The weekly weight index is a management tool that uses the historical sales data of a certain sales cycle as the basis and calculates the sales weight analysis on a weekly basis. Huang Chengming has a detailed description of this in his book “Data-based Management: Insights into Retail and E-commerce Operations”. Here, we will combine the data of Amazon’s clothing category to explain the case of the weekly weight index.
Amazon’s sales still belong to the category of retail, so they will fluctuate periodically on a weekly basis. If managers directly compare sales data, although fluctuations can be seen, it is difficult to summarize the rules. In this case, managers need to reprocess the data, divide daily sales by the total sales of the week to obtain the percentage coefficient, and then make a weekly sales percentage fluctuation diagram. In this way, it can be clearly seen that the store has the highest sales on Wednesdays, the lowest sales on Sundays, and a slight increase in sales on Mondays and Saturdays. The sales trend is obviously downward at other times.
After data verification, we found that sales do meet the characteristics of weekly cycles, so we can use the weekly weight index as an indicator for analysis. The calculation method of the weekly weight index is as follows.
(1) Collect the sales data of each store for the most recent full year.
(2) Eliminate outlier data such as Prime Day, “Black Friday”, and Deals through variance.
(3) Sort the remaining data by week, with the row label as week and the column label as day, and calculate the average daily sales.
(4) Take the sales data with the lowest average daily sales and set its daily sales weight index to 1.00. Divide the other 6 sales data by this data to obtain the daily sales weight index of the other 6 days.
(5) Add up the daily weight index to get the final weekly weight index. The minimum weekly weight index should be 7.00. The larger the value, the more unstable the sales.
After statistics, we can get the weekly weight index of a store.
Using the same method, we can get the weekly weight index of a single store. It should be noted that because the daily weight index of a single store has a small range of values, it estimates the sales of one month based on 3 months of data, so it needs to be updated once a month to ensure the reference of the data.
Finally, managers need to specify appropriate daily weight indexes for holidays and promotion days. For the US site, most holidays are defined by week. Since there are peak shopping periods before holidays and low periods during holidays, most promotions will also be launched before holidays. Therefore, you can simply compare the sales data of the past two or three years and use “daily weight index = daily sales/unit weight (sales) value” to calculate the daily weight index of promotion days and one week before holidays.
With a quantifiable weekly weight index, sellers can jump out of the traditional empirical operation thinking, track sales through data, set performance goals and complete real-time tracking. Taking the data in Figure 11-8 as an example, assuming that the monthly target of store A is 30,000 yuan, the sales target that store A should achieve every day can be calculated. The calculation formula is as follows.
Daily sales target = monthly sales target x (daily weight index/monthly weight index)
Among them, the monthly weight index is equal to the sum of the daily weight indexes of the whole month.
If the manager finds that the store has not met the target for one week, he should check the store products and the working status of the operator in time to ensure that the target is completed on time. Assume that today is July 18, the sales of store A is 17,510 yuan, and the target completion rate is 58.4%. At this time, the manager can follow up in time according to the weekly weight index to observe the sales completion.
We can see that if the current sales status is maintained, it is still a little short of the target completion this month. At this time, the operator can increase the number of orders appropriately through advertising and other means to ensure the target completion.
For some special sales nodes, such as Prime Day, “Black Friday” and “Cyber Monday”, the operator needs to set performance targets separately. At this time, the operator needs to make a comprehensive comparison of sales over the years. In addition to guiding the formulation of appropriate goals, it can also guide reasonable stocking to avoid out-of-stock or backlogs.
In addition, we calculate the unit weight value of each day over a period of time and draw a unit weight value curve, which Huang Chengming calls the “Huang’s Curve” in the book “Data Management Insights into Retail and E-commerce Operations”.
The absolute weight curve should be a straight line, but under realistic conditions, this unit weight curve will fluctuate around a certain value. By tracking and analyzing the distribution of these values, we can analyze and evaluate the impact of store promotions, special events, new product launches, out-of-stocks, and other factors.
Because the unit weight curve needs to be updated and tracked in real time, it is not recommended for small sellers and single store operators. Sellers who already have their own team and store group and have been operating for more than 1 year can try using this curve.