This project analyzes inbound call center data to evaluate call volume patterns, customer sentiment, and service response performance across multiple channels and call centers. Using operational metrics such as call distribution, response time categories, and sentiment trends, the analysis identifies peak demand periods, service inefficiencies, and customer experience gaps. The insights generated support data-driven decisions for staffing optimization, performance improvement, and enhanced customer service delivery. This project demonstrates how Excel-based analytics can support operational decision-making in a call center environment.

• Microsoft Excel: (Data cleaning, PivotTables, Dashboard)
• GitHub: (Version control and documentation)

• Data cleaning and preparation
• Exploratory data analysis (EDA)
• Business-focused dashboard design
• Translating data into actionable insights

The call center receives a high volume of inbound customer calls across multiple channels, locations, and time periods. This analysis aims to understand call distribution, customer sentiment, and response-time performance to identify operational bottlenecks and opportunities to improve service efficiency and customer experience.

Dashboard snapshot provided below summarizing call volume, sentiment, and response-time performance.

• Created a working copy of the dataset to preserve the raw data
• Checked for duplicate records (none found)
• Converted Score and Call Duration to numeric format (no decimals)
• Split call timestamp into date format using Text to Columns
• Added a helper column (Call Day) to extract day from call timestamp
• Converted the dataset into an Excel Table for easier analysis and pivoting

The analysis focuses on call volume distribution, service performance, and customer experience.

• Calls by sentiment
• Calls by reason
• Calls by channel
• Calls by call center
• Calls by state
• Calls by day of month

• Calls by response time
• Response time distribution by call center

• Sentiment distribution by call center

• Total inbound calls

• Call volume peaks on specific days of the month, indicating predictable demand cycles
• Certain call centers consistently experience slower response times, contributing to negative sentiment
• Phone and chat channels account for most inbound calls, suggesting opportunities for self-service optimization
• Negative sentiment is concentrated around a small set of call reasons, Pointing to process or communication gaps.

• Optimize staffing during peak periods identified through call volume by day and response time analysis.
• Improve response times at call centers with higher volumes of slow responses to reduce customer frustration.
• Investigate drivers of negative sentiment, particularly at call centers where unfavorable sentiment is concentrated.
• Prioritize high-volume call reasons by improving self-service options or proactive communication to reduce inbound demand.

• Analyze agent-level performance to identify training and workload balancing opportunities.
• Incorporate average handling time and abandonment rate for deeper service performance insights.
• Extend the analysis to include time-of-day patterns for more granular staffing optimization.
• Build predictive models to forecast call volume and improve capacity planning.

The dataset used in this project was obtained from Kaggle and is publicly available for educational and analytical purposes.