Academic journal article The Journal of Business Forecasting

Forecasting New Products in Consumer Goods

Academic journal article The Journal of Business Forecasting

Forecasting New Products in Consumer Goods

Article excerpt

EXECUTIVE SUMMARY Forecasting new products continues to be one of the biggest planning challenges in consumer packaged goods. There are no magical algorithms, forecasting tools, or proprietary process solutions that offer much more than a "like as" or analog-based planning solution. The companies that do the best job in forecasting new products work the details in a methodical way, challenge underlying assumptions, and examine all available data to give themselves the best possible chance for accurate new product forecasts.

One of the toughest demand planning tasks is forecasting new products - particularly in the consumer goods world. Why? First, we don't really have good math to/Help us. It would be great if there were a forecasting algorithm that reads consumers' minds, but there isn't. Consumer goods manufacturers are at the mercy and whim of the consumer. Their wants, needs, and tastes can change quickly depending on trends, pop culture, the economic environment, and even the price of gas. Second, the statistics we have on new product introductions suggest that most new products will fail within a short period of time. A 2004 Best Practices Study by the Product Development & Management Association (PDMA) found a 49% failure rate for fast-moving consumer goods.

Of course, failure is a subjective term and can rangefrom being delisted (removed from shelves by a retailer) to just failing to meet expectations. Nevertheless, the perceived failure rate of new products is high. Third, it is difficult for facts to trump aspirations, and new products carry lots of aspirations - high hopes for success - which are normally reflected in a bias toward over-forecasting and, consequently, forecast error. Marketers' hopes and dreams for their products show up in their forecasts, and while the focus on failure tends to center around new products not achieving expectations, even highly successful new product launches tend to have a higher-than-normal forecast error.

In the demand planning community - within the social networking cloud and at conferences - there is an awful lot of chatter about new product forecasting getting more and more difficult. In response to an inquiry for this article, Lora Cecere of offered an interesting statistic: New product forecasts average an 80% MAPE (Mean Absolute Percentage Error). That huge level of forecast error epitomizes the difficulty in achieving a reasonable level of accuracy for new product forecasts.

Of course, there are many theories as to why it is so hard to forecast new products. Some of these theories focus on internal reasons for such high error rates, while others concentrate on external reasons. They range from ideas about consumers being worn out by claims of "new and improved" to manufacturers putting out mediocre line extensions. Sometimes, it is the market research used to drive a new product forecast that is faulty or overly aspirational. Other times the marketing execution, messaging, positioning, product quality, or design is off the mark. Certainly the tough economy of late has reduced the likelihood of consumers spending their limited resources on new, untried products; furthermore, retailers are decidedly more hesitant to take on additional SKUs, along with the resulting inventory that comes with them. Retailers want proven winners, and some will wait until consumer "consumption" or point of sale (POS) activity demonstrates success at other retailers before they decide to carry a new item. Even the traditional definition of retailers is changing. With more and more consumers buying online, the much anticipated e-retailing environment is coming of age. Whatever the dynamics or combination of reasons, there is certainly more awareness about the challenges of forecasting new products in the consumer goods space.

As both a consultant and practitioner, I have spent a considerable amount of time trying to refine a methodology for forecasting new products in the consumer packaged goods marketplace. …

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.