Foreword
This research paper independently presents the main product analysed of the case company Apple inc; The I phone. What follows is a description of the sales forecast, sorting out the applicable theories of sales forecast that will give a better understanding of the literature. And finally, the qualitative and quantitative analysis sales method that suits our case company are analysed.
SALES FORECASTING
- Definition and concept of demand forecast
According to Mentzer & Moon, (2004), the sales forecast is dependent on market surveys and different statistical data that are analysed through qualitative and quantitative methods. Advancement is the supply and demand inclinations and other factors are given an in-depth analysis, calculations and speculation and a conclusion are made that defines the economic business strategy and tactics (Davis & Mentzer, 2007).
Fan, Chen & Chen, (2017) considers demand analysis as a way of fulfilling two management objectives: an understanding of the demand trends and forecasting of sales and revenue. Guo & Cheng (2018) denotes that the sales forecast can be deemed as a data exploration technology with the application of different statistics or regression models that use previous historical data to determine its sales prediction model.
Fildes & Goodwin, (2007) believe forecast to be the speculation concerning the probable future development, the result of which can be applied in planning. According to a research survey done by Choudhury (2018), if a company manages to achieve 1% more accuracy in sales prediction, its fulfilment rate in managing its demand increases by 2%.
- Factors affecting demand
According to Hofmann & Rutschmann (2018) sales forecast is the projected value of sales volume of an independent product in a specified time in the future. It is an essential tool for planning such as in budget making, production, procurement, and inventory as well as in making capacity schedules. A good prediction is paramount in making the decision and it assists managers to have a correct grasp of the forthcoming changes (Căpătînă & Drăghescu, 2015).
Most companies understand the essence of sales forecasting, however, to achieve a high quality, high precision forecast is challenging (Mentzer & Moon, 2004). Various factors that affect demand have to be considered and understood before making sales forecasts
Xing & Detert (2010) considers consumer income as one of the most essential factors. If a products price remains constant while the consumer’s income is increased, the sales volume of the product rises accordingly. The price of other similar products also affects demand. With substitute products, an increase in the price of a product the demand for the substitute product will rise. When the price of a product that is complementary to another decreases the quantity demanded one reacts by rising while the demand for the complement rises. According to Tran (2018), other factors include advertising and the brand of the product, national income levels as well as producing the population.
- Conditions for good prediction
Laugesen & Yuan (2010) believes that with sales forecasting the company’s external and internal factors ought to be considered. Among the external factors are demand trends, economic changes as well as industry competition. According to Davis & Mentzer (2007), demand trends are key factors influenced mainly by life patterns such as changes in population that result in a change in demand. As a result, is a company desires to achieve high-quality forecasts, it needs to gather market data. This can be achieved through statistical data surveys as well as understanding the consumer purchase motivation (Fildes & Goodwin, 2007).
Bashir & Verma, (2019) denoted that the economic changes directly influence the sale volume of any product. A decreased economic growth results in decreased income to consumers which decreases their consumption levels. Ittner & Michels, (2017), believes that to achieve an accurate product forecast, the government, as well as financial sector economic policies, have to be considered with detailed attention given to applicable indexes such as the GDP.
As described by Anand, Bansal & Aggrawal (2018) a closer look at the industry competition is essential as it offers an overview of all relevant operations of competitors. Information to consider includes product sale level, the target consumers and promotional plan of the competitors.
Among the intrinsic factors are influenced in the sales force, marketing strategy as well as product status. Bashir & Verma, (2019) describes marketing strategy as policies such as the product market positioning, advertising as well as promotions. If any of these policies changes, the sales levels are affected. According to Fildes & Goodwin, (2007), the sale of a product is a human-centric operation, therefore, the professionalism and motivation of the salesperson influence the consumer sales levels. Choosing the right sales personnel is critical in any business operation (Grafova et.al 2017)
Ittner & Michels, (2017 defines product status as personnel’s, raw materials as well as machines that are needed to ensure sales forecasts are met and delivered on time. Consequently, in forecasting sales, it is paramount to think through the product status of the company to prevent non-delivery of orders that can affect goodwill.
