Eniac Sales Optimization via Strategic Discounts

Note

This work is part of a data science project using an artificial dataset provided by the project owner.
The goal was to practice extracting insights from data and presenting them in a clear, structured way.

Purpose

As part of Eniac’s ongoing effort to boost revenue through data-driven strategies, this project explores the impact of discounts and product categorization on sales performance. With a total revenue of €7.8M and strong seasonal spikes in demand, the goal is to develop a discounting and product segmentation framework that enhances sales performance without eroding margins.

Problem Statement:

Eniac has observed fluctuations in sales that align more with seasonal events than with its discounting strategy. This raised the need to evaluate:

• How effective discounts are across different product types and price segments.
• Whether categorization by product type or price basket yields better strategic insight.
• How to structure data pipelines to monitor and optimize performance.

Solution:

This analysis utilized sales data to:

• Categorize products by price baskets and product types.
• Quantify the impact of seasonality and discount levels on sales.
• Identify data pipeline improvements for better tracking of discounts, returns, and inventory.

The insights inform targeted marketing, pricing, and inventory strategies for sales optimization.

Summary

This project evaluates Eniac’s discount strategy and categorization approaches using structured sales data. It provides insights into revenue drivers, highlights seasonality trends, and offers recommendations for refining the data pipeline to support smarter pricing and promotions.

Data Analysis & Findings

1. Sales Performance by Price Basket

• All the products which were sold were grouped into different price baskets for effective categorization of products.
• Top 3 price baskets (Mid-range, Upper Mid-range, Premium) generated ~€6.2M, about 79% of total revenue.
• Basket 4 (Upper Mid-range) alone generated ~€2.6M, accounting for 31.6% of total revenue.
• Combined average discount for top baskets was around 17.0%.

2. Sales Performance by Product Type

• Categorizing sold products based on product type revealed patterns like:
  • Higher responsiveness to discounts in some categories (e.g., accessories).
  • Others (e.g., laptops) showed volume stability independent of discounting.

3. Impact of Seasonality

• Black Week and Christmas dominate sales surges.
• Overall discounts showed limited effect on revenue increases — external events were primary drivers.
• However, within specific product categories, targeted discounting drove high volumes.

4. Correlation Analysis

To understand interdependencies between discount levels, product categories, seasonality, and revenue, a Pearson correlation matrix was used.

• Key Observations:

  • Revenue positively correlates with seasonal events, indicating event-driven sales surges.
  • Discount percentage shows a moderate correlation with sales volume but not with total revenue, suggesting discounts may be attracting buyers but not necessarily maximizing earnings.
  • Product type categories show distinct correlation patterns with revenue and discounts, reinforcing the value of segmentation.

  

This analysis supports the need for targeted discount strategies rather than blanket promotions, emphasizing the role of category-specific behavior in strategic planning.

5. Data Pipeline & Operational Gaps

• Current Issues Identified:
  • Lack of defined product categories.
  • No tracking of promo prices or customer returns.
  • Poor metadata and inventory visibility.
• Proposed Enhancements:
  • Track discounts & promotions explicitly.
  • Centralize product metadata and category definitions.
  • Implement return/reason logging for customer feedback.

Key Learnings

• Strategic Categorization is essential for actionable insights — both price and product type should be used in tandem.
• Seasonality outweighs discounting as a sales driver; however, targeted discounting still boosts volume in key areas.
• Clean, structured data pipelines are critical to support real-time marketing and inventory decisions.

Challenges Overcame

• Lack of initial product categorization and promo tracking resolved via custom data processing.
• Revenue attribution by category required data joins and basket creation logic.
• Needed to infer discount effectiveness from fragmented metadata.
• Enhanced the framework for measuring sales impact per basket & category.

Additional Reflections

• Greater automation of data enrichment (e.g., auto-categorization) would improve scalability.
• Strong product metadata hygiene is required to sustain ongoing performance analysis.
• Long-term benefit lies in centralized analytics dashboards tracking promo efficacy, seasonal peaks, and product-level ROI.

Repository Contents

• Presentation Slides: Summary of findings, visualizations, and strategic takeaways.
• Data Analysis Reports: Data processing logic and metrics calculation.
• README File: Executive summary and key learnings.

Deployment & Contribution

How to Use This Repository:

1. Clone the repository:

   git clone https://github.com/hsjoi0214/eniac-discount-analysis.git

2. Navigate to the project folder:

   cd eniac-discount-analysis

3. View the presentation and analysis reports.

Contributions:

• Suggestions on improving categorization logic or discount effectiveness models are welcome.
• Pull requests and issues are encouraged to further enrich the project.

Languages and Libraries Used

• Programming Languages: Python
• Tools: VScode, Google Slides

Credits

• Data Sources:
  • Eniac sales data.
• Analysis Conducted by: [Prakash Joshi]

Acknowledgements

Thanks to my mentors for supporting the initiative and contributing to the refinement of discounting strategy and analytics pipeline.