This research investigates three approaches to new product sales forecasting: statistical, judgmental and the integration of these two approaches. The aim of the research is to find a simple, easy-to-use, low cost and accurate tool which can be used by managers to forecast the sales of new products. A review of the literature suggested that the Bass diffusion model was an appropriate statistical method for new product sales forecasting. For the judgmental approach, after considering different methods and constraints, such as bias, complexity, lack of accuracy, high cost and time involvement, the Delphi method was identified from the literature as a method, which has the potential to mitigate bias and produces accurate predictions at a low cost in a relatively short time. However, the literature also revealed that neither of the methods: statistical or judgmental, can be guaranteed to give the best forecasts independently, and a combination of them is the often best approach to obtaining the most accurate predictions. The study aims to compare these three approaches by applying them to actual sales data. To forecast the sales of new products, the Bass diffusion model was fitted to the sales history of similar (analogous) products that had been launched in the past and the resulting model was used to produce forecasts for the new products at the time of their launch. These forecasts were compared with forecasts produced through the Delphi method and also through a combination of statistical and judgmental methods. All results were also compared to the benchmark levels of accuracy, based on previous research and forecasts based on various combinations of the analogous products’ historic sales data. Although no statistically significant difference was found in the accuracy of forecasts, produced by the three approaches, the results were more accurate than those obtained using parameters suggested by previous researchers. The limitations of the research are discussed at the end of the thesis, together with suggestions for future research.

Fashion and the contemporary environment as a whole, is a complex environment that requires retailers and wholesalers to adapt to the changes that constantly occurring. This adaptation is in a bid to ensure that more profits than loses are realized by the company. For this reason, companies have to use various methods to determine the best ways to improve their products. Companies resolve to introduction of new product to the market but the acceptance of new products to the fashion industry is not an assured factor but rather a gamble. This is mainly because of the industry’s characteristics. The main aim of this thesis is to analyze the methods that may be used to improve the accuracy of new products. The fashion industry has characteristics that may be considered as challenges because for instance, when a product is launched, one has to determine whether it is by a reputable designer or whether it is a trend, and with the fashion industry, trends are mainly turned into such by celebrities who introduce a certain design to the world for adoption. These challenges or characteristics are carefully analyzed and examined with the necessity of the introduction of new products analyzed. Data collection, being the main backbone of this thesis and multiple-case study method, is used to answer the research question as “How can structured analogy method be used to improve the forecast accuracy for the footwear products in the fashion industry “.Samples for case study have been chosen from footwear category. Structured analogy method is used to determine the accuracy of the information gathered from literature review.

The purpose of this research is to investigate current company business process from sales forecasting perspective and provide potential improvements of how to deal with unstable market demand and increase overall precision of forecasting. The problem which company face is an unstable market demand and not enough precision in sales forecasting process. Therefore the research questions are:  How current forecasting process can be improved?  What methods, can be implemented in order to increase the precision of forecasting? Study can be described as an action research using an abductive approach supported by combination of quantitative and qualitative analysis practices. In order to achieve high degree of reliability the study was based on verified scientific literature and data collected from the case company while collaborating with company’s COO. Research exposed the current forecasting process of the case company. Different forecasting methods were chosen according to the existing circumstances and analyzed in order to figure out which could be implemented in order to increase forecasting precision and forecasting as a whole. Simple exponential smoothing showed the most promising accuracy results, which were measured by applying MAD, MSE and MAPE measurement techniques. Moreover, trend line analysis was applied as well, as a supplementary method. For the reason that the case company presents new products to the market limited amount of historical data was available. Therefore simple exponential smoothing technique did not show accurate results as desired. However, suggested methods can be applied for testing and learning purposes, supported by currently applied qualitative methods.

One of the major demand related risks for companies that produce consumer electronics goods is change in consumer preferences over time as reflected in the weights they attach to the attributes of products. This contributes to the difficulty of predicting whether consumers will purchase a new product or not and the accuracy of such forecasts can have significant ramifications for companies’ strategies, profitability and even their chances of survival. Knowledge of attribute-weights and accurate forecasts of new products can give companies better insights during the product development stages, inform go-no-go decisions on whether to launch a developed product and also support decisions on whether a recently launched product should be withdrawn or not due to poor early stage sales. Despite the important implications of change in attribute-weights, no research has investigated the extent to which such changes occur and impact on the accuracy of forecasts of the future market share of these products. Prior to the current research, it was assumed that the weights are constant over time – even when the nature of the attributes was assumed to change. To investigate these concerns choice based conjoint (CBC) was applied to data gathered in a longitudinal survey of consumer choices relating a range of consumer electronic products, where innovation has different rates and the product life cycles are various. This allowed an assessment of the extent to which the weights of attributes of choice-based conjoint models change over a six months period for consumer durable products and the degree to which this variability is dependent on the nature of the product. It demonstrates that the change in weights is greater for products that have high technological complexity and shorter lifecycles and also links the changeability of weights to the characteristics of potential consumers. The results of thesis demonstrate that the assumption of constant weights can potentially lead to inaccurate market share forecast for high-tech, short life-cycle products that are launched several months after the choice-based modelling has been conducted.