Different levels of understanding of market information necessitate varying approaches to sales forecasting, as an analyst depends on concrete numbers to procure reliable results. Access to historical sales and promotional response data allows identifying the applicable methods and utilizing them to the best of their extent. This process permits appraising and adopting an effective forecasting mechanism for various marketing mix allocation scenarios considering the outlined case.
Differences Between Forecasting Methods
There are various approaches to marketing forecasting, each possibly yielding different results. The average method, reliant on mean historical data, and the naive one, which considers only the last observation, should not produce similar numbers (Hyndman & Athanasopoulos, 2018). The seasonal naive method allows utilizing the previous time of the year’s numbers, and the drift method relies on “the average change seen in the historical data” (Hyndman & Athanasopoulos, 2018, p. 49). Thus, four of the most commonly used forecasting processes depend on earlier achieved results.
Evaluation Concerning the Outlined Scenario
Evaluating these methods becomes possible when relying on the outlined scenario, affected by the availability of historical sales and, more decisively, promotional response data. According to the research by Trapero, Kourentzes, and Fildes (2015), this factor necessitates manipulating the data achieved by either method and making expert modifications. Assuming that the previous marketing period also utilized promotions, the seasonal naive method and the drift method may allow achieving the most accurate pre-adjustment results. Conversely, the average and naive approach may output numbers that are too broad or too narrow correspondingly.
Choosing a Forecasting Method
Deciding between the seasonal naive and drift method requires further analyzing the differences in their applicability. The latter method could be more relevant, as Hyndman and Athanasopoulos (2018) state it allows a more flexible time-based approach. Thus, since the only implemented assumption is the company’s previous use of promotional materials, research indicates that the drift method may be more useful to estimate the level of sales.
Conclusion
Since the drift method allows considering both the past and present observations to formulate future results, this rationale may make it most useful in different marketing mix allocation scenarios. For example, the results may be used when creating a case within which the company’s promotional activity is increased, forecasting the effect this may have based on appropriately identified previous time brackets. From trade promotions to pricing strategies, the drift method allows accounting for all previously achieved results to decide these tactics’ planned effectivity.
References
Hyndman, R. J., & Athanasopoulos, G. (2018). Forecasting: Principles and practice (2nd ed.). Web.
Trapero, J. R., Kourentzes, N., & Fildes, R. (2015). On the identification of sales forecasting models in the presence of promotions. Journal of the Operational Research Society, 66(2), 299-307. Web.