Demand Forecasting
Pivotal to operations demand management and PSI planning An unbelievable amount of information exists Multiple methods always deepen understanding … and lower risk. Precision is usually more apparent than real Goal: get close and have contingency plans
Forecasting Approaches Statistical analysis Regression, Time Series, etc. Market research Conceptual models Expert judgment Complementary … not mutually exclusive
QuantitativeQualitative Numbers Judgment Used when situation is vague & little data exist –New products –New technology Intuition, experience e.g., Internet sales Qualitative Methods Qualitative Methods Used in stable situations when historical data exist –Existing products –Current technology Math / stats techniques e.g., color televisions Quantitative Methods Quantitative Methods
QuantitativeQualitative Extrapolate Model Roll-up Disaggregate Bottom-up Top-down Numbers Judgment Demand Forecasting
QuantitativeQualitative Extrapolate Model Roll-up Disaggregate Bottom-up Top-down Numbers Judgment Demand Forecasting
Top – Down Disaggregation Industry Category Product Item
Top – Down Disaggregation Industry Company Product Item
“Tyranny of 100” Share gains must come at the expense of specific competitors (who are very likely to retaliate) Which competitor(s)? Why? How?
QuantitativeQualitative Extrapolate Model Roll-up Disaggregate Bottom-up Top-down Numbers Judgment Demand Forecasting
Bottom-up Aggregation Customer 1 Item Customer 2 Customer 3 Item
QuantitativeQualitative Extrapolate Model Roll-up Disaggregate Bottom-up Top-down Numbers Judgment Demand Forecasting
Years Penetration % Time Series Analysis Actual Projected
Years Penetration % Analogous Product New Product Time Series Analysis Analogous Products
QuantitativeQualitative Extrapolate Model Roll-up Disaggregate Bottom-up Top-down Numbers Judgment Demand Forecasting
ILLUSTRATIVE L TRANSLATION PROSPECTS PERCENT WEIGHT PROFILE BUYERS Definitely 90%10%9% Probably40%20%8% Might or might not10%20%2% Probably not015%0 Definitely not035% 0 19% Intent Translation Model Source: Thomas, p.206
YX ii ab Shows linear relationship between dependent & explanatory variables –Example: Diapers & # Babies (not time) Dependent (response) variable Independent (explanatory) variable SlopeY-intercept ^ Linear Regression Model
Regression Issues Illusory correlation –No cause and effect Meaningless coefficients –Unexplainable variations
Sequential Factoring Total TV Households Baseball Fanatics Wired For Cable Homes Cable/ Baseball Population Premium Service Buyers Baseball Pay Per View Market * A.K.A. “Factor Decomposition”, “Factor Analysis”
For example … How much dog food sold annually in the U.S.? Express answer in $$$$
Sequential Factoring How much dog food? How many people? How many homes? Homes with dogs? Number of dogs per home? Proportion of big & little dogs ? Daily consumption ? (ounces) Ounces per can ? Price per can ?
# Big # Little Little Eats # Dogs Homes % Dogs Homes w/ dogs Dogs / Home Big/little split Big Eats Popul- ation People / House Dog Food How Much Dog Food ?
Demand Forecasting Market Factoring MARKET POTENTIAL SALES MARKET SHARE MARKET PENETRATION MARKET SIZE
Market Forecasting Time Dimension
Keys to Success Practical precision Structured approach Multiple methods Iterative convergence
Demand Forecasting General Principles Errors are a certainty Aggregate series most stable Tendency to over-correct (especially short-run)
Demand Forecasting MARKET POTENTIAL SALES MARKET SHARE MARKET PENETRATION MARKET SIZE Market Disaggregation Time Series Analogies Regression Analysis Diffusion Model Intent Translation A-T-R Model Bottom-up Composites Value Function Conjoint Analysis Tyranny of 100 Majority Fallacy Cannibalization Effect
Demand Forecasting