	Requires good data and assumptions
Test marketing	New product demand	Real-world evidence	Costly and may alert competitors

Time Series Components

Historical demand may contain:

• Trend: long-term direction
• Seasonality: repeated pattern within a year
• Cyclical movement: broader economic cycles
• Irregular variation: unexpected events

Example: Ice cream demand may rise in summer due to seasonality, while overall premium dessert demand may fall during an economic slowdown. US retailers see clear seasonal spikes around Thanksgiving and Christmas; Indian FMCG companies see demand surges during Diwali.

Forecasting Example

A retailer sells fans. Monthly sales rise during summer and fall during winter.

The manager should not simply average all months. A better forecast separates:

• Base demand
• Seasonal effect
• Expected temperature
• Price changes
• Competitor promotions
• Inventory availability

If last May sales were 10,000 units and this year a hotter summer plus stronger advertising is expected, forecast demand may be adjusted upward. If a competitor cuts prices, the forecast may be revised downward.

Forecast Error

Forecasts should be evaluated. Common error measures include:

• Forecast error = Actual demand - Forecast demand
• Mean absolute error
• Mean absolute percentage error

Managers should track forecast error because repeated overforecasting creates excess inventory, while repeated underforecasting creates stockouts and lost sales. The US Congressional Budget Office (CBO) routinely publishes its forecast errors for GDP and deficits — a model of transparent forecast accountability that business managers can emulate internally.

Managerial Uses

Demand forecasting supports:

• Production planning
• Inventory decisions
• Pricing
• Hiring and scheduling
• Marketing budgets
• Capacity expansion
• Cash flow planning
• Supplier contracts

Key Terms

Term	Definition	Related Concept
Demand	Quantity buyers are willing and able to purchase at various prices during a period	Law of demand, price
Law of demand	As price rises, quantity demanded falls, other things equal	Price elasticity, substitutes
Price elasticity of demand	Percentage change in quantity demanded divided by percentage change in price	Revenue, pricing strategy
Elastic demand	Elasticity greater than 1; quantity responds strongly to price changes	Revenue, substitutes
Inelastic demand	Elasticity less than 1; quantity responds weakly to price changes	Necessities, addiction
Income elasticity	Responsiveness of demand to changes in consumer income	Normal goods, inferior goods
Cross-price elasticity	Responsiveness of demand for one product to a price change in another	Substitutes, complements
Time series	Historical data ordered in time, used to identify trends and seasonal patterns	Forecasting, trend analysis
Forecast error	Difference between actual demand and forecasted demand	MAPE, accuracy tracking
Regression analysis	Statistical method to quantify the relationship between demand and its drivers	Forecasting, elasticity

Common Mistakes

Misconception: Sales figures and demand figures are the same thing. Why it's wrong: Sales are constrained by what the firm had available to sell. If a product is out of stock, recorded sales are lower than actual demand. Using sales data alone underestimates true demand. Correct understanding: Demand is what customers want to buy; sales are what they actually bought. Managers should track stockouts, back-orders, and lost-sale reports to adjust for the gap.

Misconception: If past demand was stable, the same forecast will work in the future. Why it's wrong: Structural changes — a new competitor, a change in consumer preference, a macroeconomic shock — can break historical patterns. Time-series methods that rely solely on the past are blind to structural breaks. Correct understanding: Forecasters should regularly test whether the model still fits recent data and should incorporate leading indicators, market intelligence, and scenario planning alongside historical patterns.

Misconception: A higher price always reduces demand, so raising price is always a bad idea. Why it's wrong: For inelastic products — insulin in the US, cooking gas in India — a price increase reduces quantity demanded only slightly, so total revenue rises. For luxury or status goods, a higher price can even increase perceived quality and demand. Correct understanding: Whether raising price helps or hurts revenue depends on the price elasticity of demand for that specific product, market, and price range. Managers must estimate elasticity before changing price.

Comparison and Connections

Dimension	Qualitative Forecasting	Quantitative Forecasting
Basis	Human judgment, expert opinion, surveys	Historical data, statistical models
Best for	New products, entering new markets, major disruptions	Established products with stable demand history
Speed	Fast	Slower; requires data collection and analysis
Accuracy risk	Subjective bias, overconfidence	Model misspecification, ignoring structural change
Cost	Low to moderate	Moderate to high (data, analytics tools)
Examples	Delphi method, sales force estimates, focus groups	Moving averages, regression, ARIMA models

Practice Questions

Recall

1. List five factors other than price that can shift demand for a product. Answer guidance: Consumer income, prices of substitutes and complements, consumer preferences, advertising and brand strength, population size, credit availability, government policy, expectations of future prices.

2. What does a price elasticity of demand of -2.0 mean in practical terms? Answer guidance: A 1% increase in price causes a 2% decrease in quantity demanded. Demand is elastic. Raising price will reduce total revenue because the quantity loss is proportionately larger than the price gain.

Understanding

3. Explain why a furniture company should use different forecasting methods for its existing product lines versus a brand-new product it has never sold before. Answer guidance: Existing products have historical sales data, so time-series analysis and regression against economic drivers are appropriate. A new product has no history, so the company must rely on consumer surveys, expert judgment, test marketing, or analogies from similar products.

