When Discounts Hurt: Surprising Insights from Online Gifting Solutions’ Journey to Optimize Inventory

At Max Vision Solutions, we pride ourselves on delivering cutting-edge Business Intelligence (BI) and analytics services. Recently, we uncovered an unexpected insight while helping Online Gifting Solutions, a major player in the Indian gifting industry, optimize their inventory and sales strategy. In a competitive market where discounts are often seen as an essential driver for boosting sales, we discovered something surprising: discounts don’t always work.

The Challenge

Online Gifting Solutions is an online platform offering personalized gifts and flowers across India. As the company prepared to expand, they aimed to enhance their inventory management and profitability by analyzing their sales patterns. However, rather than confirming that discounts boosted sales, our analysis revealed an unexpected trend—higher discounts were leading to a drop in sales.

The Process

To understand this phenomenon, we began by collecting a comprehensive dataset of sales transactions spanning the full year—from January to December 2023. Our team meticulously cleaned and processed this data to ensure its accuracy. We analyzed various aspects of the sales data, including order size distribution, regional sales trends, category-wise sales performance, and seasonal spikes in demand.

Data Collection and Preparation

1. Data Collection: We gathered sales data directly from Online Gifting Solutions' online platform. This dataset included detailed information on each order, such as product details, pricing, discounts, and customer locations.
2. Data Cleaning and Processing: Date Transformation: Standardized order dates into a consistent datetime format using Python’s Pandas library. Removing Redundant Columns: Simplified the dataset by eliminating unnecessary fields. Handling Missing Values: Addressed gaps in data using Pandas’ fillna() function to replace missing values. Standardizing Data: Ensured consistency in text fields and corrected discrepancies in discount and price fields. Creating Category Columns: Added a column to classify products into distinct categories based on attributes. Price Consolidation: Ensured uniform pricing data by consolidating and standardizing price entries. Grouping Data by Order ID: Aggregated data for better analysis. Validation: Cross-checked the cleaned data against original records to verify accuracy.

Analytical Methods

We applied several analytical methods to gain insights into the sales patterns:

1. Diagnostic Analytics: Identified patterns and performance trends by analyzing total sales, average order values, and top-performing products.
2. Predictive Analytics: Projected future sales trends based on historical data, identifying peak periods and seasonal variations.
3. Prescriptive Analytics: Recommended strategies for inventory management and pricing based on diagnostic and predictive insights.

Results and Findings

Our analysis uncovered several key findings:

1- Descriptive Analysis

In the descriptive analysis, we examine the data from Online Gifting Solutions. to summarize key metrics and trends. This approach provides a clear view of the company’s performance, highlighting important aspects such as total sales, order volume, and average order value. By analyzing these details, we gain valuable insights into operational efficiency, identify areas of strength and potential improvement, and guide future strategic decisions.

Sales Overview: Total Sales: ?26,166,030.38

Number of Orders: 14,150

Average Order Value (AOV): ?1,849.19

2- Top-Selling SKUs

Popular items included products like "Escorts Gifts" and "Color It Happy," which drove significant revenue.

Low-Selling SKUs: Items such as "Cashew 250 gm" and "Crushed Pineapple Cake (1kg)" generated minimal revenue, indicating potential areas for inventory optimization.

3- Monthly Sales Trends:

Monthly Sales Trends The monthly sales data reveals patterns of performance throughout the year. Peak periods are evident, such as significant sales increases during festive seasons like Valentine's Day and Raksha Bandhan.

Seasonal Peaks: The months from July to September exhibited the highest sales, likely due to a combination of multiple festivals and events during this period. Understanding these trends helps in planning inventory and marketing efforts to align with high-demand periods.

Off-Peak Periods: Certain months showed lower sales, indicating periods of reduced customer activity. These slower periods can be used to plan promotions and adjust inventory to avoid overstocking.

4- Discount Impact Analysis

Impact of Discount on Sales, we observed an unexpected trend: higher discounts resulted in lower sales volumes. Conversely, sales actually increased when no discount was provided. This indicates that discounts may not always boost sales and could potentially harm profitability.

Unexpected Trend: Higher discounts were associated with lower sales volumes. Surprisingly, non-discounted items performed better. This counterintuitive finding suggests that discounts may not always drive increased sales and could harm profitability.

5- Time Series Analysis

Time series analysis looks at sales data over time to spot patterns and trends

Monthly Sales Trends with Rolling Mean: Fluctuations: Significant sales fluctuations were observed, with peaks in July 2023 followed by a slight decline. The rolling mean highlighted an overall upward trend with notable peaks during holidays. Future Projections: Sales are projected to grow, especially during peak times, necessitating inventory and marketing adjustments.

6- Event Performance Analysis

Event performance analysis helps us understand how different types of parties or events affect sales. This analysis shows which events are most profitable and how their sales vary.

Impact of Events on Sales: Festival Influence: Events such as Valentine’s Day, Holi, Raksha Bandhan, and Diwali significantly impacted sales. For instance, Diwali saw a substantial spike in sales due to high demand for gifts and festive items.

7- Geographical Analysis

Geographical analysis examines how sales vary by location to understand which regions perform best and where there might be opportunities for growth. This analysis helps in identifying high-performing areas and tailoring strategies to improve sales in different regions.

Sales Distribution by Region: Top-Performing Regions: Maharashtra, Delhi, and Uttar Pradesh emerged as high-performing states, showing strong market presence and potential for growth. Underperforming Regions: States like Andaman and Nicobar, Arunachal Pradesh, and Puducherry showed lower revenue, indicating a need for revised strategies in these areas.

8- Category-Wise Sales Analysis

Category-wise sales analysis evaluates sales performance across different product categories. This analysis helps understand which categories are most popular and profitable, and how sales trends are likely to evolve in the future.

Product Categories: Top Categories: Flowers and cakes were top sellers, with seasonal peaks observed in categories like flowers during Valentine’s Day and Diwali. Low-Performing Categories: Items such as dry fruits and teddy bears had lower sales, suggesting a need for targeted promotions or product strategy adjustments.

To improve inventory management, adjust stock levels based on sales trends for each category, ensuring that high-demand products are consistently available. Implement stock rotation strategies to manage products with declining sales and avoid overstocking.

For marketing strategy, focus on promoting top-performing categories to maximize revenue and leverage their popularity. For lower-performing categories, develop targeted promotions or introduce new products to increase sales and renew customer interest.

In strategic planning, prepare for growth in high-demand categories by scaling up production or sourcing. Address potential sales declines by diversifying product offerings or enhancing product quality to sustain and boost performance

SWOT Analysis

• Strengths: Strong brand reputation, skilled team, reliable suppliers, and high-performing products.
• Weaknesses: Challenges with excess inventory, discount strategies, and seasonal stock management.
• Opportunities: Potential for new markets, advanced analytics use, and CRM system enhancements.
• Threats: Competition, economic downturns, supply chain issues, and ineffective discount implementation.

Technology Used

To conduct this analysis, we employed a suite of Business Intelligence tools and BI software:

• Data Visualization: Tools like Tableau and Power BI enabled us to create interactive BI dashboards, allowing us to visualize trends and patterns clearly.
• BI Reporting Tools: We utilized SQL and Excel for comprehensive BI reporting and data exploration.
• Data Warehousing: Our analysis involved a robust data warehousing solution to consolidate data from various sources.
• ETL (Extract, Transform, Load): We used ETL processes to ensure data integrity and prepare it for analysis.
• Predictive Analytics: Python and R were used for predictive analytics, helping us forecast sales trends and customer behavior.
• Self-Service BI: Tools like QlikView were implemented to empower users with self-service BI, enabling them to explore data independently.
• Data Mining: We employed data mining techniques to uncover hidden patterns and insights.
• Cloud BI: Leveraging cloud BI solutions ensured scalability and accessibility of data across teams.
• Artificial Intelligence in BI: AI was used to enhance data analysis and generate actionable insights.

Why Did This Happen?

Several factors might explain why deep discounts backfired:

1. Perceived Value: Customers might have questioned the quality of discounted items, assuming that high discounts were tied to lower-quality products.
2. Brand Positioning: Online Gifting Solutions markets itself as a premium gifting service. Deep discounts could have clashed with this brand perception, leading customers to look elsewhere.
3. Customer Behavior: Many customers used the platform to purchase gifts for important events. These purchases were driven more by emotional value than price. As a result, offering deep discounts didn’t align with customer priorities.

The Strategic Solution

To address the issues identified, we recommended a more targeted approach:

1. Optimize Inventory for Peak Seasons: Ensure sufficient stock of top-selling items during high-demand periods.
2. Focus on High-Performing Regions: Tailor marketing efforts and inventory distribution to regions with higher sales.
3. Rethink Discount Strategies: Implement selective, well-timed discounts rather than blanket reductions.
4. Leverage Data for Smarter Decisions: Use advanced analytics to predict customer demand and optimize inventory.

Conclusion: The Power of Data-Driven Decision Making

This case study highlights a critical lesson for businesses: more isn’t always better, especially when it comes to discounts. A well-crafted strategy, based on careful data analysis and customer behavior insights, can unlock hidden opportunities to enhance profitability and optimize operations.

Our Free Business Insight sessions, combined with our Business Intelligence (BI) and analytics services, are designed to empower businesses with the data they need to make informed, impactful decisions. Whether you're looking to optimize inventory, enhance profitability, or leverage advanced BI solutions, get in touch with us today.

Important Links

• Raw Data: https://docs.google.com/spreadsheets/d/1b_9bcgkMJ2X4co1thuf4-CXcT1PGaM7QoZ42m096dHE/edit?gid=1904289295#gid=1904289295?
• BDM Project Link: https://colab.research.google.com/drive/1WrPwDgFBxZwqur8XLflC0UI3FybnfS2M?usp=sharing

Cleaned Data: https://drive.google.com/file/d/1-KBjAGlQbYuqhQZWplW2B7WwMpwNtf6g/view?usp=sharing