According to Hofmann & Rutschmann (2018) integrating the above mentions factors will impact the forecasting results positively. The thresholds that need to be met for high-quality sales forecast are timeliness and accuracy of the forces as well as reliability of predictions. Fildes & Goodwin, (2007) goes ahead to add other thresholds that include, the simplicity of the predictive technology used and that the predictions ought to show meaningful units.
Sales Forecast Method
Based on the Ittner, & Michels (2017) work, sales forecasting methods are categorized into two, the qualitative and quantitative forecasting subject to the analysis process. Choudhury, (2018) presents an all-inclusive explanation of forecasting techniques based on principles, tools, macroeconomic models as well as business cycle analysis. Davis & Mentzer (2007) gives a theoretical touch to these techniques while Xing & Detert (2010) discusses the contributors to forecast inaccuracies and ways of improving accuracy. Tran, (2018) goes ahead to give an introduction forecasting of business cycles and its relevant theories. It also offers an introduction to the history of business cycle indicators and lagging indicators.
Hofmann & Rutschmann (2018) argues that the most popular qualitative prediction methods are the Delphi method, Jury of Executive Opinion, leading Index technique, shopper expectation technique as well as Historical Analogy. With the quantitative predictors Choudhury (2018), analyses the methods as the time series model, averaging, exponential smoothing, as well as regression analysis.
- Qualitative approach
In this paper, when the qualitative prediction method is concerned, we have analysed our forecasts based on previous experiences and extensive professional knowledge of the sales of our product.
In their book Xing
& Detert (2010), applied the search engine of Google Trends in forecasting the values of economic indicators such as sales. The query indices most of the times are correlated with several economic indicators that are essential in short term economic prediction. In their book, they indicate that modest seasonal AR models that have relevant Google Trend data can outshine models without such date by 5 to 20 per cent.
The Conference Board’s Business Cycle Indicators Handbook (2001) gives a broad view of the Coincident, Leading as well as Lagging Economic Indicators. It encompasses the history, interpretation, evaluation and also constriction of these indicators. Worth noting is the comprehensive description of the Leading indicators and their selection process and the procedure involved in constructing the leading index. Selection of the cyclical indicators is done through a sequence of statistical and economic tests depending on economic implication, conformity, smoothness, statistical appropriateness and lastly consistent timing.
- Quantitative prediction method
Šindelář (2019) describes quantitative prediction method as the use of historical gathered data as well as factor variables in the establishment of mathematical models used to forecasting demand. On the other hand, Ghauri, Grønhaug & Strange (2020) defines is historical statistical data analysis, using the modern mathematical statistical technique in the prediction and organization of the data, and the construction of product prediction models. He further explains the objective of the method is that of speculating on its market trends and figuring out the correlation between the variables of the forecast product and its associated variables.
Also, Fildes, Goodwin & Önkal (2019) enlightens readers on the three objects that can be achieved by econometric models. The first is that the quantified formula retrieved for historical data leads to the prevention of accidents brought about by management or even policy adjustments. The second is that objective behavior is likely to be more accurate in reflecting the forecasted sales. Lastly, not only is it possible to foresee the direction of changes in the economy, but the intensity of such changes can also be foreseen.
As discussed by Ghauri, Grønhaug & Strange (2020) the measurement model can be applied in the modification and adjustment of the parameters once the comparison of the predicted and the actual data values are made. That being said, the premise of applying the quantitative prediction method in the construction of the forecasting model is that there must exist substantial usable data.
According to Demir & Akkas (2018), quantitative prediction methods are categorized into two. The first is the casual analysis method while the other is the time series analysis method. Šindelář, (2019 describes causal analysis method as the application of connections in the development of products to speculate their development tendencies based on previous years data. Such data is used to figure out variable as well as dependencies in the forecasted product to determine the mathematical model of prediction.
Majid & Mir (2018), introduces the most popular and simplest technique of the analysis namely the regression analysis. In multiple regression analysis, it involves the study primary of the relationship between two or more factors. From the analysis, the correlation between variables is acquired depending on the forecasted sales and the volume of sales of the product. Some variables such as population growth and GPD are correlated and their such variables can be applied in establishing the prediction of sales.