Megaline Statistical Data Analysis

Business Problem

Megaline, a telecom operator offering two prepaid plans (Surf and Ultimate), needed to determine which plan generates more revenue to allocate its advertising budget more effectively. Using a sample of 500 customers, this analysis examines call, message, and internet usage throughout 2018 and compares plan performance through rigorous statistical methods. The project demonstrates expertise in applying data science to solve real business challenges, translating complex datasets into clear, actionable recommendations that impact marketing strategy and resource allocation.

Note: All data in this project is simulated.

Key Findings

Revenue Comparison

Ultimate generates consistently higher and more stable revenue than Surf. While Surf revenue depends heavily on overage charges (especially for data usage), Ultimate provides predictable monthly revenue through fixed plan pricing.

Data Usage Patterns

After June, Surf users consistently exceed their plan limits, particularly for data usage, creating revenue variability. Ultimate users demonstrate more consistent usage patterns and rarely exceed their allotted data, resulting in stable, predictable revenue and lower strain on network resources.

Statistical Validation

An independent samples t-test (α = 0.05) revealed statistically significant differences in average revenue between plans, with a p-value of 2.86e-08. This provides strong evidence to support prioritizing Ultimate in marketing decisions.

Business Recommendations

• Allocate more advertising budget to Ultimate — the higher-value, more predictable plan
• Target high-usage Surf customers for upgrades — especially those with large data overages
• Reevaluate plan structure — consider a mid-tier plan or higher data allowances to better capture heavy users
• Geographic targeting — statistical analysis identified significant regional differences (NY-NJ area) suggesting the need for localized marketing strategies
• Explore pricing adjustments — overage fees could be revisited to improve revenue predictability while remaining competitive

Analytical Approach

• Data cleaning and preprocessing to ensure data quality
• Monthly aggregation of customer usage patterns
• Exploratory data analysis of calls, messages, data consumption, and revenue
• Comparison of customer behavior across Surf and Ultimate plans
• Statistical hypothesis testing using independent samples t-test
• Regional analysis to identify geographic revenue variations

Key Insights

• Ultimate is the stronger revenue-generating plan and should be prioritized in marketing and pricing strategy
• Surf users more frequently exceed plan limits, leading to higher revenue variability driven by overage charges
• High-usage Surf customers represent a clear opportunity for conversion to higher-value plans
• Ultimate users benefit from predictable costs, while Surf appeals to lower-usage customers seeking lower base prices
• Regional differences suggest the need for geographically targeted strategies rather than one-size-fits-all approaches

Tools & Technologies

View Full Project

Explore the complete notebook with all code, detailed statistical testing, visualizations, and business recommendations:

View on GitHub →