Future Ready by Steve Morlidge & Player Steve

Author:Steve Morlidge & Player, Steve
Language: eng
Format: epub
Publisher: John Wiley & Sons, Ltd.
Published: 2010-05-10T16:00:00+00:00

There is extensive literature on modeling techniques, so in this chapter we have focused on giving guidance on how to choose the right model and how to overcome practical problems managers might face in implementing their choice. Before we leave the topic there are two more problems worth discussing.

Problem: How do we decide what to treat as ‘interventions’ for the purposes of forecasting?

Every day in every business managers make decisions. Many will involve making changes to existing plans and each of these could be treated as an intervention for the purposes of forecasting. An obvious merit of this approach is that each intervention is treated (quite correctly) as an investment rather than a cost; the accountant’s distinction between capital and revenue has led to an unhealthy focus of attention on those expenditures that happen to end up on the balance sheet to the detriment of those that do not. For example, a company like Unilever will typically invest 10 times as much on advertising and promotion as on capital, but only the latter will appear on the balance sheet. AMEX calculates that as much as 25-35% of their operating expenses are discretionary in nature.

On the other hand, by treating every decision as an intervention it is easy to overwhelm the forecast process with detail — in large enterprises there may be thousands of projects active at any one time. Management is unlikely to have the detailed knowledge to make considered judgments about such a large number of projects, few of which will have a significant impact on the performance trajectory of the business. As a result we recommend you apply the 80:20 rule. The incremental impact of the projects that do not ‘move the dial’ should be treated as part of ‘momentum’: business as usual.18

Wherever the line between momentum and intervention is drawn it is important to ensure that the assumptions that are made are consistent. If, for instance, interventions made in the past have had a ‘one off’ impact, then that impact will have to be stripped out of the historical record used for the statistical modeling of momentum (see Figure 4.6). If this is not done you would be double counting the impact of initiatives.19

Figure 4.6 Combining models.This example demonstrates the use of two different types of models to produce a combined ‘momentum plus interventions’forecast. Here the impact which interventions made in the past has been excluded before extrapolating a trend statistically toproduce a momentum forecast. This is supplemented by a judgmental forecast of the incremental impact of interventions toproduce a combined forecast.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.