The challenges of pre-launch forecasting of adoption time series for new durable products

Abstract

The successful introduction of new durable products is important in helping companies to stay ahead of their competitors. Decisions relating to these products can be improved by the availability of reliable pre-launch forecasts of their adoption time series. However, producing such forecasts is a difficult, complex and challenging task mainly because of non-availability of past time series data relating to the product and the multiple factors that can affect adoptions such as customer heterogeneity, macro-economic conditions following the product launch and technological developments which may lead to the product's premature obsolescence. This paper critically reviews the literature to examine what it can tell us about the relative effectiveness of three fundamental approaches to filling the data void : i) management judgment, ii) analysis of judgments by potential customers and iii) formal models of the diffusion process. It then shows that the task of producing pre-launch timeseries forecasts of adoption levels involves a set of sub-tasks, which involve either quantitative estimation or choice, and argues that the different nature of these tasks means that forecasts are unlikely to be accurate if a single method is employed. Nevertheless, formal models, rather than unstructured judgment should be at the core of the forecasting process.Gaps in the literature are identified and the paper concludes by suggesting a research agenda to indicate where future research efforts might be most profitably employed.

Retail forecasting: Research and practice

“…The literature on this topic is vast and even by the mid-1980s, Assmus (1984) found the number of methods too numerous to include in his review paper. Since then, many new product forecasting methods (and models) have been developed, and a number of recent reviews on this topic have been made from different perspectives though none focus any attention on the retail new product decision (Chandrasekaran and Tellis, 2007;Goodwin, Meeran, and Dyussekeneva, 2014;Machuca, Sainz, and Costa, 2014;Meade and Islam, 2006). In general, new product sales forecasting methods can be grouped into three broad categories: (i) the judgmental approach, which entails management judgment based in part on past experience; (ii) the market research approach, where survey data is used to forecast customers' purchasing potential; and (iii) the analogical approach, whereby the forecaster assumes the product will behave as "comparable products" have behaved, a comparison which entails the identification of such comparators and which itself is heavily judgmental.…”

Supply chain forecasting: Theory, practice, their gap and the future

“…Indeed, forecasting for new products is often performed by selecting appropriate "predecessor" or "similar" products to the one to be forecasted (NOTE 9) . By fitting appropriate models to these longer time series of sales of "similar" products, one can forecast a new fashion product's entire lifecycle, or a new durable good's adoption curve (Goodwin et al, 2014). A related approach is not to use similar products, but similar locations, by offering a new product initially at a small sample of test stores and extrapolating to other stores (e.g., Fisher and Rajaram, 2000), which is, however, only possible if the goods already physically exist, so requires separate production runs or can only be used for short-or medium-term forecasts.…”

The influence of product involvement and emotion on short-term product demand forecasting

“…For example, when determining the market potential of a new fashion product, managers may base their estimate on an analogy; that is, they may take a similar product, launched in the past, as a reference and then adjust its sales on the basis of contextual information that is relevant to the new product (Abernathy, Dunlop, Hammond, & Weil, 1999). In doing this, they can be overinfluenced by the cases that are most easily retrieved from their memory (Goodwin, Meeran, & Dyussekeneva, 2014;Lovallo, Clarke, & Camerer, 2012;Tversky & Kahneman, 1974). Even expertise is no guarantee of accurate forecasts.…”