Linearity of Sale is a concept denoting the variability of the consumption, and thereby the demand, of any product. The concept is closely related to seasonality of demand. Understanding of linearity of sale has helped many retail marketing companies in India optimize their retail sales and value chain expenditure over the years. Evaluation of linearity becomes more important in the context of FMCG sector as the demand is highly disaggregated and collecting direct information on demand variation becomes extremely complicated.
But most attempts at understanding the linearity of sale end with larger time frames of around a year or so. But, what about the linearity of sale within a month? How does the consumption or purchase of products vary within a month? Answers to these questions might not help in readjusting the value chain but there are other ways to use this information. Ways which are gaining more and more importance by the day.
In the context of retail, by identifying products whose demand tapers off in specific periods within a month, their merchandising strategy can be tweaked to account for that. Linearity of Sale of a particular product within a month can also help in understanding the shopper behavior as it can be seen as a parameter which can explain the impulse generation and other such purchase influencers.
We had analyzed the sales of few FMCG outlets in Delhi to understand this linearity. Below chart shows the sales trends of these outlets.
The X-axis is the day of the month and the Y-axis shows contribution of that day to monthly business. This data validates a few hypotheses that most would have. The first week of a month would always have high business as pay-checks are just being cashed. But what explains the rise in the sale in the last week of the month? The lowest period of sale is actually not at the end of the month but it is around the end of third week.
Let us now look at similar charts for each outlet separately.
4 of the 5 outlets broadly follow similar pattern as the overall chart, with variation only at the extent of the rise in the last week and timing of the dip. Only the green colored outlet follows a radically different pattern. Curiously the green colored outlet is a Chemist outlet, which has completely different kind of shopper behavior and purchase motivations as expected theoretically.
Purely through analyzing the data, we could identify the exception outlet. We hope to extend the analysis on larger set of outlets and cluster the outlets by the shape of the curve to group them. We believe such clustering would be able to group outlets with similar shopper behavior much more effectively than any qualitative exercise can ever hope to do.
It is important to measure the linearity of sale across categories, outlets and geographies. It can be used to understand the shopper behavior and help in effectively classifying the outlets and geographies to developing marketing plans accordingly. We have developed a parameter to measure the skew of the demand called “Intra Month Skew” (IMS). This is defined as the ration between average of top 3 3-day moving contribution windows and bottom 3 3-day moving contribution windows. An IMS of 1 would indicate equal demand on all the days while an IMS of 2 would indicate that the peak purchase periods have double the sale of bottom purchase periods.
The IMS of the Overall sales is 1.72. This means that the peak periods have around 72% higher sales than the bottom periods in these stores. IMS for individual outlets is depicted in the following chart.
It would be interesting to explore the causes of differences in shapes of the curves and IMS values across outlets. These are the first steps towards a system which could tell you much more about the shopper behavior in an outlet than qualitative research could ever do.
The most interesting part of this study was to see how IMS varies across categories. The next article in this series would dive deeper into the data to present category level insights in linearity of sale.