Move Up the Value Chain for AI in Customer Service

Artificial intelligence is the topic of the day for customer service. But let's define what AI really means. Pure AI, often featured in science-fiction movies and doomsday scenarios, aims to build machines whose overall intellectual ability is indistinguishable from or even surpasses that of humans. These machines will learn, reason, problem-solve, perceive, and understand language. But, even with early excitement and five decades of investment, only slow progress has been made as pure AI represents a hard problem.

However, applied AI—dotherwise known as pragmatic AI—has a different history. Applied AI aims to produce commercially viable "smart" systems. Today, applied AI pervasively powers consumer experiences. Amazon and Netflix recommend products based on our historie., Waze and Google get us to our destinations more effectively. Lyft and Uber let us know when our cars will arrive.

Applied AI is not a single technology. Instead, it's composed of discrete technologies that either individually or in combination, are advanced enough to add intelligence to applications. They can learn, predict adapt, and potentially operate autonomously. This intelligence produces quantifiable business outcomes that can be exploited today. So how do you get started? Follow the value chain of AI for customer service.

Step 1: Use AI for improved customer service efficiency.

Use speech or text analytics to automatically classify cases and assign disposition codes, shaving dozens of seconds off handle times per call. Improve inquiry routing by using predictive models based not only on agent skills but also on real-time analysis of behavioral characteristics, like performance, personal strengths, and communication styles, to improve contact times and customer experiences.

Step 2: Use AI to make customer journeys easier.

Use real-time satisfaction predictors for incoming incidents to identify in-flight issues and customers who need immediate attention. Under the covers, AI models calculate satisfaction scores from attributes, such as wait times, reply times, incident details, and effort metrics, and then decide which actions to take if they receive poor scores. Go a step further and create a health score from customer and product usage data that tracks the success of onboarding activities and long-term usage. Big data analysis surfaces key elements that inhibit adoption and lead to customer churn. Triggers alert organizations to sudden changes in their health scores so they can intervene to deepen customer relationships.

Step 3: Use AI to better empower customers.

Use knowledge management solutions to arm agents and customers with answers. Modern knowledge solutions use pragmatic AI capabilities, like natural language processing and text analytics to extract topics and automatically classify ingested content to understand the intent of queries; they use machine learning to optimize the organization and relevance of search results based on customer profiles, histories and context. Use AI-powered chatbots to automate customer conversations and help agents be more effective and get smarter over time.

Step 4: Use AI to proactively—even pre-emptively—engage customers.

Track and intervene in customer journeys via an invitation to chat or co-browse at points of struggle. Intervene opportunistically at points in the journey best suited for customers to accept a recommendation, coupon, offer, or additional advice. Do this by using intent models to determine the best outcomes and machine learning to refine them over time.

Kate Leggett is a vice president and principal analyst at Forrester Research, serving application development and delivery professionals.