Your wifi is out. You try resetting your modem. The reconnection attempt fails. Frustrated, you call the customer service department of your telecommunications company. The average wait time to reach a call center agent for a telecom company is 5 minutes. When the agent finally joins, she helps you troubleshoot what turns out to be a relatively simple and common issue. The total time-to-resolution, though, is close to 7 minutes.

Telecom customer experience is defined by frustration, friction and fragmentation which is why telcos, cable companies and internet providers consistently rank last in the American Customer Satisfaction Survey. In a signal of just how bad the telecom industry is, OnHoldWith.com collected more than 165,000 posts over a 12-month period mentioning the phrase “on hold with” from Twitter. The phrase “on hold with” is among the more commonly used expressions of frustration in the customer lexicon. Verizon was the most likely to be mentioned, with both AT&T and Telus showing up in the top 10.

Telecom companies are inundated with customer needs, whether it’s issues related to billing, being charged for a service a customer never signed up for, service issues, installation requests, or collecting personal information. Telecom customer service queries make up over 78% of the top five customer service queries on Google and 63% of the top 10. Every month, Verizon alone has 919,000 people try to understand what’s wrong with their Verizon service.

While ticket volume alone presents a major challenge, this is only compounded by the fact that segmentation of technology and functions within the business causes customers to be passed from one department to another. Oftentimes, customers are asked to repeat the same basic information.

Understanding mounting frustrations, telecom providers have made efforts to improve the customer experience by launching chatbots. However, many of these efforts have come up short. Can Modern AI revolutionize support and improve CSATs for telcos?

Why telecom chatbots are failing

Many telecom companies have launched chatbots: Vodafone has Tobi; Telefonica has Aura; and AT&T, Verizon and CenturyLink all offer chatbots to assist with customer service and handle initial sales queries.

Many of these, though, have not delivered on the promise of AI and chatbots because they cannot understand natural conversational dialogue. If a person asks a question in a way that deviates even slightly from the script that the bot has been trained on, the bot fails to understand a person’s intent. The problem is that everyone has their own way of asking a question, so chatbots often get confused. Comcast, for example, found that its customers had asked the simple question “I want to see my bill” in 7,500 unique word and phrase combinations. Unless a chatbot was trained to understand these 7,500 combinations, it would not be able to answer the question effectively. Most bots are not designed to handle that number of variations, and therefore result in frustrating user experiences.

The opportunity of Modern Conversational AI for CX in telecom

Since the first generation of rudimentary chatbots, there have been advances in Natural Language Processing and machine learning enabling AI-powered agents to better comprehend conversations and continuously improve over time. Modern AI platforms can now learn from past email tickets and phone logs to accurately respond to different sorts of queries and language cues and even understand misspellings, slang and short-form. This enables companies to automate resolutions to everyday issues more effectively.

Conversational AI systems can understand when it needs to ask for new information and when information requests are repetitive and annoying to customers. Whereas chatbots are used for low-level triage and redirection, Conversational AI might be used to provide a zero-touch way to:

  • Modify or upgrade customer subscription plans
  • Resolve technical problems
  • Automate repair scheduling
  • Update payment or profile information

In addition to this, conversational AI agents that connect to other business systems can anticipate when a person is likely to churn based on their behavior, customer profile attributes or other lifecycle trends, and preemptively intervene with a specific incentive to renew a contract.

Customer-centricity will be the defining factor for telecom companies

The telco industry is rapidly evolving. The cost to switch to another provider is lower than ever, and there is more competition and choice today than ever before. Just consider the rise of cord-cutting, as cable providers lose a staggering 14,000 customers per day.

Customer service now directly impacts customer loyalty. Negative customer experiences boost churn and cause brand degradation over time. Modern conversational AI can help telecoms meet increasing expectations for immediate and effortless support across a rising number of channels.

Interested in learning more about the opportunity of customer service AI for telcos? Download the whitepaper Beyond Chatbots: Conversational AI for Telecommunications Companies today.