Skip to content

pallavising48/customer-support-analytics

Repository files navigation

Dashboards (Tableau Public)

customer-support-analytics

Customer support call analytics, forecasting, and vendor optimization case study

Executive Summary

  • Analyzed 4.8M+ customer support calls across shared and dedicated queues to understand 2024 volume patterns, vendor performance, and service quality.
  • Found that Vendor A handled ~59% of volume with 90.5% SLA at a lower cost per call, while Vendor B achieved higher satisfaction (~39.5%) despite weaker SLA.
  • Quantified that outages occurred on ~29% of days but reduced SLA by <1 percentage point, showing strong operational resilience in the support model.
  • Built a Q1–Q2 2025 forecast (~2.5M calls) with event-based adjustments (e.g., Easter, Mother’s Day) and translated it into staffing plans of ~1,200 agents in Q1 and ~800 in Q2.
  • Recommended a phased shift in vendor allocation toward Vendor A to capture cost savings while maintaining an overall SLA target of 88%+ and tracking satisfaction closely.

Excel Version

In addition to the Python notebooks, this repo includes an Excel-based implementation of the same analysis and forecasting logic, plus the original written report:

  • excel-solution/Customer_Support_Forecast_Excel.xlsx – 2024 summary, Q1–Q2 2025 forecast, and staffing model built entirely in Excel.
  • excel-solution/ Word report – case-study write‑up answering the original assessment questions.

About

Analytics & forecasting of customer support calls (vendor performance, SLA, staffing)

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •