The evolution of AI: Are chatbots driving a new customer experience?

two vintage robots with old phone on wooden floor

We’ve all had our frustrations dealing with a telco, bank, utility, insurance company or other large organisation when trying to get information, complete a transaction, or request service. Typically, we start our search on an organisation’s website, but we might end up dealing with a long winded IVR system with long wait times, or eventually talking to a customer representative who may not be geographically co-located and thus only partially understands local nuances.

More customers are now engaging with service providers through digital devices such as smartphones, tablets and laptops, and I’ve often observed that they can switch devices almost three times per hour. The ‘connected’ generation of Millennials and the iGeneration prefer to interact with organisations on mobile applications and ever evolving social media channels. Customer engagement is increasingly through text rather than voice, and the option for self-service is becoming the favoured approach assuming it is easy to use, it works, and we get the answer to our question.

Only 36% of organisations have some visibility into multichannel data…

But why is it so hard? Why do so few companies have a viable multi-channel, customer engagement platform that allows context, intent and integration across channels providing a seamless experience? There is nothing more annoying than being passed from one division to another and hearing the words “we are not able to access that information, please restate your problem.” It shouldn’t matter where we start or where we need to go; we should not have to start all over again.

Call centre engagement is no longer the preferred approach to service customers. The cost of engaging through the call centre is the most expensive, time consuming, and least effective method for delivering customer satisfaction. Indeed, the cost per call for a chatbot is usually under one dollar whereas a CSR typically costs between $10 – $25 per call.

There are many organisations that have now engaged chat facilities using chat agents, but again they typically fail to deliver the ideal customer engagement experience. Although, a chat agent can typically handle 3 -5 client interactions simultaneously; they can be cumbersome, there are time delays, and are only viable for the time the chat window is open.

Customer interaction and the goal of providing viable synchronous self-service capability is now a 24-hour requirement. However, having a voice or chat agent available 24 /7 is not commercially practical, so moving to a virtual agent or chatbot (a computer based system using AI to provide 24/7 customer experience) is quickly becoming a preferred option.

Refining the customer experience through less intervention

Chatbots are essentially digital personal assistants programmed to simulate human conversations via voice or text providing a “virtual” communication partner. They can start off in a simple version whereby they are dealing with either voice or text through natural language interfaces enabling the computer to understand the question and provide the best answer in real time. Chatbots have the advantage of being multi-channel, with context and integration across systems; and Gartner has even predicted over the next few years 85% of customer engagements will be fielded without human intervention.

Another advantage of chatbots is that they learn and get better over time, which is why it is essential to train them before using them operationally. It is also important to manage customers’ expectations, particularly at the beginning, while the chatbots are still developing and expanding their learnings – just like humans. However, once fully trained, chatbots can become the standard means to engage with many customers, for example:

  • Banking and finance – provide better self-service functions around personal banking, credit cards and home loans applications
  • Insurance – choose the right coverage, manage claims, provide product information
  • Health – patient care, provide appropriate remedies for specific disorders
  • Retail – help customers search for products and place their orders
  • Utilities – add new services and products, report issues, faults and problem resolution
  • Telco – get to the right division within the organisation, add new services, faults and problem resolution

Chatbots are capable of anticipating and acting on customer intent, improving time of resolution, and rarely need to engage with a human for problem resolution and self-service functions. The more a chatbot is deployed, the more it effectively learns from one conversation to the next as it takes note of shared dialogue and content. The result: chatbots naturally evolve over time especially when they begin to access the vast amount of data from the systems they have available and the link between AI and Big Data only grows stronger.

Where is this all heading?

Almost 80% of companies are considering chatbots for their customer care strategy by 2020. 62% will implement chatbots by the end of 2018.

There is even speculation that chatbots will replace search engines. What this illustrates is the simple and natural way chatbots access information from multiple sources, their potential power, and their extraordinary ability to discover and learn. .

Many articles are also being written about why, and when, chatbots will replace apps. This is largely due to the fact that chatbots mimic century-old human communication, in a natural manner, rather than forcing us humans to interact in a prescribed way. This is consistent with the trend of the declining use of dedicated mobile apps, and greater uptake of social media interfaces. The obvious example is Facebook Messenger which is especially friendly to chatbots, and many chatbots are instinctively gravitating towards this channel. It’s on this premise we will likely start to see chatbots become more embedded in social media and begin to replace dedicated mobile apps.

Why are Chatbots so hot?

24 x 7 access to customer engagement is fast becoming a mandatory requirement for today’s organisations. Chatbots fulfil this requirement by offering 24 x 7 x 365 coverage, far cheaper, and more effectively, than human agents especially in peak periods. Australian airline, JetStar,  and their virtual assistant “Jess” is a great recent example of the effectiveness of chatbots. When Bali’s Mt Agung erupted at the end 2017, Jess was able to assist 3,000 customer with flights, changes and cancellations – all of which could not have been so efficiently handled by a human call centre.

There are many vendors offering chatbot/virtual agent solutions, some are very easy to use and supposedly quick to develop. They are aimed at the simple use case where the volume of queries might be high but the number of questions is limited and well known, with expected answers and short conversations.

However, users are quickly going to become used to interacting with chatbots and will soon become more demanding of their capabilities (such as the Jetstar example above). As such, organisations should consider carefully their solution of choice to avoid locking themselves with a solution that cannot scale. There are greater platform offerings which can do the simple and fast components, but they can also evolve into more complex scenarios requiring lots of integration and self-learning with complex questions and longer conversations.

Happy AI-ing !

This article is the second of a series about AI from Focus the Way Forward. Watch out for our next fortnightly installment!

Michael Ritterman

Michael Ritterman has career spanning over 30 years combining academic and IT pursuits. He has held senior sales and management positions in a broad range of technologies working for both Australian and international SI and Software vendors. Over the last 10 years he has provide thought leadership to many organisations with their go to market strategies and sales enablement. Currently he is the Sales Director for DXC in the IBM Practice within the Emerging Technologies Group.


  1. […] article is the first of a series about AI from Focus the Way Forward. Watch out for our next fortnightly […]

  2. […] enterprises operate by improving business processes, supply-chain management, financial modeling, customer service, workplace safety, and digital […]

  3. […] The evolution of AI: Are chatbots driving a new customer experience? […]

  4. […] rapid increase in the number of chatbots that pop up on many web sites with the goal of helping to increase customer satisfaction with that digital experience. This is also increasingly the case when communicating with voice enabled devices like […]

  5. […] with productive dialogues and meaningful information exchange between the robot and the user. These good bot experiences mean the customer gets the results they want and the company involved has a happy and loyal […]

  6. […] The evolution of AI: Are chatbots driving a new customer experience? […]

  7. […] dialoghi produttivi e scambio di informazione significativa tra il robot e l’utente. Questo tipo di esperienze significa che il cliente ottiene i risultati desiderati e che l’azienda coinvolta ha un cliente […]

  8. […] et à un échange significatif d’information entre le robot et l’utilisateur. Ces bonnes expériences de bots mettent en évidence qu’un client obtenant des résultats attendus est un client heureux et […]

Speak Your Mind


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