The pandemic will continue to crush customer call centers. New technology can improve productivity, boost training and reduce call volumes.
When COVID-19 shut down much of the world’s economy in March, it didn’t shut down the influx of customer service needs. Whether at home or unemployed, many suddenly had more questions and more time to wait on hold while using other means of connection outside of the phone, including websites, apps, text, social media, and email. Meanwhile, companies struggled to transfer complex in-office operations to remote locations, oftentimes at home.
Now that the U.S. is reopening, everything is changing once again, driving new customer questions.
We’re in an environment where change is the only thing that is real. No one knows what ‘next’ means or when it is going to happen. Organizations have to adapt quickly in order to stay afloat.
– Kathy Sobus, Senior Director of Customer Experience Strategy, ConvergeOne
Artificial intelligence is emerging as a popular tool to handle customer service requests and manage contact centers. Chatbots have grown in popularity, both those that are customer-facing and agent-facing. Over the next two years, expect to see more bots, or virtual assistants, rolled out by organizations that address customer questions via text, Facebook messaging and chat. Research firm Gartner projects that by the end of this year, Chatbots will handle as many as 85% of all customer service queries. In fact, CIOs identified them as the main AI-based application used in their enterprises—a 160% increase in interest over the year prior, due to an increase in demand for customer service, knowledge management and user support, according to Gartner.
There's a good reason: These bots can surely take the pressure off agents. Some predict that chatbots could cut business costs by as much as $8 billion per year in certain industries. These chatbots can provide self-service capabilities for customers and even assist less-experienced agents who need on-the-job advice while helping to solve customer needs 24/7. Agents working from home can no longer tap teammates sitting next to them when puzzled about a customer query. What happens in most cases is that they “guess” at what they should be doing, which oftentimes creates errors. More conscientious agents will ask questions via IM to their buddy. But because their buddy is also busy with a customer, this process takes longer to answer and interrupts other customer calls. This scenario, which is repeated all over the world, causes a trickle-down of slower, less satisfactory customer support, says Sobus.
The big game-changer for chatbots is natural language processing, a branch of artificial intelligence that allows computers to process, analyze and make sense of human language and text. Ultimately, the technology makes bots conversational and makes it more difficult for customers to distinguish whether they’re interacting with a real person or a bot.
As large companies like Apple and Google rush to make conversational virtual assistants that can respond quickly to customer inquiries and offload the toll on human agents to artificial intelligence, the global chatbot market will continue to grow fast, with some projecting the market will surpass $1.34 billion by 2024.
Even a simple chatbot can ease the pressure on your customer service team, says Sobus. A basic chatbot with 50 common questions and answers can be built in five days for an affordable price, she says, dramatically saving time and boosting customer satisfaction—both amid today’s pandemic remote workforce and anytime there’s constant change.
Agent-facing AI is also being used to assist in onboarding, training, form filling, compliance or disclaimer reminders, and the like. For years, contact centers have used classroom-style role-playing and one-on-one coaching to train agents—helpful, but they can be time-consuming and inconsistent because there are so many human factors involved that make it a complicated training process.
Over the last few years, more companies have embraced AI to keep agents engaged—which can be key when turnover at contact centers can be as high as 25% to 35%. AI-powered gamification software rewards employees for hours worked, lessons learned or speed to answer questions. For instance, a number of companies now use such software to motivate employees. They can track performance, including the number of cases resolved, average handle times, and aims to increase collaboration and can help agents learn what to do, how to do it and respond to customer enquiries faster.
Meanwhile, digital buddies or virtual employee assistants can help with productivity by cutting down on the time spent on repetitive tasks that don’t require human judgment, such as filling in documents, alerting employees about what needs to be done next, and even writing emails and embed case details. The bots can fill out HR forms, handle IT issues, and even walk agents through the customer interaction, providing guidance and performing tasks like filling in information in many systems. This reduces errors and increases both agent and customer satisfaction.
By 2021, as many as a quarter of digital workers will be using these kinds of virtual employee assistants, according to estimates by Gartner, a marked leap from just 2% in 2019.
Proactive Customer Service
Despite the AI hype, call centers can reduce their call load by simply getting to know their customers better, says Seth Earley, the author of The AI Powered Enterprise: Harness the Power of Ontologies to Make Your Business Smarter, Faster, and More Profitable. By understanding and anticipating your clients’ needs, you can organize your chatbot or your website to easily address them—and ultimately reduce the number of phone calls your call center receives.
“Don’t go out and just build a bot,” says Earley. Instead, harness customer data to truly understand them and their needs. Then, use your website to more intuitively and proactively answer those needs so customers don’t have to contact a real human agent. This means ensuring your site can be easily searched—and that customer data is sliced and diced to ensure each person is shown the most relevant search results.
Smart companies will successfully offload to the Internet any kind of rote service—think: checking on an account balance, resetting passwords, making a payment, understanding new policies and procedures due to our changing situations. That leaves only the calls that need human cognition and decision-making. You then turn your customer support into a customer experience, Earley says. “You have to understand them, understand their needs.”
Earley’s book highlights Allstate as an example of a company that did this well, ultimately allowing agents to spend more with their customers and build a stronger relationship. He also tells the story of a precast concrete manufacturer that requires complex details and requirements for custom-fabricated products. Salespeople once spent time walking their customers through sometimes very routine and repeatable configurations. But when those routine products could be ordered via an online helper tool, it freed about 24% of sales peoples’ time. “With more than 50 salespeople, that was like hiring 12 new headcount,” says Earley. “The salespeople could spend more time actually selling, while the tool became the order taker.”
More companies are also implementing AI to anticipate customer questions, rather than waiting for them to go to the site to ask an agent. Instead, the company automatically sends anticipated wait times, cancelations, and alternative solutions to them without any human intervention. The airlines pioneered this technology, by anticipating customer questions like, “Where is my baggage?” and “Is my flight delayed?”
Using Biometrics, History and Data to Improve Call Centers
Call center AI is getting more sophisticated, thanks to AI and biometrics. It may sound like science fiction, but your voice can provide hints about who you are—such as mood, social status, ethnicity or weight. New software that learns from customer service training recordings can detect micro signatures in a voice that can tell agents more about a person. This kind of biometric hardware and software is expected to hit $15.1 billion in revenues by 2025, according to a market report by Tratica, a Boulder market research firm that analyzes automation trends.
New AI systems also analyze customer behavior online—clicks, views and purchases—and previous conversations with agents to assess their personalities and predict needs and desires and route a person's call or chat to the appropriate agent. Tools also include emotional analytics, which identifies a customer’s mood by analyzing verbal and non-verbal communication. Companies might use the technology to understand a customer’s experience and even spot potential problems that caused a negative reaction. Clues like these also provide organizations with information on the “next best action” that should be recommended on behalf of the customer.
No doubt there is opportunity ahead for call centers to improve customer and agent experiences. The pandemic is causing all organizations to retool their processes, procedures, and customer interactions. AI is helping many companies alleviate this burden by offering another means to help both customers and agents. Says Sobus: "AI has enormous potential to increase customer satisfaction and ultimately improve the overall efficiency of contact centers."