Analyse the descriptive, predictive and prescriptive for DialRiver data

Introduction 

DialRiver is a small call centre which provides customer services and sales for corporate clients. They  are based in a renovated warehouse in Melbourne’s outer Western suburbs with seating for 8 call  centre agents plus offices for management staff. The call centre agents are in charge of answering  customer calls, listening to their concerns and solving problems. 

The business is jointly owned by business partners Ashley, Thomas and Maryam who met while they  were students during their undergrad at RMIT and were also working part time in a call centre, so  they have some understanding of how a call centre works. Now that they have started their own  business, they are finding that managing a call centre is challenging.  

Their clients are various electronic retail businesses who sell similar categories of electronic goods.  DialRiver provides sales and customer support services for their clients by acting as their call centre  (outsourcing to a third party). DialRiver has many phone numbers which all come into the same call  centre and each phone number is allocated to a different client.  

Our client’s customers do not know that they have called a third-party call centre, as far as the  customer is concerned, they have called the retail business directly.  

The clients advertise their allocated phone number to their customers and the agents have a phone  display which displays the phone number the customer dialed (the one allocated to a client) so it is  important the agent makes note of the phone number on their display so they know which client the  call will be about and greet the customer accordingly. Occasionally this has caused some confusion.  If an agent forgets to look at the display or forgets which client the phone number was allocated to,  they might answer with the wrong client greeting or discuss another client’s products.  

Agents are in a virtual queue to receive the next call that comes through to the call centre. After each  call the agent is given at least 10 seconds before they get another call so they can wrap up any  leftover work from the previous call. If they need more time to finish a complex call the agent can  put themself at the back of the queue or make themself unavailable until they are ready. 

The call centre only operates as an inbound call centre (i.e. a customer initiates the call rather than  outbound calls where the call centre tries to generate business by calling customers). Operating  hours are 9am – 5pm seven days a week.  

DialRiver are paid a fee directly by their clients. $15 for calls that are resolved and $2 for calls which  the agent could not resolve, e.g. could not find the information the customer needed or did not make  a sale. The agent records the outcome of the call as resolved or unresolved in the Customer  Relationship Management (CRM) system to keep information of whether it is a sale or a simple  service enquiry call.  

Ashley takes care of accounting and finance, Maryam handles sales (of our call centre services to our  clients) and forecasting whilst David manages the daily operations of the call centre. Call centre agents are rostered onto variable shifts depending on how many clients they have and the number  of calls they are expecting.  

Matching the staff numbers to their expected call volumes is a key challenge of any call centre. Too  many staff and they sit idle eating into the profits. Not enough staff and the customer’s waiting times  increase making them frustrated and less likely to make purchases. The challenges of getting this  right have been a source of tension for the three partners. As the Operations manager, David is keen  to roster many staff on as he is responsible for making sure calls are handled quickly and customer  waiting times are low. Ashley is pressuring David to reduce the staff however, as idle staff reduce  their profit margins, “We can’t be paying for people to read magazines and complete their uni  assignments in between calls.” Ashley also thinks Maryam is not doing a good enough job at  forecasting the number of client campaigns and the number of calls each campaign should expect.  Maryam thinks having more people than they need is better than not having enough and is already  concerned that they are missing too many calls as it is with a number of calls not answered. Ashley thinks there are more than enough staff, they are just not working efficiently.  

The management team has not really been paying attention to the Customer Satisfaction ratings.  Customers are directed to an automated voice prompt to rate the call (1 Poor - 5 Excellent) once they  have finished talking to an agent. The average satisfaction ratings grouped by product are provided  directly to the clients at the end of every month via an automated report without the call centre  management team’s involvement. David had a recent informal conversation with one of their clients  who was concerned about the customer satisfaction ratings. The client also queried why they were  charged for some unresolved calls with an unknown product. The client warned that without  improvements the client would be looking for a new call centre. David thinks he should have further  information before he raises the issue with the other partners and clients.  

DialRiver are not paid by the customers calling the call centre but by their clients, so they need to  make sure they are responding to their client’s needs. On the other hand, if they do not provide a  good service to the customers, then their clients will not want to continue to use their services.  

The Insights Needed for Decision-Making Ashley, David and Maryam seek your consultation to help them understand the performance of their  Call Centre and how well their business has been running. What should they be measuring? Average  Wait Time (time the customer waits in a queue before they speak to a real person) and Average Talk  Time (how long it takes the operator to complete the call once they start talking) or is it just important  to know how many calls were resolved successfully and try to improve revenue? Are there enough  or too many staff rostered on to manage the number and duration of calls? How well are they serving  their clients, and could this be improved?  

A spreadsheet has been provided to you containing call record data for the month of Jan 2021.  

Some of the data and calculations have been completed for you and you are guided to some of the  other calculations and tools. Costs have not been provided so you are not required to make  calculations and predictions regarding profitability of the business.  

A summary sheet has been started for you. You may make your calculations using any Excel  methods/tools you are comfortable with. 

You are expected to undertake descriptive, predictive and prescriptive analysis of the data that  DialRiver has provided and use Excel to analyse the data. The results of the analysis are then required  to be presented in a professionally organised report that analyses and discusses both operations so  far and the various options available to help improve future operations. Some specific expectations  are as follows:  

a) Analyse the Jan call records to identify the key variables/features (e.g. products, client  satisfaction, agents, call durations, call outcomes, etc). Has the Call Centre been well run?

b) Discuss at least 2 options to improve the performance of the variables/features identified in  (a). 

c) Predict the likely outcomes of these key variables/features for the first two weeks of Feb (assuming no changes to business operations since Jan) and then compare the effect of the  changes discussed in (b) for the same two weeks.  

d) Identify at least one business issue which the business should further explore to improve their  business. Clearly explain your reasoning using data to support your discussion. Justify how  this would improve the business. 

e) Make two specific recommendations to the owners to improve the management of the  business. Briefly justify these recommendations (i.e. prescribe what should be done and why  this would be helpful, drawing on your previous predictive and descriptive analysis).  Demonstrate the benefits to the business of acting on these recommendations.