Overcoming the Perception of Precision
                        Parts 1 & 2

Part #1

Making Executive S&OP work is not an extension of past experience. It is not doing the same things better, but rather doing things differently to be better. This means change, and that sometimes is hard to do.

One of the more difficult issues is to overcome the ‘perception of precision’ that many companies have – the notion that if more certainty in a forecast is desired, more granular detail is necessary. Actually, the opposite is true. If you’re too close to the forest, all you see is trees and become distracted by the detail. This is not to say that no detail is necessary, but full granular detail is a non-value-adding distraction.

One company I know, getting started with Executive S&OP, derived their aggregate family forecasts by rolling up the detail. Well, it soon became clear that that highly detailed forecast was just wrong, even to a casual observer. The forecasters needed to learn that “less is more.”

Executive S&OP needs to be roughly right and not precisely wrong. While there is much data to be managed in an Executive S&OP implementation, it must be understood that human judgment and simplification will beat a formula every time. The role of the computer is to enhance human judgment, not replace it.

In a future column, we’ll talk about how that’s done.

Part #2

In our last column, we promised to talk about how to overcome the perceptions of precision associated with forecasting at a very detailed level. We recommended forecasting at a family or sub-family level (except for close-in, where SKU detail is often needed to buy and make the right items). There are two things that contribute to accomplishing this goal of forecasting at a higher level:

First is developing the mechanics and simplifying assumptions to convert a family forecast into resource requirements and revenue projections.

The second element is dealing with the organizational behavioral issues surrounding what can be a dramatic change from past practice.

This second element is far more difficult. For this reason, we recommend an approach that is rooted in taking a low risk path to gaining experience and confidence; one that proves this approach is a simpler and better way to forecast the future. We sometimes call this the pilot/parallel approach. It has the following characteristics:

•    It totally parallels what’s being done today, thereby eliminating all risk
•    It is a small slice of the business so the development work is not overwhelming. As a result:
•    It takes 90 days or less to complete
•    It enables top management to participate, signaling their willingness to change

Learning that ‘less is more’ with regard to forecasting cannot happen without effectively dealing with the counter-experiential and counter-intuitive nature of the change. This approach will demonstrate and lead an organization to learn, in a low risk environment, that it’s far better to be roughly right than precisely wrong.