Project 2 - Supply and Demand Gap Analysis

Project – 2 : Demand and Supply gap analysis of call centre

AIM:

• To identify the agent who performs well and with the highest satisfaction rate in the team.
• To identify the total calls answered and out of which the no. of calls resolved.
• To identify the sector which receives the highest call rate.

INTRODUCTION:

Supply and demand gap analysis is a useful tool for accessing the balance between the availability of resources and the level of demand for those resources in a given market. In the context of call centres, supply and demand gap analysis can be used to identify gaps between the number of agents available to handle incoming calls and the volume of calls that need to be handled. By examining these gaps, call centre managers can make informed decisions about how to allocate resources, including hiring additional agents, implementing new technologies, or adjusting staffing schedules to better meet demand. Ultimately, supply and demand gap analysis can help call centres optimize their operations and improve their ability to deliver high-quality customer service.

In this case study, a detailed analysis is done which is for a period of three months. Data analysis included importing data from web, data cleaning using power query editor and transformed the cleaned data into multiple visualizations in the form of charts, graphs, slicers etc. The goals of this case study is to find the best agent and also analysing the calls for different sectors during different time zones.

PROBLEM STATEMENT:

It is very essential to understand the workflow in a call centre such as, finding out the best agent so that he/she can provide few skills to the others or provide the best training. Identifying the sector or category receiving the highest number of calls, so the management can appoint more agents to that sector and work on it. To minimize the no. of unresolved cases and strive for excellence. To provide recognition or awards to the best performing agents so that it inculcates competitive spirit within the others.

The dataset contains multiple duplicate values and errors. In order to meet the goals of this case study, it is important to have the data cleaned and perform the necessary analysis and visualization.

METHODOLOGY:

• Data was imported from Kaggle and the dataset was a .csv file.
• Imported the dataset to Power query editor to perform few cleaning which was replacing the null values with 0 in the column of whole number datatype, separated the column of call duration into minutes and seconds and then merged those columns.
• Created a rough idea of how the analysis should be done depending on the goals of the case study.
• Created new columns of calls answered, calls rejected, resolved calls and not resolved calls
• Created measures calculating the percentage of calls answered and rejected.
• Data visualization was done using the calculated columns and measures.
• From the visualization, the goals of this case study are easily met.

ANALYSIS:

Data analysis is the process of examining and interpreting large sets of data in order to extract meaningful insights and conclusions. It involves a range of techniques and tools, including statistical analysis, data mining, and machine learning, that help identify patterns and relationships in the data. The following are the analysis done in this case study:

• Eight cards have been created displaying the Total no. of calls, No. of agents working, Total calls answered and rejected, percentage of answered calls and rejected calls, agent who has the highest call records and the agent who has got the highest satisfaction rate.
• A stacked column chart displaying the total calls by resolution.
• Stacked bar chart describing the calls received for different sectors.
• Stacked column chart with the sum of duration on calls by each agent.
• Pie chart describing the sum of calls answered by each agent.
• Area chart detailing the total calls by month
• Doughnut chart expressing the sum of calls rejected by each agent.
• Two slicers in which one has names of first three months of the year and another slicer describing each dates of those months.

INSIGHTS:

Call centre datasets contain a wealth of information that can provide insights into customer behaviour, call centre performance, and overall business operations. Some key insights that were obtained from analysing this call centre data included:

• Call centre performance
• Customer demographics such as the call duration, types of issues they are calling about.
• Staffing and scheduling by identifying the peak call time and adjusting staffing levels accordingly.

RECOMMENDATION:

The goal of recommendation of call centre dataset is to identify actionable insights that can help call centres optimize their operations and provide a better customer experience. Based on call volume data and customer feedback, recommendations may be made to adjust staffing levels or provide additional training to call centre agents to improve their ability to handle customer inquiries. Analysis of call centre data may reveal areas where technology can be leveraged to improve call routing, reduce wait times, or provide self-service options for customers.

CONCLUSION:

From the analysis and visualization done using power BI desktop, it was found that a total of 5000 calls were received over a period of three months and there were eights agents in total to handle these calls. Different agents performed well during each month and using the slicer, we can find out the best agents.

Total no. of calls resolved and not resolved were found using the bar chart and the management can take necessary actions according to the data. Also, from the analysis, it was found that streaming had the highest number of calls received. Number of calls rejected by each agent is also analysed using a pie chart.

Taking a summary of the analysis done, the operations of the call centre can be improved.