Demand Forecasting that Reflects the Minds of Customers
Challenges in Demand Forecasting
One of the most significant challenges in the demand forecasting landscape is the low responsiveness to real-time data analysis tools. Senior leadership in this landscape endeavors to finalize a solution that has all the answers to their data analytics queries. Folks, let’s be honest. We are in a world that values the excessive acquisition of data and our inability or unwillingness to discard this vast amount of data makes using it challenging. Just like collecting stamps or sporting cards, eventually you’ve collected too many of them and you have trouble knowing exactly what you have collected. This is where I believe that metadata comes into play by adding value to this amorphous body of data that we collect in order to make it meaningful for future use. I view this ‘metadata’ as a packaging around data that helps to define it. This is similar to how packaging indicates the contents within a carton or how people used metadata by way of cards in a library catalog. In this instance, the metadata consisted of all the contents of a library, organized with a single card for each item in that library. Until we understand how to leverage this metadata at our fingertips, and apply these learnings across our unwieldy datasets, a distinct fix for this intricate challenge to unlock usefulness of real-time data cannot be achieved.
After spending a considerable amount of time in the supply chain ecosystem, particularly in the retail and manufacturing industry, I have observed that more and more businesses operating in the demand forecasting landscape want to leverage data analytics for financial planning, sales forecasting, promotional activities, and even building future roadmaps. There is also a different aspect of data analytics, if seen from the vantage point of data insight. I have observed that there are chunks of useful information hidden under the humongous mountain of collected data. As demand forecasting professionals, we must quickly unearth the ‘diamonds from the rough’ within the multiple datasets we have at our disposal. It falls to us to derive a sustainable methodology that can shed light on the most valuable insights to be used within the different facets our organizations. This must be our prime objective and can add significant value to our companies.
Challenges of Managing Huge Datasets
The biggest obstacle to managing huge datasets, especially while dealing with a vast number of stores across the country, is efficient inventory management. Today’s customers have numerous options; hence a store cannot afford to have customers disappointed when looking for a product. Technology can be the driving force in determining the purchasing trend of a given day or week or month, and it can enable stores to display high demand products that at the forefront, thus enhancing the customer experience. All in all, not just understanding the needs of customers but also anticipating them through predictive analytics is critical to successful customer engagement, which in turn is essential for demand planning and forecasting.
Metadata comes into play by adding value to this amorphous body of data that we collect in order to make it meaningful for future use
Factors that Drives Sales
For our organization, there are many factors that determine our sales curve. I have observed that causal factors such as ‘lottery and weather’ play a significant role in driving customers to our storefront. Lottery is a volatile factor, with its contours marked by uncertainties. However when the potential financial windfall crosses a particular threshold, engagement from customer’s side increases rapidly. Weather also plays a vital role in customer engagement. Pleasant weather conditions would drive more customers into stores when compared to inclement weather. Therefore, an important objective of 7-Eleven is to identify ways to counter the adverse impact that comes with these factors. In doing so, predictive analytics, data analysis, and other forms of technology need to be employed. Digital and social media, for example, has great potential in acquiring customer’s trust and loyalty; therefore this information is quintessential to supply chain, demand planning, and forecasting.
Advice to Upcoming Demand Forecasting Leaders
The typical adage of ‘Supply Chain as a Service’ has now been marginalized. The idea that supply chain should own the data, data analytics, and data mining is a valid proposition; however, the action these data insights dictate has changed drastically in the last 15-20 years. As leaders in demand forecasting, we should make proactive choices to understand insights gleaned from data and make that information available to other parts of the organization. In my experience, after working in multiple industries, they all employ similar techniques in data analytics. Typically, the companies in these industries pour in massive amounts of data, slice and dice it, prepare a pseudo plan before producing the actual plan, and review it at the end. This is a template type approach that yields savings at the cost of efficiency. Having the right product in the right place at the right time is the model for convenience and leads to an enjoyable customer experience. That being said, demand forecasting leaders must analyze the data and organize it in a format that makes sense to other areas of the organization. This in turn helps those business units drive the idea of true customer demand. Data should reflect the minds of customers, and it should be leveraged to maximize customer engagement.