🎧 Call Center Customer Satisfaction & Operational EDA — Python

An exploratory data analysis (EDA) project evaluating inbound call center traffic, service level agreements (SLAs), handle times, and customer satisfaction (CSAT) scores across communication channels.

📋 Business Problem & Objective

Call center operations needed to identify key friction points causing customer dissatisfaction, determine optimal staffing levels across support channels, and evaluate agent resolution efficiency against SLA targets.

📊 Dataset & Scope

• Source: Enterprise Customer Service Ticket & Interaction Dataset (data/call_center_sentiment_data.xlsx).
• Scale: 10,500 customer support interactions across Phone, Chat, Email, and Web Portal channels.
• Attributes: Ticket_ID, Channel, Customer_Sentiment, Handle_Time_Min, Resolution_Status, Agent_ID, CSAT_Rating, Queue_Wait_Sec.

📈 Operational Performance Overview

Ticket Volume & CSAT Ratings by Channel

🔬 Methodology Summary

1. Data Preprocessing & Type Cleaning: Formatted timestamps, normalized missing CSAT ratings, and filtered outlier handle times.
2. Channel Benchmarking: Compared ticket volumes, average speed of answer (ASA), and first contact resolution (FCR) rates.
3. Handle Time vs. CSAT Correlation: Quantified customer satisfaction drop-off curves relative to queue wait times.
4. Agent Tier Productivity: Visualized call resolution rates across junior, intermediate, and senior agent tiers.

🔍 Key Findings & Insights

• Channel Performance: Live Chat recorded the highest CSAT rating (4.42/5.0) and fastest average resolution time (4.2 minutes).
• Handle Time Threshold: Identified that handle times exceeding 8 minutes cause an immediate 45% drop in CSAT ratings.
• Channel Volume Distribution: Inbound phone inquiries represented 46.2% of total ticket volume, followed by Live Chat (29.7%) and Email (16.0%).

💻 Tech Stack

• Language: Python 3.11+
• Libraries: Pandas, NumPy, Matplotlib, Seaborn, Jupyter Notebook

🚀 How to Run / Reproduce

1. Clone the repository:

   git clone https://github.com/mailmetanmaymandal92/Python-Call-Center-Satisfaction-EDA.git
   cd Python-Call-Center-Satisfaction-EDA

2. Run the Jupyter Notebook:

   pip install pandas openpyxl matplotlib seaborn jupyter
   jupyter notebook notebooks/call_center_satisfaction_eda.ipynb

---

👤 Author & Contact

Tanmay Mandal • Data Analyst | Business Intelligence & Analytics Specialist

• 🌐 GitHub: mailmetanmaymandal92
• 💼 LinkedIn: Tanmay Mandal
• 📧 Email: mailme.tanmaymandal@gmail.com