Can Chatbots Help You Win the Customer Support Challenge

With Amazon Alexa Skills, Facebook Messenger and RCS, companies can finally leverage chatbots and voicebots to conquer the customer support challenge.  Why? Because, these innovations have made today’s economy more “on-demand” than ever. From entertainment to food delivery, consumers are conditioned to believe they can get anything, anytime, anywhere at the push of a button on their mobile phones. The impact on customer service has been enormous, as people want the information they need or their issues addressed in an incredibly short span of time.

The result for customer service, support, and care is that consumer expectations are higher than ever. If people don’t get the answers they need right away, they’re likely to switch over to a competitor. In fact, Microsoft’s 2016 Global State of Customer Service Report indicates that 93% of consumers think customer service is a key factor in determining their loyalty to a brand.

Digital channels will be the future starting point for most customer care interactions. Over 50% of customers now believe they should be able to resolve their issues on their own, without relying on a live customer support agent. This according a study conducted by Zendesk.  In another survey, Harris Poll found that 64% of customers would rather text than call a business. The increasing reliance on self-service reflects rapid growth in digital-savvy users. These people have grown up communicating, and accessing data, over the internet. They see self-service as the most time, and cost, efficient method of solving problems.

To provide rapid (yet still personalized) service, companies are finding one technology particularly revolutionary: Chatbots. In layman’s terms, a chatbot is a computer program that simulates human conversation through artificial intelligence and guided conversation. If someone sends a customer service inquiry via SMS message, for instance, a chatbot can interpret the message and come up with an immediate response that is likely to address the issue. Some even predict that by 2020, 85% of all customer service interactions will be managed by non-human technologies like chatbots.

Today, we’re at an inflection point for chatbot technology. Adoption, innovation, and customer acceptance of intelligent agents is increasing. Both businesses, and customers, are recognizing the user cases and value of in terms of customer service and support. In this white paper, we’ll dive into why now is the time for chatbots, how companies can phase them into their support processes, and areas to achieve quick (and long term) ROI wins by utilizing chatbots for customer care.

Customer Service Chatbots – Why Now?

For prior generations, a phone call was the likeliest starting point for customer service interactions. Today, most customer service communications begin on digital channels. Per Microsoft, 55% of consumers initiate customer service interactions via digital channels such as email, SMS message or social media.

This allows the modern customer service environment ripe for chatbot technology. Brands that can address those digital interactions quickly and effectively will generate greater trust and loyalty among their customers. Acceptance among consumers is growing, with over 50 percent of people now trusting chatbots for basic customer service tasks like bookings and purchases.

As a chatbot developer and a builder of Alexa Skills we have watched as the infrastructure for chatbots and voicebots has matured and fused into channels that are equally familiar to most people. Facebook Messenger, the messaging extension for Facebook users, has emerged as a clear nexus point for handling customer inquiries. Today many companies use Messenger as the gateway for digital customer support with both humans and bots as agents. Over time as these interactions become more understood, these digital channels will convert to largely chatbot-first engagements.

The natural language processing (NLP) technology on which chatbots rely on is getting cheaper and more easily accessible. For instance, IBM’s Watson Developer Cloud. is a collection of ready-to-use chatbot artificial intelligence (AI) and NLP technology. It’s purchased as a subscription, and can be implemented in most any customer service or call center.  Google and Facebook among others offer terrific “AI-as-a-Service” kits for developers as well. Microsoft’s Bot Framework and Microsoft LUIS are other great examples. Bot Framework is a web service that lets companies design, build and customize service conversations based on software already developed by Microsoft. This “drag and drop” approach allows developers to build out sophisticated bot interactions.

Bot Framework also offers a FAQ-like bot service. The goal is to build a chatbot to answer FAQs that a business may try to answer for users of their websites in. This service is in a developmental phase. Yet it represents another digital channel use case for bots as intelligent agents. If you are thinking of this approach as a first step, we recommend a blended approach.

You should rely on both Bot Framework capabilities, as well as a thoughtful approach to the design of the conversational interface.

