Chatbots have evolved on the back of AI, ML and speech recognition, integrating into consumer and business application at a dizzying speed. This note takes stock of the application areas, companies of interest, and changes ahead.
‘Hello. How can I help you today?’, a conversation once between two humans, increasingly finds a smart software conversing (text or voice) with a human, like a human. From Eliza in 1966, KITT in the 1982 Knight Rider Series to Jarvis working alongside Tony Stark in Iron man (2008), and AI-powered Alexa, Siri and Google Assistant of the 21st century, these have made digital assistants and chatbots both ubiquitous and acceptable. At the core, they remain a software code designed to recognized input data or voice and respond in predicated ways (that’s changing as well).
Since 2010, chatbots have evolved from being mere digital tools that performed generic, repeatable tasks, to digital assistants that use Artificial Intelligence (AI), Machine Learning (ML) and Natural Language Processing (NLP) to engage better and more efficiently with customers and employees alike. It is expected that, by 2020, chatbots may well understand sentiment (knowing customer’s mood from tone of voice and manner of speaking), recognise individuals through facial recognition, enable secure transactions through blockchain and find wider applications through internet of things (IoT).
Their adoption is being fuelled by interplay of tech evolution, need for superior customer service, pervasive desire to rein costs (potential at over $8 billion by 20221) and automate services to the extent possible. Businesses spend over $1 trillion on customer service calls and chatbots can take over 80% of included routine questions2. Equally, the rise of text messaging and live chat has changed how customer seek to interact with enterprises (replicating the P2P behaviour), relegating the mobile app approach. There are an estimated 100,000 bots on Facebook Messenger alone3. Gartner estimates that, by 2021, more than 50% of enterprise will spend more on bots and chatbots over traditional mobile app development4. And all of this is driving funding, estimated by some at $24 billion5 till date, to the chatbot landscape, already valued at over $1 billion.
Chatbots have found increasing usage across customer facing industries, primarily because of their ability to answer queries of multiple customers simultaneously, round the clock. AI-enabled chatbots are already providing personalised solutions and recommendations to customers, in the quest for higher customer satisfaction and stickiness.
In healthcare, Chatbots have helped streamline diagnosis of patients as well as disseminate information related to ailments. Melody, the chatbot in Baidu’s Doctor app, and India’s online pharmacy 1mg, provide two way conversational features to help patients zero down on their medical issue. Boston Children’s Hospital partnered Amazon Echo to create KidsMD that provides parents with basic health advice for their children during ailments and even shares information on age and weight specific dosing guidelines for over-the-counter drugs.
Bank of America was one of the chatbot pioneers in the finance, introducing Erica in 2016. Erica helps customers pay bills, provides updates on FICO score and identifies areas of savings based on their financial behaviour. Other players like MasterCard and American Express use bots on Facebook Messenger to facilitate payments and provides updates and benefits based on loyalty and spending patterns. And then, COIN at JPMorgan helped save 360,000 hours of manpower by streamlining its back-end operations (categorises employee emails, sifting through complex legal contracts and routine IT tasks).
The retail industry has been one of the biggest beneficiaries of Chatbots, primarily because of bots providing a more personalised and more efficient user experience. Popular clothing brand H&M, French cosmetic chain Sephora and women’s lingerie brand, Victoria’s Secret Pink, have been early-adopters of messaging application Kik’s fashion and beauty bot category. These bots provide users with outfit and design recommendations, allow them to chat with friends and enable them to browse new styles – thus providing a more social, customised and enhanced shopping experience.
There is an almost unanimous consensus on the tremendous role that chatbots are poised to play in the foreseeable future. And yet, the biggest challenge comes from the status quo – customer’s preference for a human interaction. There are also concerns around data breach and security of information shared with chatbots. And the question of chatbot failure, brought to fore after bots at Facebook failed to fulfil user requests. While initial chatbot outcomes were sketchy, the current wave of deployments and move towards AI-powered agents and a personalized service together with rising automation does hold one promise – chatbots are here to stay.
1. Top Strategic Predictions for 2018 and beyond (Gartner, Oct. 2017)
2. Chatbot Conversations to deliver $8 billion in Cost savings by 2022 (Juniper Research, July 2017)
3. How chatbots can help reduce customer service costs by 30% (IBM, Oct. 2017)
4. Conversational Artificial Intelligence. We Need To Talk About It (Gartner, Oct. 2017)
5. Bots shift towards AI and garner $24 billion of investment (VentureBeat, Mar. 2017)