Last week, I was alarmed to get a text message from my LPG supplier that my ‘cash memo for the refill I had ordered had been cancelled’. Did it mean I wouldn’t be getting my next cylinder for some reason? I called the company and pressed a button in the interactive voice response system (IVR) to learn about the current status. A recorded message told me the query could not be answered at the moment and the call got disconnected. I called again and pressed another button to make a complaint.
I got the same response. There was no option to speak to a customer service executive so I called my gas supply agency to follow up on the problem.
Sounds familiar? Many times, it’s hard to get through to an actual person on the other side. When you do, your problem does not always follow a script familiar to them. Often, the person can sound like an automaton themselves. While most of them are polite and helpful (in my experience), they fall short of the capability to deal with unfamiliar problems and are also anxious to end the call, probably to keep in tune with some efficiency metric.
Shashank Mehrotra, General Manager and Business Head of web hosting company Big Rock, which hosts websites for small and medium businesses, says average handling time (AHT) of over eight minutes spent on a call is considered a sign of inefficiency on the part of a customer service agent. Most companies use metrics such as AHT, first call resolution (FCR) and net promoter score (NPS) to rate efficiency and impact.
Firms balance cost considerations against the number of people needed to solve customer problems and get the NPS right. NPS is a measure of how likely customers are to recommend the company on a scale of 0-10. Mehrotra says, “Interestingly, we find that agents who have a very high AHT have a high NPS score as well – customers tend to think that an agent treating them with patience is a sign of good service.”
In these days of automated customer service, it is only natural to wonder whether reaching a real person through the maze of IVR systems is a measure of efficiency too. Vivek Mathur, Chief Commercial Officer, Vodafone, says it takes 8-10 seconds at his company. He, however, makes the point that many customers want speedy solutions, not human interaction, and much of this is enabled by automation. The automated menu of problems and solutions is topped up as and when new situations arise, he explains. All this automation goes into freeing up resources for when actual help is required, says Mathur.
What’s the script?The scripts for conversations with customers are also updated. Whenever there is a new problem, it is taken on board. The agent’s computer screen will show him the pertinent questions and answers to ask and offer and the goal is for the next customer with the same problem to have an easier time of it.
Big Rock’s Mehrotra points to instances where frustrated customers get unreasonable and abusive. The service agent is expected to keep quiet and listen, which is not fair, he says, adding that in such cases, his company empowers its staff to shut down the conversation. “It behoves the customer to be professional as well. Rude behaviour puts off the agent too,” he says, pointing out that in India, customers expect agents to be servile.
But won’t the shutting down result in a deadlock? The problem can be escalated, says Mehrotra, and the company has to provide the means to do so by letting customers know whom to approach. Mehrotra earmarks 30 minutes a day to read complaints and queries. “Having a senior review and reach out to the customer helps solve problems faster,” he asserts. That said, it helps to employ people with good attitude. Vodafone’s Mathur says his company hires people with a positive outlook and supplements it with training and refresher courses to help them perform beyond customer expectations.
Ashwin Mittal, who is President of market intelligence agency Blueocean, says customer service can benefit from big data analytics which segment and link customers and agents based on the nature of the complaints. This means that when Customer A calls, based on past interaction (satisfaction level, nature of the problem and even the kind of customer she is) data analytics put her on to the most suitable agent to resolve her problem.
There are newer technologies to help make customer service more human and efficient, adds Mittal. Speech analytics, where the computer can listen in on the conversation and suggest a solution, is one. Proactive customer service, where a company can reach a customer even before they call with a complaint, is another – enabled by mapping past calls with current behaviour, in the case of credit card companies, for instance. Lakshmi Rao was recently advised by her credit card company to cancel the current one for “security reasons” and apply for a new one, advice that she took. Proactive? She thinks so, though she does not know why, and neither did the agent who answered her call. But this was achieved at a much swifter pace than her dealings with another credit card company which blocked her other card for security reasons over a year ago. Admittedly, she did not notice it as she used it only rarely, but when she did, had a hard time getting through to the call centre, an agent and finally, getting it reissued.
Oh well, all’s well that ends well!
Vitamin C is a weekly dose of consumer empowerment