GfK HealthCare November 2011
	Eight Simple Ways to Improve Your Forecasts By Rudiger Papsch, Managing Director, Asia, and Dulyalarp Nanthawatsiri, Research Director Forecasting is a complex and rather technical matter. To make things even worse, it can cause substantial financial damage to companies if they get their forecasts wrong. Unfortunately, there are numerous examples of pharmaceutical companies that over- or underestimated the market potential of their drugs, which resulted in misallocation of resources or lost opportunities for them. This poses not only a significant challenge to a company, but also to the career of the person preparing the forecasts, as it is fairly easy to evaluate the accuracy of a forecast post hoc. That is probably why one finds only two groups of people: Those who are experts in forecasting and those who prefer to stay away from it. Rarely one finds people in the middle, that is, who see the need and value of good forecasting, but who at the same time have a pragmatic approach to it and do not wish (or simply do not have the time) to become experts in it. This view is confirmed by a recent survey that GfK HealthCare conducted with 31 clients from the pharmaceutical industry. Sixty-one percent indicated that they are not satisfied with the way forecasting is conducted in their organization, and 68 percent stated that their organization does not have a systematic and functional forecasting process in place. Many different challenges were reported by the participants, and the most often mentioned challenges were these: A systematic approach to the challenges of forecasting It is actually not difficult to group these challenges systematically along two dimensions: nature of challenge (Technical versus Business) and source of challenge (Forecaster versus Stakeholder). Based on these two dimensions it is possible to derive the following matrix, which allows classifying each challenge into one of its quadrants: Forecasting Techniques Positioning (of the forecasting process in the business) Communication Quality & Monitoring Thinking about these challenges, there is no reason why forecasting has to be such a problem and sometimes cause so much discomfort to the forecaster. No doubt, forecasting can be resource-intensive and challenging and usually involves many technicalities which may not be easy to understand. From our point of view what causes this situation is not something inherent to forecasting per se, but rather how we teach and use forecasting. We often seem to stand in our own way and not utilize all those possibilities that can greatly help us make forecasting a much smoother process (and at the same time yield more accurate forecasts). Good advice and best practice have been given before, but they are usually limited to the technical aspect of forecasting (quadrant Forecasting Techniques). Even though this aspect is very important, it is not the only aspect that can pose challenges to the forecaster. Looking at the four quadrants we know there are three other aspects that need to be addressed. Forecasts are not just made for spreadsheets but for use by people in marketing, management and many other functions of the company to make better business decisions. Good forecasting needs to address these aspects too. Just by following a limited set of eight rules and best practices it is possible to address the challenges on all four quadrants and to improve the forecasting process and communication with stakeholders. Eight good practices for the forecasting process and for communication with stakeholders The following rules are suggested to improve the forecasting process and the communication and involvement of the various stakeholders across the organization: Quadrant 1: Forecasting Techniques Keep it simple. Various studies on forecasting methods have confirmed that simple methods forecast at least as accurate as complex methods (e.g., M3-Competition). Furthermore, simplicity keeps the forecasting process understandable, which facilitates the involvement of other stakeholders. Have a proper forecasting process in place, including documentation and presentation. This will not only increase the accuracy and efficiency of the forecasting process, but also ensure that the process is transparent. Quadrant 2: Positioning Keep the business objectives in mind – always. Forecasting is a business activity with a clear business objective. Therefore it has to be matched to the business decision and the target audience. Accept that forecasts may not always be popular. The objective of forecasting is not to be popular, but to give the most accurate estimation of the future. Of course this is easier said than done, but if you follow the other seven rules, you have already started to manage this problem effectively. Quadrant 3: Communication Make your forecasts transparent and involve the stakeholders openly and often. Because forecasts influence major business decisions many stakeholders are involved. Naturally they have different point of views and interests. Involve them early to get their buy in and expertise. The key is to focus on inputs (assumptions), not on the output (results). Manage expectations of your internal clients and management. Future prediction with precise numbers always involves a certain degree of error. Clearly communicate this fact to stakeholders. Even a forecast that is not perfectly accurate will still allow for making better business decisions than those based on gut feeling. Quadrant 4: Quality & Monitoring Follow your organization’s best practices and use Excel if you have a choice. If your organization has best forecasting practices in place, it is usually advisable to follow them and to work closely with the internal forecasting experts. The software package Excel has been proven a useful forecasting tool as its spreadsheet approach allows for preparing transparent and flexible forecasting models. Evaluate your forecasts and build a track record. Comparison of forecasts with actual outcomes is the most effective way to improve them. Various measures for forecasting errors are available; the Mean Absolute Percentage Error (MAPE) is usually the recommended choice (simple and meaningful). Keeping a track record will also help you to build your reputation. Want to learn more on this topic? Please contact:

Contact  |  GfK HealthCare Copyright © 2011 GfK HealthCare Published by GfK HealthCare | North America | Europe | Latin America | Asia Reproduction or any other form of duplication only with the written permission of the publisher. All rights reserved.