A Customer Support Natural for Users

For today’s consumer, chatbot support is a logical use case. Customers are now migrating towards self-service, preferring not to deal with human agents if they don’t have to. In fact, 90% of consumers say they now expect brands and organizations to have a self-service offering. Because customers either want to solve their issue quickly on their own, or receive an immediate response should they contact the brand, chatbots can help companies meet these expectations.

Chatbots also make sense for users because chatbots occupy the digital spaces and channels that consumers spend most of their time. Recent data indicates that 40% of consumers would rather use a messaging app for customer service than make a phone call. Add this to the prevalence of email, social, and self-service, and you have a near perfect use case for consumers when it comes to chatbots.

Most interactions in the future will flow through digital channels first. A personalized experience is something users will come to appreciate. Chatbots will be able to conduct conversations and transactions based on customer preferences. A one-on-one dialog reflecting user history with the business will benefit the customer.

A Customer Support Natural for Businesses

For companies, chatbots represent an opportunity to be the voice of their brand 24 hours a day, seven days a week. Even though brands might not have agents on call to respond around the clock, automated chatbot conversations can still provide brand ambassadorship. Chatbots also can serve to automate “low value” interactions such as simple informational inquiries such balance checks or store hours or product availability. The reduced need to service these lower value calls and interactions creates more time and resources to focus on adequately addressing higher value interactions. In fact, per McKinsey Consulting, surveyed executives said that their top customer service priority over the next five years is to reduce call demand and increase self-service.

Offloading some of these low value interactions also results in huge return on investment (ROI) for service departments. Per Forrester, on average the cost of a chatbot interaction is about 25 cents, while hiring a live agent to handle that same interaction would cost anywhere from $6-20 dollars. This is becoming more and more crucial, as 60 percent of leading contact centers now use “cost reduction” as one of their main key performance indicators (KPIs). The ability to quickly handle simple inquires at any time of day or night, along with the reduced overhead per interaction, is making chatbots a valuable enterprise asset for brands across all industries.

Finally, enterprises should start with a proof of concept when considering chatbots as a change agent for their customer care operations. First, they should clearly map out the value proposition that chatbots would bring to both internal and external stakeholders. Second, focus on specific use cases, products and scenarios where chatbots can provide value quickly and immediately.

Make sure the chatbot is integrated to all the relevant data and workflows that it needs to be, and then make sure to utilize data generated by bot interactions to improve and adapt usage on an ongoing basis. By taking these steps, and approaching chatbot technology with a mindset of constant improvement, companies can serve their customers better and impact their bottom line at the same time

Chatbot Customer Support Success Stories

Chatbots aren’t just a “wish list” technology for brands and service teams. Chatbots are positively impacting both the customer experience and the bottom line for a great many companies.

Here are a just a few:


    • One of the leading global providers of health insurance
    • Challenge:
      • Sought automated way to help members complete a secure, HIPAA-compliant (and therefore complex, difficult) account registration process
    • Action:
      • Developed the “Ask Ann” chatbot to assist members in the process
    • Results:
      • More than half of those registering on member Web site engage with Ann
      • Ann handles an average of 20,000+ chats per day
      • 29% reduction in calls to member service helpdesk
    • The largest passenger railroad service in the United States
    • Challenge:
      • Needed to find a way to both simplify online booking process and reduce customer support overhead
    • Action:
      • Created the “Ask Julie” intelligent agent to provide both written and vocalized responses
    • Results:
      • Went live in six months and now answers more than 5 million inquiries per year
      • Amtrak experienced an 8X ROI multiplier due to both cost savings and increased bookings
      • Currently saving $1 million per year in customer service costs
  • Alaska Airlines
    • One of the largest regional airlines in America, carrying over 17 million passengers per year
    • Challenge:
      • Alaska’s knowledge base of customer service solutions had grown in terms of both amount of information and complexity, creating customer care productivity bottlenecks.
    • Action
      • Launched an intelligent assistant chatbot named Jenn at and Super Jenn within Alaska’s internal systems.
    • Results
      • Jenn increased engagement, and decreased call-center, traffic for Alaska’s customer service operations.
      • The need for an online “live chat” function was eliminated within the first year of Jenn’s implementation.
      • Jenn is now the most preferred option by customers on the Contact Us page

