One place where large amount of data is generated is the Call Centres. The moment a call is made to the call centre a large number of records are created as Interactive Voice Response System (IVR) and Automatic Call Distributor (ACD) use data to monitor the flow of calls. The call centres are nowadays recording these calls for various purposes and in turn creating big data. Tools are being used by experts to analyse this big data which is used for different purposes like improving overall efficiency, improve the quality of calls and measuring the performance of the call centre executive etc. Some of the data created by call centres are:
- Operational data which can be classified into operational customer data and operational agent data
- Marketing or business data
- Human Resources data &
- Psychological data
There are various types of data models which are used in analysing this data. The analysis can be divided into three. They are:
- Descriptive Analysis
- Explanatory Analysis &
- Theoretical Analysis
One of the extensively used tools for the study of call centres is the Queuing theory. Poison process is one such statistical tool used to predict the arrival of calls. This model was used when the data generated by call centres was used only for understanding the quantitative data regarding nature of calls received, average time of each call, average call holding time etc. Nowadays the requirements of call centres in terms of analysis are changing because of large scale integration of multiple processes.
The call centres are now integrating new technologies like Dashboard and Cloud Computing to face the challenges arising out of multiple locations, real time adherence etc. The business models adopted by call centres are also changing and they are moving away from being just voice based service providers. The call centre agents have to handle multiple issues thereby increasing the challenge for experts in analysing large quantity of data.
The call centre companies have now started using new models of data analytics with the change in the business models. They have started using tools like sentiment and text analysis on unstructured data which is created in call centres. Recently IBM has started using a tool called Text Analysis and Knowledge Mining (TAKMI) tool in its call centres to:
- Understand the key customer issues
- Make out trends and patterns
- Detect early warning signs
- Deeper analysis of agent records
- Identify best practices of agents
- Identify the causes of dissatisfaction
- Isolate revenue related calls
- Identify the reasons of why costly repeat calls are made
There are many call centres who cannot afford big data analytic tools which will need a huge investment. They can make use of even tools like low cost social media listening tools to identify complaints on global issues. In the years to come we are going to see the use of more advanced data analytics tools like IBM’s Watson as the complexities of the call centre business increases.