Data

A forecast is only as good as the data or assumptions it is based on.

Good data and assumptions are the foundation of every good forecast.
Data (such as historical performance, behaviour, etc) must be carefully collected and checked for potential errors, which need to be eliminated.
Statistical techniques for examining data series, and identifying underlying trends within data series can be employed. These form part of the area known as time series modeling.

Assumptions will include elements for which data are not available, or are uneconomical to collect. Also included are variables which are to be altered as part of a ‘what-if’ analysis.

Assumptions must be gathered and checked just as carefully as data, and must be fully documented in the forecast. Several techniques have been developed to gather assumptions in a rigorous manner.
These techniques include methods such as :

• ‘Delphi’ groups
• task-focused consensus groups
• exploratory workshops

For more information, please contact us.