Phases for Building Out a Customer Service Bot

For companies, call centers and service teams looking to build their own customer service bot, a phased approach is recommended. Here are the steps you should take to implement a chat bot, start to finish:

  • Identify a high volume  use case, along with a channel. This could be SMS messaging, web chat, email or social. They key is to pick one channel to focus on in the beginning. Review customer service data, call and chat transcripts for keywords to utilize.
  • Design the bot’s conversation flow. Identify systems integration requirements needed to enable the bot. Take time to sketch out specs, including flow designs for all necessary areas.
  • Plan for onboarding users to the chatbot and how each interaction will trigger, complete or escalate. For instance, understand the logical design elements available for various messaging channels such as the web,Facebook messenger or SMS. When a live agent becomes necessary, have a plan to loop them into the conversation.
  • Test the bot on an ongoing basis, and track its performance over time. Before it goes live, make sure to conduct a user acceptance tests (UAT) in a demo environment to correct obvious flaws. This will give you an idea of how to optimize performance and conversation flow. Always instrument the bot with a bot-specific analytic packages to understand interactions. We recommend implementing more than one vendor’s solution to take advantage of the different features each includes in their non-premium offerings.

Quick Wins for Businesses

By implementing chatbots, brands can create “quick wins” in multiple areas of service for their customers. Here are some of the key areas to focus on:

  • FAQs. There are many businesses focused on helping companies create FAQ bots that read your website and relevant customer interactions to produce conversational bots. The FAQ bot is a first step in creating a bot to handle web or social media based customer inquiries. It works when a customer asks a question which can be answered through an automated data lookup of available of information provided to the bot. More complex questions can be routed to live agents. 
  • Check account balance. Today, 22 percent of consumers say they use chatbots to interact with their financial services provider for tasks like checking their balance or activating a new account.
  • Status updates. Whether it be a delayed flight or a change in hotel booking, brands can notify customers of status/news updates with push notifications from chatbots. The intelligent agent can then ask if the customer wants more information or would like to take an action based on the new information. 
  • Bill payment. Today, customers see many lines and charges in their bill payment interface. This all depends on the service or vendor. Yet the introduction of mobile payment platforms is making the payment process easier. Chatbots integrated with mobile payments enhance customers’ ability to transact from anywhere.  
  • Mobile Texting. Texting is replacing the phone call as the preferred first service channel. Implementing basic rules-based, or advanced machine learning, chatbots to respond to mobile queries can solve many initial queries. 
  • Social Service. With customers turning to social media as a convenient means of customer service, chatbots integrated with apps like Facebook Messenger can be highly valuable.

It’s clear that chatbot technology is set to play a major role in the future of customer service, support, and care. The use cases for both users and enterprises is becoming more compelling than ever, as customers’ demands are rising along with companies’ desires to contain costs. Moving forward, companies will need to build a business case for chatbot use within their organizations, develop a phased implementation roadmap, and start by utilizing chatbots in areas where they can see an immediate impact on their customers’ experience.

In our experience, developing a roadmap to improve customer care and support with chatbot technology is the first step towards success. At Azumo, our focus on delivering world-class chatbot solutions is the result of putting many of the concepts mentioned in this paper into practice for clients like Twitter and others. If you’re interested in discussing how to take your customer service and support operations to the next level with chatbot technology, please reach out to one of our experts.

Share On:

You Might Also Like

Case Study: Twitter


Video: Panel Discussion on Speech, NLP and Chatbots

We’ve been following the chatbot opportunity closely for some time now and re...

Transcript: Panel Discussion on NLP and Chatbots

Below is the transcript from our panel discussion on NLP. smart group of pane...

Leave a Reply

Work with the Leading Software Development
Services Company for the Modern Age