Effects Influence of Social Media Constructs on Shopping: An Empirical Study on the Prediction of Retail Clothing Sales

In a market characterized by frequent new product placement and their rapid exit, accurate forecasting of the sales is an important goal to guarantee the profitability and long-term survival of companies. This paper follows this line of research and explores the sales data of an Italian clothing company. The objective of this research is to improve sales forecasts. The data under study were extracted from the corporate datalake of the clothing company analyzed as well as from the company’s official Facebook page using Facebook’s Graph API. The sales forecasting should try to consider all the possible demand influencing factors and also explanatory variables. In this way, the time series perspective is combined with qualitative indicators on the demand side. The approach taken confirms the influence of sales from exogenous variables: there is a delay of 15 days between social communication and the sale of clothing. Social communication, as well as some regular Italian holiday, negatively affects sales. On the other hand, school holidays and some regular Italian holidays have a positive impact on sales. As a result, the advance knowledge of revenue forecasts allows managers to choose suitable marketing strategies.