Academic journal article The Journal of Business Forecasting Methods & Systems

Choosing the Right Forecasting Software and System

Academic journal article The Journal of Business Forecasting Methods & Systems

Choosing the Right Forecasting Software and System

Article excerpt

The best forecasting system is the one which is multidimensional, can handle multiple data sources, is built for a broad community of business users, enables collaboration between various business partners, has an open architecture, includes a large variety of modeling techniques and offers rich analysis capabilities ... in buying a software or system, think long term, not only what you need today but also what you will be needing in the future.

With an increasing recognition of the forecasting function, more and more companies are searching for a forecasting software or system. In an effort to find the right software or system, the question often comes up what I should be looking at in making a selection, which is the topic of this article. MULTIDIMENSIONAL MODELING

Businesses often operate along at least three hierarchies: customer, geography and product. For example, a PC manufacturer can group its customers into two channels: direct and indirect. The direct channel includes selling directly to Corporate, Government and Educational clients. The indirect channel, on the other hand, includes selling to distributors and value-addedresellers. As such a hierarchy can easily run into four to six levels.

The business is becoming more and more dynamic. Every day new developments are occurring in different market segments. The business can respond quickly and effectively only if it can forecast these developments ahead of time. Consider two regions: Western and Eastern. The sales forecast of the Western region is usually lower than the Eastern region. If forecasts are tracked at the country level, the unique developments of each region will be overlooked. If you include suppliers, demographics and other dimensions in the model, you'll quickly run into six to eight business dimensions. Therefore, to determine and analyze the development in each dimension, the forecasting software/ system must have a multidimensional modeling and forecasting capability. Furthermore, these dimensions are interconnected. Each combination of customer, geography and product identifies a unique market segment which needs to be forecast separately. The forecast that is uniquely created for each market segment reveals the emerging trends, thereby allowing a company to respond right away.

This is not the be all and end all of the multidimensional forecasting. Within each dimension, a hierarchical model is required to align the forecast that is generated by different teams at different levels of the organization. For example strategic planning tends to generate long-range forecasts at the region, product line and channel levels. Tactical planning, on the other hand, calls for short-term forecasts by product groups and territory. The question one has to ask is: "Does the software/system offer a flexible framework to model, forecast and reconcile the forecast of all the relevant market segments?"


Have you ever lost a customer? If you have, you need a system that enables you to update the information easily so that it reflects the change. On the other hand, if you gain a new customer you should be able to add to your system. Furthermore, the system should allow you to add or remove an exhaustive list of products and territories that correspond to a customer that you are adding or dropping. If the system does not have that kind of flexibility but you have hundreds of products sold in dozens of territories, you may wind up spending a good portion of your time in adding products, removing customers, and realigning territories and sales reps.

If a forecasting system is flexible and easy to maintain, it will be a snap to update the information. Data driven and configurable forecasting system enhances team productivity, lowers the cost of maintenance, and improves the responsiveness to business changes.


Access to relevant historical data is the key to generate accurate forecasts. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.