The Growing Use of Chatbots in Customer Service


Thanks to substantial advances in machine learning and natural language processing, chatbots have become increasingly popular in the last few years. Since all businesses exists to serve customers, the technology used to serve said customers has gotten considerable attention and as a result has developed at a rapid pace.  We’ve come a long way since since Cleverbot’s strained responses in the earliest stages. The Turing Test may have been a misleading standard when it comes to creating the best performing AI software, but nevertheless you can see for yourself how human like chatbots have become, with Mitsuku winning the Loebner Prize for the fourth time in 2018.

What is a chatbot?

A chatbot is a software program that is able to conduct a conversation via auditory or textual channels. They’re intended to mimic how a human would carry the same conversation as convincingly as possible. Chatbots are used for numerous practical purposes such as customer service and data collection. Many make use of complex natural language processing systems but many less intricate ones scan for keywords and pull the most appropriate answer from a database. In 1994, Michael Mauldin (created the first Verbot) coined the term “chatterbot” to describe such conversational programs. Most chatbots are presently accessible via virtual assistants like Google Assistant or Amazon’s Alexa, via messaging apps like Facebook Messenger or WeChat or via apps and websites from individual organizations.

Chatbots can be grouped under usage categories as follows:

  • Chat e-commerce
  • Analytics
  • Customer support
  • Communication
  • Developer tools
  • Design
  • Entertainment
  • Education
  • Finance
  • Human resources
  • Marketing
  • Productivity
  • Shopping
  • Social
  • Travel and utilities.

We’ll be examining the reasons this type of software is becoming so prevalent and the impact it will have on how we communicate, conduct business and interact with each other online. Our world is culturally moving away from human interaction and is now developing a preference for self-service options when it comes to customer service. This applies especially to millennials, a recent study uncovering that 72% of them don’t believe that the best way to handle their customer service inquiries is via telephone call.  It can be predicted that the following generations will rely even more on self-service and quick responses which can be managed by incorporating a chatbot.

According to the experts at, chatbots increase CSAT (Customer Satisfaction Score), decrease customer churn and drive revenue, whilst freeing live agents to have a greater impact. Businesses are consequently beginning to gain value from using this technology as we see in the case of Coca-Cola, Sephora, Marriott or Hilton. In fact, Mariott now has three different chatbots and is working on acquiring more, 1-800 Flowers is working on developing bots for each major channel and Oracle did a survey showing that 80% of companies intend to use chatbot software by the year 2020.

McKinsey also conducted an analysis and demonstrated that businesses save up to 29% on customer service through automation and BI Intelligence estimates annual savings to $23 billion.

Why are chatbots becoming so ubiquitous in customer service?

Given that one member of the customer service staff can answer just one query at a time and cannot operate 24/7, chatbots can make an excellent investment.  The bulk of customer service calls can be split into the 80/20 rule, as 80% of the time the same questions are being asked. At present, these programs are being used to automate simple queries that are made repeatedly, with a live agent taking over if the bot doesn’t have enough information to answer a particular request.  When the bot is unable to provide an answer, it automatically passes it to a live agents and records the answer thus expanding its database for the future. The best aspect of using this kind of technology is that the program gets smarter with time and can handle more and more requests on its own.

Even when the live agent doesn’t know how to answer a particular question, the chatbot records the answer given by the internal help desk and includes it in its database, effectively decreasing the response time to the customer. Hiring and training customer support agents is very expensive, the hiring process for call center employee reaching upwards of $4,000 and the training programs $4,800.  Since this profession has a high turnover rate, these expenses make it very hard for companies to sustain a necessary operation.

Even though a chatbot has to be consistently monitored, it can carry an unlimited amount of requests at the same time, allowing the human agents to direct their attention exclusively to high-level demands, thus helping reduce the turnover rate. It’s a win-win. Chatbots adapt to changes in scenario without needing unaffordable training and never gets cranky, never comes late to work or becomes over-worked. By reducing the strain on your staff, you’d also be improving customer interaction which is shown to lead to repeat business.

According to a recent Zendesk study, 42%  of customers made more purchases after a positive customer service interaction, while a bad one led 52% of them to cease buying anything from that company. Not only this, but Globe Telecom managed to also increase employee productivity by 350%, all while reducing call volume by 50% and increasing customer satisfaction by 22%. Activision collaborated with Facebook in 2016, when launching Call of Duty: Infinite Warfare. They would direct players through a short video message to a Facebook Messenger based chatbot which was playing the role of Lt. Reyes, the main protagonist of the game. The bot had 6 million conversations within the first 24 hours.

All things considered…

It’s clear by now that using chatbot software is spreading like wildfire in every industry and it makes perfect sense. Even if the stage we’re at when it comes to AI development does not satisfy the standards to pass the Turing Test, we’re still get very convincing results. Current generations seem to prefer interacting with non-human agents and companies make huge savings from this trend. In the past AI was shrouded in mystery and seemed like the final frontier but now it’s here to stay and seems to be making itself comfortable in the Customer Service sector.

Similar Posts

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.