For this case study, we have generated data through software as we cannot use real data of our clients
The dataset contains sales data for 12 months. It has 13 columns. The details of these columns are given below:
(1) Product: Product ID
(2) APR: Sales Amount for the month of April
(3) MAY: Sales Amount for the month of May
(4) JUN: Sales Amount for the month of June
(5) JUL: Sales Amount for the month of July
(6) AUG: Sales Amount for the month of August
(7) SEP: Sales Amount for the month of September
(8) OCT: Sales Amount for the month of October
(9) NOV: Sales Amount for the month of November
(10) DEC: Sales Amount for the month of December
(11) JAN: Sales Amount for the month of January
(12) FEB: Sales Amount for the month of February
(13) MAR: Sales Amount for the month of March
The sales data from April to March (for one year) is given as amount with Product ID. The data are given from April to March because in India financial year starts from April. In any case, we require data of last 12 months so that we can find products whose sales is high
Below, we give screenshot of Sales Data:
The basic idea to find products who sales is very low is to find whether these products can be discarded and put forth best marketing efforts on the best-selling products to improve total sales and the overall profitability of the company.
Result 1: Yearly Data Analysis Low Sales Products
We have selected the dataset in "Discover" and asked to find out top 1000 products whose sale is very low. We have not selected Product ID as it is only for identification. Below, we give the screenshot of "Discover":
When we click on "Result" button, we see the following result (given screenshot of Result):
Analysis of Result
We find that top Product ID is "AFV546269". Its overall sale is lower than other products. The score is 94.97%. The Key Column 1 is JAN_low. The sale of this product for the month of January is 3,733 that is lower than other months. However, the score is computed based on all the months.
Note: Please note that it took less than 5 seconds on a moderate hardware to find top 1,000 lowest selling products from 10,000 products.
Result 2: Third Quarter Low Sales Products
For this purpose, we have selected all columns except OCT, NOV and DEC so that all other columns are ignored, and computation is done of these three months. This gives the low sales product result for Third Quarter (for the months of October, November and December). Below, we give the screenshot of "Discover":
When we click on "Result" button, we see the following result (given screenshot of Result):
Analysis of Result
In the given screenshot, we can see the top 28 products with very low sales though "Discover" has computed top 1,000 products. We find that Product ID of top "low sale product" is "CUF425855". Its score is 93.30%. The Key Column 1 is DEC_low. The sale of December month is 3,589 that is lower than other months of the Third quarter. Of course, the score is computed based on all the months of the Third quarter (Months of October, November and December).
Result 3: Fourth Quarter Low Sales Products
For this purpose, we have selected all columns except OCT, NOV and DEC so that all other columns are ignored, and computation is done of these three months. This gives the low sales product result for Fourth Quarter (for the months of January, February and March). Below, we give the screenshot of "Discover":
When we click on "Result" button, we see the following result (given screenshot of Result):
Analysis of Result
We find that Product ID of top "low sale product" is "XQN042398". Its score is 94.24%. The Key Column 1 is FEB_low. The sale of the February month of this product is 3,674 that is lower than other months of the Fourth quarter. Of course, the score is computed based on all the months of the Fourth quarter (Months of January, February and March).
Result 4: First Half-Year Low Sales Products
For this purpose, we have selected all columns except APR, MAY, JUN, JUL, AUG and SEP so that all other columns are ignored, and computation is done of these six months. This gives the low sales product result for First Half-Year (for the months of April, May, June, July, August and September). Below, we give the screenshot of "Discover":
When we click on "Result" button, we see the following result (given screenshot of Result):
Analysis of Result
The Product ID of top "low sale product" is "GNG546986". Its score is 93.95%. The Key Column 1 is AUG_low. The sale of August month is 3,750 that is lower than other months of the First Half-Year. However, please note that the score is computed based on all the months of the First Half-Year (Months of April, May, June, July, August and September).
Result 5: Second Half-Year Low Sales Products
For this purpose, we have selected all columns except OCT, NOV, DEC, JAN, FEB and MAR so that all other columns are ignored, and computation is done of these six months. This gives the low sales product result for Second Half-Year (for the months of October, November, December, January, February and March). Below, we give the screenshot of "Discover":
When we click on "Result" button, we see the following result (given screenshot of Result):
Analysis of Result
The Product ID of top "low sale product" is "XQN042398". Its score is 94.90%. The Key Column 1 is FEB_low. The sale of the February month is 3,674 that is lower than other months of the Second Half-Year. However, please note that the score is computed based on all the months of the Second Half-Year (Months of October, November, December, January, February and March).
Note 1: The time taken for each type of result is less than 5 seconds on a moderate hardware.
Note 2: The basic idea to find products who sales is very low is to find whether these products can be discarded and put forth best marketing efforts on the best-selling products to improve total sales and the overall profitability of the company.
