As we enter a new year, it’s natural for customer experience (CX), sales, and customer service leaders to ask: What went well? What didn’t? And where can we improve? If you’re the “futurist” type, you might even ask: should I adopt AI to improve business results next year?
The problem with AI today is that it’s hard to discern between fact and fiction. In CX, we see companies using AI to augment human interactions, increase customer satisfaction, and respond to customer service issues faster with minimal or zero need for a live customer service agent.
In sales, managers, and reps tout AI as the new way to forecast sales, boost sales engagement to drive bottom-line results and orchestrate live calls with customers. But, while it sounds forward-looking, you can’t throw AI at just any problem.
Take containment — a long-heralded and oft-misused metric — as a prime example of where AI can go wrong. High containment rates are often deemed good because it means that automated systems — think AI chatbots — resolve customer issues without the need for human interaction. This means the company saves on resources.
But keeping customers away from reps is not always good for the customer, nor for add-on sales. If the AI chatbot isn’t properly designed or trained it can frustrate callers. Can a well-designed system work in these situations? Yes. But it has to be done very carefully. To be clear, it’s not all bad.
Here are some examples where AI is showing promise.
Boosting sales & revenue with AI
Companies have realized AI has tangible and practical applications in sales and, if used correctly, can drive gains on the bottom line. The analyst community agrees. Gartner research predicts that 70% of customer experiences will involve some kind of machine-learning component in the next three years.
AI can force-multiply sales reps by automating repetitive tasks like CRM data entry and scheduling meetings. It helps reps prioritize, so they can focus on the high-probability, high-payoff deals, and customers that matter most. AI can also provide analytics on communications between sales reps and potential clients including emails, phone calls and chats. Other use case examples include sales forecasting, lead qualification, routing/matching, and coaching sales and CX reps.
Fueling contact center performance
Customer expectations often intensify after the holidays when consumers seek out deals. Now is a great time to find out if your contact centers are prepared to deal with demanding customers who may be stressed out as holiday bills catch up to them during tax season, or by slow turnaround on items they returned several weeks back – or sticker shock when services and renewals leap in price.
Enter predictive call routing, which helps improve CX by looking at past customer behavior and preferences to predict the skills and personality traits an agent needs to provide the best service possible. In this use case, AI cuts escalation time and improves satisfaction for both the caller and the agents.
Today, there are many proven examples where AI is transforming the customer experience. AI customer service bots and virtual employee assistants like Cortana are becoming more commonplace and capable. According to Salesforce’s recent State of Service report, more than half of service providers will be adding chatbots to their lineup in the next year and a half.
Contact centers continually deal with the challenges of training agents at scale. Coaching and development for agents is a top priority, and AI can help by analyzing agent metrics and caller responses to recommend specific on-demand learning and reinforce great performance. AI-enabled performance intelligence helps managers identify issues with agent performance sooner, and in a survey we conducted, 85% of managers said this feedback allowed them to manage more people simultaneously.
AI will get better for sales and more
Going to the tried-and-true baseball analogy, we are probably only in the second inning of AI utilization. AI will become progressively more useful to CX, CS, and sales teams in the coming years because the source data it consumes will become richer as departmental data silos become further integrated and more broadly shared enterprise resources.