Leveraging Social Media for Improving Forecast Reliability
One of the greatest challenges that complex supply chains of the companies today face are accurate forecasting and demand planning. In this context, the social media can be leveraged effectively to analyze customer data and achieve more insight on forecasting, planning, scheduling and inventory management. For example, imagine if it was possible to predict the percentage of customers that were interested in buying a black colored Reebok T-shirt against a brown colored one or to predict the percentage of customers interested in buying a sportz model of a car versus an another model. This would help to stock appropriate quantity of inventory and also facilitate in computing the right value of safety stock.
Social media helps in bringing real time data in terms of the percentage likes or dislikes posted by users of a social media site. Also the informal interaction enabled through social media helps consumers to express their candid views about a product. Customers who are interested to buy a product online or are sure that the product cannot be returned are particularly concerned about the quality and the correct utility of the product. These consumers prefer to take an informed decision based on viewing the product reviews in social media sites. Also since personal interactions don't happen much these days, an informal chat over a social media site or reading product reviews by first hand consumers can greatly affect the buying decision.
Social media not only could help building leaner supply chains but also help in bringing companies closer to their customers. Since social media is something that is used by majority of consumers on a daily basis, it is slowing evolving as the most powerful tool to know the current demand trends of products. However, the social media analytics should be brought into the supply chain strategy only after performing a due diligence on the customer demographics, location, culture etc. since the user profile of social media may bias the kind of conclusions drawn. For example, it would make more sense to use the customer data of users located in Northern US, Finland, Germany for a winter jacket. The reason being this group of users are more likely to buy such a product as against Asian or African users.
Also, there are some more points to look at before using the social media data. What sample size should be considered and how should the percentage likes and dislikes be extrapolated? Also the users using the social media sites can change their likes and dislikes. Hence, when should the data be integrated into the system?
Demand forecasting necessarily needs a mathematical model to suggest the right numbers for the demand of products. Social media can only provide a trend. However it needs to be seen as to how effective can a tool/software be that has the ability to translate such trends and patterns into numbers
However, the final question still remains as to whether the views in terms of likes/dislikes expressed over social media are reliable enough to integrate the social media analytics into a tool like ERP to drive forecasting ? Isn't an expression of interest not more than just mere excitement of a product?