How do you know your business is going in the right direction?<\/p>\n
With so much data we have in hand, it is very easy for the problem to get lost while cleaning and analysing it. To be honest, I was in the same place. As a Data Analyst\/Data Scientist, I would try to look at the data first and make sense of it. But it not only consumed more time but also made it difficult to explain why I took certain decisions.<\/p>\n
A recent project on Business Intelligence taught me the power of problem-solving. For this tutorial, we will be exploring a dataset of a global store. To keep it general, I am going to assume our problem statment here is The framework breaks this down into two major strategic approaches:<\/p>\n There are a few ways to boost profits. This framework highlights three core strategies:<\/p>\n \u27a4 Increase the Quantity Sold More units sold usually means more revenue. To do this effectively: – Ensure demand is met<\/strong> \u2013 Make sure your supply chain and stock levels can support customer demand. We need more data to do this. – Focus on region-specific performance<\/strong> \u2013 Some regions may outperform others, and identifying where sales are strongest helps target efforts better.<\/p>\n \u27a4 Sell More Products with High Profit Margins Not all products are created equal. Shifting focus toward high-margin items can improve profitability. – Use product-level<\/strong> data to identify top performers. – Understand regional trends<\/strong> to see where high-margin products sell best.<\/p>\n \u27a4 Reduce the Shipping Cost Shipping and logistics can quietly erode margins, so optimizing this is crucial. – Analyze which products<\/strong> incur the highest shipping costs. – Check which regions<\/strong> are driving those costs and explore smarter routing or fulfillment solutions.<\/p>\n While discounts can drive volume, excessive discounting eats into profits. Two key tactics can help manage this:<\/p>\n \u27a4 Identify Best-Selling Products Best-sellers may not need heavy discounting to move. – Reassess discount policies at the product<\/strong> level. – Consider performance variations across regions<\/strong>.<\/p>\n \u27a4 Uncover Bottlenecks Inefficiencies or slow-moving inventory often trigger unnecessary discounts. – Identifying and addressing bottlenecks<\/strong> can reduce dependency on markdowns.<\/p>\n Now that we have a basic framework in place, let us look at each aspect and decide on the metrics.<\/p>\n The following libraries are required for the below code to execute:<\/p>\n GridDB’s GitHub Page<\/a> covers the installation in detail. Please go through it to be able to interact with the GridDB Server in Python.<\/p>\n Great! Let’s go ahead and import the libraries<\/p>\n We will be using an open-source SuperStore dataset that can be downloaded from here<\/a>. Once we download it and put it in the same folder as the code, we can then use pandas to read it and clean it.<\/p>\nHow to grow our business<\/code>. Then, we will work through the problem statement in a tree-like framework. This will help in structuring our analysis and also get specific KPIs.<\/p>\nFramework<\/h2>\n
1. Increase Our Profit<\/h3>\n
2. Decrease Our Discount<\/h3>\n
What metrics are we tracking?<\/h2>\n
\n
\n
\n
\n
\n
Pre-requisites<\/h2>\n
\n
Importing Libraries<\/h2>\n
#this is optional - in case of errors during execution even though the python client is installed, it is likely that this could be a path issue and can be resolved as follows\nimport sys\nsys.path.append(\"\/home\/shripriya\/python_client\")\n#sys.path.append(<path_to_python_client>)<\/path_to_python_client><\/code><\/pre>\n<\/div>\nimport griddb_python as griddb\nimport pandas as pd\nimport seaborn as sns\nimport matplotlib.pyplot as plt<\/code><\/pre>\n<\/div>\nDataset<\/h2>\n
Reading the Dataframe<\/h2>\n
superstore_df = pd.read_excel(\"global_superstore_2016.xlsx\", sheet_name=\"Orders\")\nsuperstore_df.head()<\/code><\/pre>\n<\/div>\n