Demand Forecasting: Meaning, Objectives, Importance & Methods

1. Meaning and Definition of Demand Forecasting

Every business produces goods today to sell them in the future. Since the future is uncertain, managers cannot rely on blind guesswork (“Chalo 10,000 pieces बना लेते हैं”).

• Meaning: Demand Forecasting is a scientific and systematic process of estimating the future demand for a product or service for a specified future period, based on past sales data, current market trends, and consumer surveys.
• Definition:

  “Demand forecasting is an estimate of sales during a specified future period based on a proposed marketing plan and a set of particular uncontrollable and competitive forces.” — Cundiff and Still

2. Objectives of Demand Forecasting (Short-Term vs. Long-Term)

Why do companies forecast demand? The objectives are divided into Short-Term Objectives (up to 1 year) and Long-Term Objectives (3 to 10 years):

3. Steps in Demand Forecasting

4. Methods (Techniques) of Demand Forecasting

All methods of Demand Forecasting are divided into two grand categories:

1. Qualitative / Survey / Opinion Methods (Based on human judgment & surveys — best for new products or short-term forecasts)
2. Quantitative / Statistical Methods (Based on past numerical data & math — best for established products)

Part A: Survey / Opinion (Qualitative) Methods

1. Consumers’ Interview / Survey Method

Directly asking potential buyers what and how much they plan to buy in the coming period! It has three sub-types:

• a) Complete Enumeration Method: Surveying 100% of all buyers in the market. (Practical only when buyers are very few — e.g., an industrial steel machine maker asking 15 car factories).
• b) Sample Survey Method: Selecting a small representative sample of consumers (say, 1,000 households across 10 cities), interviewing them, and multiplying the sample result to estimate total market demand. (Fast and economical!).
• c) End-Use Method: Used for intermediate / industrial goods (like cement, steel, yarn, milk). We calculate how much of the product will be needed by each end-using industry (e.g., how much milk will be needed by households + sweet shops + ice-cream factories + cafes).

2. Collective Opinion Method (Sales Force Composite Method)

• Also called the “Grass-Roots Approach”.
• The company asks its own field salespeople, territory managers, and dealers to estimate expected sales in their local area for the next quarter, because they talk to customers every day!
• All regional estimates are added up and reviewed by the Sales Director.

3. Experts’ Opinion / Delphi Method

• Developed by the RAND Corporation (USA).
• Instead of a face-to-face meeting (where a loud senior boss might dominate everyone), a coordinator sends anonymous questionnaires to a panel of 5–10 independent industry experts across the country.
• After Round 1, the summary of reasons is shared anonymously with all experts, allowing them to refine their forecasts in Round 2 and Round 3 until a reliable expert consensus is reached!

4. Controlled Market Experiments / Test Marketing

• Essential for brand-new products that have zero past sales history!
• The company launches the new product in 2 or 3 test cities (e.g., Indore and Pune) with real prices and ads, observes actual consumer buying for 3 months, and uses those results to forecast national demand.

Part B: Statistical / Quantitative Methods

Used when a company has 5 to 10 years of past sales data:

1. Trend Projection Method (Time Series Analysis)

Assumes that past sales trends will continue into the future. It uses three techniques:

• a) Graphical / Freehand Curve Method: Plotting past years’ sales on a graph paper and extending the line forward by hand.
• b) Moving Average Method: Taking the average of the last 3 or 5 years’ sales to smooth out ups and downs:

  Forecast for Next Year = (Sales of Year 1 + Sales of Year 2 + Sales of Year 3) ÷ 3

• c) Method of Least Squares (Linear Trend Equation): Fitting a mathematical straight line to past data:

  Y = a + bX
  (Where Y = Forecasted Sales, X = Time Period, a = Base Sales, and b = Annual Growth Rate).

2. Regression and Econometric Method

• Links demand (Y) to its actual economic cause/determinant (X) — such as Consumer Income, Population, or Advertising Spend — rather than just time!
• Example: Forecasting demand for school bags based on school enrollment numbers, or forecasting paint demand based on new house construction.

3. Barometric (Economic Indicator / Lead-Lag) Method

• Just as a barometer predicts tomorrow’s weather by measuring air pressure today, economists use Leading Economic Indicators (events that happen first) to predict demand for products that follow later!
• Examples:
  • The number of Building Permits approved today is a Leading Indicator for the demand for Cement, Bathroom Tiles, and Furniture 6 months later!
  • The number of New Car Registrations predicts the future demand for Spare Parts and Car Insurance.

💡 DAVV Exam Tip

Question (10 Marks): “What is Demand Forecasting? Explain its objectives and the various methods of demand forecasting.”

• Divide Objectives into Short-Term vs. Long-Term in a 2-column table.
• Divide Methods into A. Survey/Opinion Methods (Consumers’ Survey, Collective Opinion, Delphi Method, Test Marketing) and B. Statistical Methods (Trend Projection, Regression, Barometric Method) — examiners award full marks for this structured classification!