Chatbots: A customer experience relic
“How may I assist you?”—It’s hard to miss the popup icon at the bottom of your screen every time you land on a website. Strongly reminiscent of Clippy, the ever-helpful assistant on MS Office, website chatbots are the latest players helping to enhance customer engagement and lead generation. These conversational agents are changing the experience for customers and decision-makers alike through accelerated and accurate communication.
The scope of customer support and service desks has broadened with the advent of conversational Artificial Intelligence (AI) and chatbots. By streamlining and automating the user support process, chatbot solutions engage users in unsupervised simulated conversations and give them a personalized experience. However, contextual awareness and user expectations are critical details that can’t be overlooked.
What are chatbots?
At its core— “A chatbot is a software or computer program that simulates human conversation and interacts with users on various platforms eliminating the need for human intervention.”
However, they are far more capable than their conversing skills. When used well, these conversational agents also double as powerful business tools. The automation of basic processes enables businesses to direct their valuable resources to more critical tasks.
Chatbots are deployed on messengers, standalone applications, and websites. Based on their capabilities, they are classified into the following types of chatbots:
- Keyword recognition-based
- Machine learning
- Voice bots
How do chatbots work?
The conversational user interface is linked to various data sources via an API. The required data is retrieved in real-time from the data storage and knowledge bases. Once the data is processed through the Natural Language Processing (NLP) algorithms, the chatbot starts responding to customer queries.
In this way, chatbots are able to deliver information and services like breaking news, online purchases, weather forecasts, product information, payment management, general information, lead generation, personal assistance, and a whole range of other topics – all on demand. 1
Read about how brands are using chatbots to connect with their audiences.
The evolution of chatbots
To understand the evolution of chatbots, we must start in the 1950s when a landmark paper by Alan Turing, Computing Machinery and Intelligence, introduced a thought experiment that aimed to assess the sophistication of computers. This method, eventually, served as a baseline for future computers and machines.
The Turing Test is a deceptively simple method of determining whether a machine can demonstrate human intelligence: If a machine can engage in a conversation with a human without being detected as a machine, it has demonstrated human intelligence. The test has inspired many researchers and scientists to work on chatbots that utilized NLP and AI to simulate human conversations.
In its early stages, chatbot development employed core NLP methods; machine learning wasn’t exactly reliable back then.
Chatbots with NLP implementation
Some early attempts at implementing NLP in chatbots were seen in 1964 when Joseph Weizenbaum created, one of the first chatbots. Within two years, ELIZA appeared to have conversations with humans by mimicking and substituting their words giving the illusion of a real conversation.
Almost a decade later, Kenneth Colby worked on ELIZA’s underlying principles and created an advanced version (PARRY) that used a more conversational strategy. In 1973, a conversation between ELIZA and PARRY went something like this —
ELIZA: Good Evening. Tell me your problems. Please terminate input with a period or a question mark.
PARRY: People get on my nerves sometimes.
ELIZA: I am not sure I understand you fully.
PARRY: You should pay more attention.
ELIZA: Suppose you should pay more attention.
PARRY: You’re entitled to your own opinion.
ELIZA: What makes you think I am entitled to my own opinion?
Siri and Alexa have definitely come a long way from 1964’s ELIZA!
Chatbots with AI implementation
The first wave of AI chatbots came during the 80s and the technology advanced from there. Jabberwacky by Rollo Carpenter is one of the earliest creations in AI that used human interactions. It aimed to simulate conversations with entertainment and humor to make them more interesting. Jabberwacky was introduced in 1988.
In 1995, ALICE (Artificial Linguistic Internet Computer Entity) was introduced. While NLP was used in the original version, ALICE eventually migrated to an AI language after undergoing numerous changes.
In the early 2000s, chatbots were struggling to deliver a reliable experience. Implementation was time-consuming and the sophistication was far from what was envisioned as some of the remarks by the bots were inappropriate.
But, in 2010 when Apple’s Siri was launched, the chatbot game changed completely. Since then, the world has witnessed a new generation of intelligent conversational AI like Cortana, Alexa, Google Assistant, Netomi, BlenderBot, and various others.
The latest trends in chatbots
Chatbot technology has progressed beyond scripted resolution paths. Functionality over gimmicky attributes is trending, and contact center AI and chatbots have a brighter future ahead thanks to:
- Context Management: Bots rely on past user interactions to remember important details like employee profiles, client profiles, customer information, customer preferences, etc. This contextual awareness helps deliver a personalized experience.
- Sentiment Analysis: Chatbots today can detect the tone and emotion of a user based on their communication. They are then able to respond accordingly by either changing the direction of the conversation or the conversing style or by immediately bringing in a human agent who can carry the conversation forward. A popular example of sentiment analysis is the Grammarly tone detector that helps writers ensure that the right tone and message are conveyed through their written words. Sentiment analysis is also one of the key features of HGS Agent X that helps identify the feelings of the customers accurately.
- Dialog Management: To keep up with the complex dialog changes of human conversations, conversational AI technology enables bots to handle situations that involve entity changes, processing multiple entities in parallel, mixed dialog, etc.
- Omnichannel Support: Omnichannel support enables users to start a conversation on one channel (WhatsApp) and end it on another channel (YouTube). Context or continuity is not lost in the process and makes for very streamlined communication.
- Multilingual Support: With multilingual support, it is easier for organizations to cater to a global audience.
Read our whitepaper on AI-powered speech analytics to find out how it can optimize the customer experience.
A glimpse into the future of chatbots
AI-based systems deliver a more human-like experience with integration capabilities with other solutions like Robotic Process Automation (RPA), recommendation engines, answering questions, appointment scheduler, reservations and bookings, etc.—the possibilities are endless. It depends on how well understanding, memory, sentiment, personality, persistence, and tangents are capitalized. Each capability enhances the AI-based interaction.
Learning from each interaction and preserving that information is now the most basic function expected in chatbots. Needless to say, chatbot development and training is a skill in itself.
At HGS, our CX bots are equipped to address 85%-90% of the issues that can be resolved via automation while our agents handle all remaining problems. The agent-assist bots can optimize team performance with quick and consolidated access to customer information and internal knowledgebase systems.
Chatbots are evolving toward building an intelligent user-engagement hub. Predicting user demands has become more accurate for these AI-infused virtual assistants, which in turn, has helped them deliver very tailored and personalized suggestions and responses.
Consult our experts and get started with conversational AI and CX bots.
Prakash Hariharasubramanian, Director & Practice Lead, Intelligent Process Automation (IPA), HGS
Prakash has led various IPA implementations across multiple industry verticals in his tenure of 7 years with HGS. In his role, Prakash develops IPA practice frameworks, creates IPA solutions, and serves as a key automation evangelist for HGS.
Recent blog posts: