If you’ve ever had to call a customer help line or other call center, you probably experience a shiver of dread at the mere thought. Some customer help lines are so reviled that people would rather deal with the issues than try to get support.
But all that may be changing as companies use advances in data collection, data analysis, and artificial intelligence (AI) to improve the call center experience for customers and provide valuable insights for companies.
A wealth of data
You’ve probably heard the recorded message that “your call may be recorded for training and quality purposes,” but companies are starting to realize that those recordings are valuable for a great deal more.
With the advent of natural language processing (NLP) technologies and the ability to understand more unstructured data, like phone call recordings, companies are sitting on a wealth of information every time they record a call.
And because the field is expanding so rapidly, there are many different ways companies can put this data to use.
Speech analysis and NLP
Natural language processing is what allows an automated system to properly direct your call when you speak into the phone. It used to be that you had to use the exact word or phrase, saying “accounts” or “operator” in order to be routed properly. But with the growing sophistication of NLP algorithms systems can now interpret long strings of words, like “I have a weird charge on my bill I want to talk to someone about,” and route that caller to the correct department.
Speech analysis goes beyond what you say to understand how you say it. For customers, speech analysis can analyze the caller’s tone, vocabulary, sentiment, and even silences to gauge emotion and satisfaction. It can detect a caller’s age, which could determine the success of an age-related campaign or direct the caller to a particular agent; studies from Software Advice show that older and younger people have different preferences when it comes to the type of agent they want to deal with.
On the agent’s side, speech analysis can listen for certain keywords or phrases and prompt the agent accordingly. It can also be used to identify gaps in an agent’s knowledge, types of calls they are least comfortable handling, and other factors that can influence training and call routing.
Predictive analytics
Speech analysis, combined with predictive analytics, can also detect when a caller is on the verge of getting frustrated or angry — or even predict when a caller is lying or trying to commit fraud.
The predictive models can provide insights on the best ways to handle different types of calls, in order to boost an agent’s effectiveness and improve the customer’s experience on the call. One company, Mattersight, has developed a the solution that uses algorithms to identify the best agent to handle each caller based on the agent’s past performance and personal strengths, and the customer’s personality and other behavioral characteristics.
But it also goes outside the call center. When call center data is combined with social media data, companies can watch for complaints about specific problems, and predict a higher volume of calls for that reason. For example, if Twitter suddenly lights up with complaints that phone service is out in a certain region, the phone company call centers could be alerted to expect a higher incoming call volume, and could even change recorded messages to include information about the outage.
Another way big data is changing the call center industry is through HR. Finding and retaining qualified, patient people to staff call centers is a huge challenge for the industry. One company estimated that it requires two hours of HR staff time to find and hire a single employee — and when you need hundreds of agents for a new center, that means the process could take months.
A company called CornerstoneDemand has come up with a data-based 'Evolve' solution for this, putting applicants through an automated application process, including recording their voice and giving them simulated calls to answer. The company then uses data analytics to predict who should get an interview, who should be interviewed with an eye toward certain potential problems, and who shouldn’t get an interview.
For a large call center, this could shave hundreds of hours off the hiring process time. And companies report that employees hired through the Evolv process are more successful and stay longer in their jobs.
As technologies continue to advance, I suspect we will see even more improvements to call centers and customer service in general, perhaps making the cliché of the dreaded help line phone call a thing of the past.
