Online Support Evolves With AI, NLP, and Machine Learning
AI, NLP, and Machine Learning software can be used to help improve customer experience. Inbenta is using these tools with customer support at large call centers.
Join the DZone community and get the full member experience.Join For Free
I had a great talk with Jordi Torras, CEO of Inbenta — an Artificial Intelligence (AI) and natural language processing (NLP)-based self-service software that has been around for eight years to help improve the customer experience (CX) with customer support at large call centers.
The perception of the call center has changed, as customers would prefer to find answers to their questions themselves. This is becoming easier as AI continues to develop and "learn."
Inbenta uses techniques from AI to help provide significant savings for large B2C companies that are generating a lot of web traffic. Customers include airlines, telecommunications, Ticketmaster, and Groupon. They help to ensure the knowledge base FAQ matches users' questions with canonical questions and provides a chatbot user interface.
Inbenta indexes content that describes abstract situations and matches content using NLP to understand the question and provide the right answer. This reduces the time and effort needed to train the system while the system expands content to be able to answer more questions with greater relevance and accuracy.
This frees customer service reps (CSRs) to work on higher-level problems while knowledge managers and content editors can identify gaps in the knowledge base that need to be filled.
Inbenta develops sites with web forms inspired by paper forms, collecting name, number, model — all of the information necessary to build a 360-degree view of the customer and a better CX by only requiring the customer to enter their data one time and not having to repeat it. This enables the customer to have an intelligent conversation with the CSR, agent, or chatbot while the system is constantly learning more about the customer and product or service issues.
The return is significant. Incoming email volumes can decline up to 80% in the first three months, while chat can reduce the number of incoming calls by 5%—30% while customer satisfaction increases. Customers avoid search (which is typically inaccurate) and calling (which involves wait times), and have a better experience with live chat that works 24/7/365.
The Googlization of customer support is resulting in long-tail SEO so that the content of chatbots can be indexed by Google. As a result, customers can find answers to specific questions more easily.
Opinions expressed by DZone contributors are their own.