Evan Kohn is a leader in digital marketing and customer experience and Chief Business Officer at Pypestream. There he developed PypePro, an AI onboarding method used by Fortune 500 companies.
Companies have long relied on web analytics data such as click rates, page views and session lengths to gain insights into customer behavior. This method studies how customers react to what is presented to them and to reactions that depend on the design and the copy. However, conventional web analytics do not exactly capture customer requirements. What about the way companies are promoting broader customer experience (CX) as marketers push for predictive analytics?
Executives are increasingly turning to conversational analytics, a new paradigm for CX data. The focus will no longer be on how users react to what is presented to them, but on what "intention" they convey through natural language. Companies that can capture intent data through conversational interfaces can be proactive in customer interactions, deliver hyper-personalized experiences, and better position themselves in the marketplace.
Direct customer experiences based on the customer's disposition
Conversational AI, which powers these interfaces and automation systems and feeds data into conversational analytics engines, is expected to grow from $ 4.2 billion in 2019 to $ 15.7 billion in 2024. AI can inform CX decisions not only about how customer journeys are structured – such as curated buying experiences and purchase paths – but also about how the entire product and service offering can be further developed. This advance in knowledge could become a decisive factor and competitive advantage for early adopters.
Nowadays there are huge differences in the level of complexity between conversational solutions, from simple chatbots with only one task to secure, user-centric, scalable AI. In order to unlock meaningful conversation analytics, businesses need to ensure they have provided some key ingredients that go beyond the basics of analyzing customer intent with natural language understanding (NLU).
While intent data is valuable, companies will improve their engagements by collecting sentiment and sound data, including through emoji analytics. With such data, automation can adapt to a customer's disposition. So if an overdue invoice is found to be an issue, a quick path to resolution can be provided. If a customer expresses joy after purchasing a product, AI can respond with an upsell offer and collect more acute and actionable feedback for future customer trips.