Scenario-based forecasting involves creating multiple versions of the forecast and then varying the assumption inputs to create alternative forecasts. It is common for three scenarios to be created, a base case which is the most likely outcome, worst case which includes a credible set of negative assumptions and, best case, including a credible set of positive assumptions. The best approach to scenario-based forecasting is to first identify those assumptions that have the largest impact on the forecast and are also associated with some uncertainty.

Probability-based forecasting involves defining a probability for each of the key assumptions. This is then used to analyse the different possible outcomes and a range is produced associated with a level of likelihood – typically that a certain level of revenue will be achieved. We recommend Monte Carlo simulation a computerised mathematical technique that allows you to account for risk. A Monte Carlo simulation provides the decision-maker with a range of possible outcomes and the probabilities they will occur for any set of variables.

Econometrics takes the historical performance of a dependent variable, like unit sales, and explains its relationship to a set of independent variables, like price and/or promotional spend. The resulting mathematical model describes these historical relationships, for example that a 10% increase in price resulted in a 5% decrease in unit sales. This econometric model can then be used to help predict the future performance of a product.