In contemporary marketing, the phrase “at scale” usually means big or more. But in business, it’s becoming clear, post-pandemic, that more doesn’t always mean better. And for customer service (CS), this is especially true.
Because modern contact centers are grappling with a convergence of challenges in the wake of COVID, an unprecedented percentage of customers are now using digital channels for shopping, messaging, and, yes, engaging with customer service. With this mass adoption of digital, expectations of customer service have spiked as well. And, right or wrong, new pressures are being placed on contact centers to help make up for lost revenue.
It makes sense, then, why some are throwing the kitchen sink at problems inherent to this convergence. More apps. More data. More options. More artificial intelligence (AI) functionality like natural language processing, bots, and machine learning. But while AI technology is part of the way forward, it’s not a solution unto itself. Instead, scaling for success with modern CS involves using AI to augment human agents and the work they do in call centers while providing feedback, guidance, and coaching in real-time.
The results are striking. Done correctly, the potent combination of customer service AI, human empathy, and training is doing a whole lot more with a whole lot less.
Customer service AI creates opportunities to scale service excellence in real-time across the customer journey. Preemptively, AI, automation, and increasingly sophisticated self-service options ensure the customer issues that reach agents are those for which real-time customer care has the most impact.
Issues like forgotten passwords are important, but the potential “wow” factor inherent in these more transactional types of interactions is low. What’s more, in these cases, traditional forms of AI are increasingly able to provide service excellence on their own. In 2020, 54% of customers surveyed reported regular interactions with AI-enabled chatbots and digital assistants. Of these customers, 49% viewed these interactions with AI as trustworthy, compared to just 30% two years prior.
Unsurprisingly, the impact of making it easier for agents to address customer issues is significant. Here at Cresta, we discovered that agents spent up to 30% of the conversation on verification and basic information search. On the contrary, some contact centers leveraged AI to return this time to agents. Doing so, along with the use of AI-driven coaching and expertise solutions assisting agents, resulted in a 2.3x better first contact resolution rate (FCR) than for centers that did not.
Emerging applications of AI are enabling organizations to take proactive approaches to customer service. As an example, a leading airline built a machine-learning system to measure satisfaction and predict revenue on a daily basis for 100 million-plus customers. This was no small feat, requiring a small army of data scientists, customer service experts, and external partners.
But once completed, the airline could easily identify and prioritize customers at risk of experiencing delays or cancellations. Agents could then reach out and connect with these customers in real-time, offering personalized compensation and support. This customer care initiative created an 800% increase in satisfaction and reduced churn for priority customers by 60%.
In the wake of COVID, customers expect more understanding, attention, and empathy from contact center agents. And, left unsatisfied, these same customers are increasingly willing to jump ship and shop competitor brands. But conversely, the subsequent mass consumption of digital has crippled sales opportunities in brick and mortar locations. And, right or wrong, businesses are looking for contact centers to make up a big part of the difference.
This need to make up for lost revenue has proven difficult for CS veterans, many of whom have made a career measured only by the level of service excellence provided. For these contact center agents, the expectation to master selling relatively overnight was, at best, counter-productive.
Fortunately, modern customer service AI platforms can learn sales best practices and the behaviors of a top sales performer. This “superstar” modeling can then be leveraged to transfer knowledge and augment customer service agents, in real-time, with coaching that helps them deliver excellent customer service while also identifying and seizing opportunities to drive revenue. In the last year, we have seen this “service-to-sale” motion gain popularity.
The effects of this real-world “rising tides raise all boats” approach to customer service enablement can be eye-opening. For instance, the customer service team for a leading internet service provider increased their add-on sales conversions by 38% in the first two months by using our own platform, Cresta. Moving forward, they increased add-on sales by 2.7x overall in their next quarter compared to their pre-Cresta performance.
Finally, the impact of real-time customer service, when expertly executed, extends beyond the moment of resolving customer issues. It can also increase brand loyalty and the chances for repeat business. This is especially true in recent years, as research shows customers value “average” customer service less over time. Even then, customer satisfaction alone is no longer enough to foster brand loyalty.
So, the key to customer retention involves linking what customers say to what they actually go on to do. And with real-time AI at the core of your customer care offering, it’s possible to scale beyond the moment to impact customer service in the future.
This brings us to predictive behavior analysis—yet another way to optimize customer service. Customer service-focused AI solutions can now employ pattern logic and sentiment analysis to understand how a customer’s experience will affect their future behavior as a customer. Organizations are using customer emotion, as detected by AI, as an indicator of the customer’s likelihood to renew or churn at a future date. Long silences or voices speaking over one another underscore AI’s understanding of anger or frustration. Or distress identified at the beginning of a call could signal very different things than distress identified at the end of a call.
This ability to “predict the present” improves the effectiveness of customer journeys and algorithmic learning loops that optimize themselves over time, ultimately improving customer service, business outcomes, and a business’s ability to forecast the future.
At the end of it all, customer service has to meet customers where they’re at. And while meeting them in the present was once the ceiling, it’s now become the floor. But know you’re not alone in facing these challenges—real, lasting change is being realized through contact center AI.