4. How does knowing the income elasticity of demand help a manager plan for an economic recession? Answer guidance: If income elasticity is positive and high, demand falls sharply during a recession as incomes drop — the manager should reduce production and inventory. If income elasticity is negative (inferior goods), demand may actually rise during a recession as consumers trade down.

Application

5. A telecom company raises mobile data prices by 5% and finds that subscribers reduce usage by 8%. Calculate the price elasticity of demand. Is demand elastic or inelastic? What does this imply for revenue? Answer guidance: Elasticity = -8% / 5% = -1.6. Since the absolute value exceeds 1, demand is elastic. The price increase will reduce total revenue — the company should consider whether cost savings from lower data usage offset the revenue loss.

6. An Indian e-commerce retailer notices a clear seasonal peak in November and December. How should this affect their inventory purchasing decisions in October? Answer guidance: The retailer should increase inventory purchases ahead of the seasonal peak, using last year's November-December sales data adjusted for overall demand trend and any promotional campaigns planned. They must also buffer for forecast error to avoid stockouts during peak demand.

Analysis

7. A fast-moving consumer goods company finds that its regression demand model performs well in normal years but badly during years with major economic shocks. What does this reveal about the model, and how should the manager respond? Answer guidance: The model captures average behavior but not structural breaks caused by shocks. The manager should add scenario analysis for shock conditions, use leading indicators as early warning signals, and adjust the model after each shock by incorporating dummy variables or recalibrating coefficients.

8. Two competing airlines price economy seats differently: one uses rigid cost-plus pricing, the other uses dynamic pricing based on real-time demand. Which approach reflects better demand analysis, and what are the risks of each? Answer guidance: Dynamic pricing better reflects demand — it raises prices when demand is high and lowers them when it is weak, maximizing revenue. The risk is customer backlash if price changes feel arbitrary. Cost-plus pricing is transparent but leaves revenue on the table during peak periods and overprices during weak demand.

FAQ

What is the difference between a shift in demand and a movement along the demand curve? A movement along the demand curve occurs when the price of the product itself changes — quantity demanded goes up or down while the curve stays in place. A shift in demand moves the entire curve, meaning customers want more or less of the product at every price. Shifts are caused by changes in income, substitute prices, preferences, population, or advertising — anything other than the product's own price. Managers must distinguish between these because they require different responses: a price change triggers a quantity response, while a shift in demand may require changing production capacity, pricing, or marketing.

Why do economists say demand is a relationship, not a number? Because demand describes how quantity responds across a range of prices, not just at one price point. Saying "demand is 1,000 units" is incomplete — demand at what price? The demand curve or demand schedule captures the whole relationship: at price Rs. 100, demand is 2,000; at Rs. 150, demand is 1,200; at Rs. 200, demand is 600. Managers need the full relationship to evaluate the revenue implications of any pricing decision, not just the quantity at today's price.

How often should demand forecasts be updated? As frequently as new information becomes available and the cost of updating is justified. For fast-moving retail products, weekly updates may be appropriate. For capital investment decisions, quarterly or annual updates may suffice. The US retail industry has moved toward near-real-time demand sensing using point-of-sale data. Indian FMCG companies increasingly use weekly distributor sell-through data. The principle is that a forecast should be updated whenever new information would meaningfully change the estimate.

Can demand forecasting predict black-swan events? No — by definition, black-swan events are not predictable from historical data. However, good risk management supplements demand forecasts with scenario analysis that asks: what would our business look like if demand fell by 30% or 50%? This is not the same as predicting the event but it builds resilience to low-probability, high-impact shocks. The COVID-19 pandemic revealed how exposed firms were that had no contingency scenarios for extreme demand disruption.

What is the Delphi method and when is it useful? The Delphi method collects opinions from a panel of experts through several rounds of structured questionnaires. After each round, experts see a summary of the group's answers and can revise their estimates. The process converges toward a consensus view. It is most useful when historical data is absent — for new technologies, emerging markets, or decisions far in the future. It is less useful when experts share the same blind spots or when the question has a right answer that data could reveal. A startup entering an entirely new market in India or the US might use Delphi alongside test marketing.

Quick Revision

• Demand is quantity buyers are willing and able to purchase at different prices in a period.
• The law of demand says quantity demanded falls when price rises, other things equal.
• Demand shifts when income, substitute prices, preferences, or other non-price factors change.
• Price elasticity of demand = % change in quantity / % change in price.
• Elastic demand (elasticity > 1): price increase reduces total revenue.
• Inelastic demand (elasticity < 1): price increase raises total revenue.
• Forecasting methods include expert opinion, surveys, moving averages, regression, and test marketing.
• Time series has four components: trend, seasonality, cyclical movement, and irregular variation.
• Forecast error = actual minus forecast; track mean absolute percentage error to improve over time.
• Overforecasting creates excess inventory; underforecasting creates stockouts and lost sales.
• Income elasticity distinguishes normal goods (positive) from inferior goods (negative).
• Demand forecasting feeds production planning, inventory, pricing, hiring, and capital decisions.

Related Topics

Prerequisites

• Introduction to Managerial Economics
• Basic microeconomics (supply and demand)
• Elementary statistics and data interpretation

Related Topics

• Pricing Decisions
• Cost and Production Analysis
• Marketing Management and Consumer Behavior
• Operations and Inventory Management

Next Topics

• Cost and Production Analysis
• Market Structures
• Pricing Decisions