Using SPSS for statistical data analysis with an aim to increase the manufacturer’s profit
A renowned manufacturer and distributor of Construction products noticed a considerable decline in profits and approached us. Our team from Glorious Insights figured out the challenges the company was facing and helped them in generating profits. This case study describes in detail how the use of IBM SPSS for statistical data analysis helped this manufacturer in growing their profits considerably.
Customer background
The client is a prominent manufacturer and distributor of products for the construction business. The client offers a wide range of products for both commercial and residential construction. It is known for the reliability and durability of the products. Knowing that the construction industry rewards innovation, the company strives to maintain superior standards in performance, job quality, and efficiency.
Recently, the profits of the company have been stagnant. This is where it collaborated with Glorious Insights to find the real problem and take corrective measures. The company wished a detailed analysis of its sales data related to products, customers, and regions. This would help to identify the areas where the marketing should be focused on and areas where the operations can be improved for cost-saving.
Challenges
The client under consideration is a leading brand in construction-related products. The business of the company spans to a wide geographical range as it caters to its customers through multiple units.
Although the company has been constantly evolving in its process and products there has been no increase in the profits. This led the company to introspect and find the areas that needed improvements to overcome this problem.
Glorious Insights suggested a thorough data analysis. The challenge is to study the data related to customers, sales, location, SKU, and products. The ultimate goal was simple: to find ways to increase profit.
Solution
The company provided data for ten years. After meticulous scrutiny of the available options, we decided to use the SPSS software for accurate and dynamic statistical analysis. The solution provided organizes the data into statistical models and draws useful inferences that can suggest the necessary actions to increase the profit.
- Functional aspect
The solution system uses SPSS for both interactive and batch analysis of data. It classifies the data into different categories. The results help in analyzing trends for forecast and planning, improve efficiency, find new opportunities, and maximize revenue generation.
- SKU categorization
The categorization of SKU or store keeping units refers to the classification of each product kept in every unit. The products are classified based on the material, type, design, etc. For some products, the classification attributes were even more than ten.
- Customer categorization
Customers are classified according to their type that can be either business or individual. Business customers are further categorized as big box stores or retailers. For each customer personal, household, and demographic information is analyzed.
- Location categorization
Location here refers to the ship-to location, that is the place where the order is being shipped. This is related to the customers’ information. Location can be urban or rural, among the townhouses or in suburbs.
- Technical aspect
SPSS is a powerful package for the analysis of data. It allows us to study the data in different forms to draw out actionable information. It provides robust features to aggregate and evaluate the data to make the judgment. These features help you in making accurate predictions and take necessary actions to achieve your business goals.
- Cluster analysis
The feature helps in categorizing product groups based on their attributes. It helps in finding which product is popular in a particular region or which is closest to the profit margin.
- Segmentation analysis
This feature creates different segments of the customers and products based on their types and locations.
- Decision tree modeling
This tool is useful to divide the data into subsets that are simpler to manage and understand. The tree begins with binary subsets that are split until we reach an atomic entity.
Customers can be split into two sets, contractors and big box stores. The nest tree-level can be the products each type of customer buys. This gives information about the customer buying the most.
- Descriptive analysis
This step is to summarize the data and answer the questions like what is the average order size, which type of customer buys the maximum valued product, etc.
Results
With the information generated by the new system, the company could make some valuable decisions that contributed to a significant increase in profit.
- Some 20 percent SKUs were contributing only 1 percent to the sale. So ceasing their production helped in increasing profit.
- Surprisingly, independent contractors were found to be assets as they ordered most high-margin products.
- Interior designers contribute only 0.01 percent but ordered new products and influenced the trends.
- A significant difference is found in the products preferred by the customers in different regions.
