Overview
This lecture covers the principles of forecasting in marketing, including key steps, different forecasting methods, and strategies to improve forecast accuracy.
Forecasting Process Steps
- Determine market potential—estimate total industry-wide sales for a specific product and time period.
- Estimate company sales potential—a firm's maximum expected sales, usually a percentage of market potential.
- Adjust forecasts as factors like price, competition, and market conditions change.
- Forecast revenues and compare them with product costs and market potential.
Types of Forecasting Methods
- Judgment Techniques: Relies on opinions of customers, salespeople, executives, or external experts.
- Surveys: Gather customer or channel member intentions for buying, best for market potential estimates.
- Sales Force Composite: Aggregates sales estimates from the sales team; more accurate for short-term.
- Executive Opinion: Averages sales estimates given by company executives; fast but can be biased.
- Expert Opinion: Uses outside experts, providing insights but often inaccurate alone.
- Time Series Techniques: Analyzes past sales data to forecast trends, with adjustments for changing conditions.
- Correlational Analysis: Uses related variables (leading indicators) like housing starts to predict sales.
- Response Models: Statistical models based on past customer responses to marketing strategies.
- Market Tests: Launches a new product in a limited market to gather real-world sales data.
Improving Forecast Accuracy
- Choose forecasting methods suited to the product and business cycle.
- Combine multiple forecasting methods for a more reliable estimate.
- Forecast sales for smaller business units or segments before consolidating.
- Employ different scenarios to account for possible market changes.
- Regularly track actual sales data and adjust forecasts accordingly.
Key Terms & Definitions
- Market Potential — Total expected industry sales for a product in a given time frame.
- Sales Potential — The maximum expected sales for a company, as a fraction of market potential.
- Judgment Techniques — Forecasts based on opinions rather than quantitative data.
- Time Series Analysis — Examines historical sales data to predict future trends.
- Correlational Analysis — Links sales forecasts to trends in related variables.
- Response Model — Uses customer and sales data to predict how marketing actions influence sales.
- Market Test — Experimental product launch in a small market to estimate broader demand.
- Leading Indicator — A variable that changes before sales do, helping predict future trends.
Action Items / Next Steps
- Review and compare forecasting methods for different product categories and timeframes.
- Practice combining forecasts and adjusting them as new data arrives.
- Prepare answers for questions on method selection, expert opinion, and accuracy improvement.