Customer needs and desires do not always overlap with business objectives: for example, cutting costs as a corporate goal does not benefit the consumer directly.
As for customer service in the past, cutting costs often meant the introduction of automation in the form of Interactive Voice Response (IVR). While these systems promised 24/7 access to information and no waiting on hold, many poorly designed systems did nothing to improve the better service hopes of consumers. Whether it was confusing automated menu trees, clunky touch-tone interfaces that slowed interactions or poor speech recognition, one thing became quickly clear: that IVR has never been a friend of the consumer’s.
The Growth of Text Messages
But taking the conversation from the spoken to the written word has the potential to help both customers and businesses. The advantages for the consumer translate directly into advantages for the business.
For the first time, a service channel holds the promise of making service so easy that it is not only affordable for the business, but also the consumer’s preferred means to get help.
“Phone call first” was the predominant theme of the 80s and 90s and “web first” ruled in the early 2000s. But we are now entering the “messaging first” era, a variation of “mobile first.”
Connecting Questions to Answers
When it comes down to it, customer service is about connecting questions to answers. Everything starts with a question: “What is the status of my claim?,” “Where is my order?,” “Can I get a refund?” or “Where is your nearest branch?” Yet with websites, mobile apps or IVR systems, we never let the consumer just ask the question. Instead we force them to transform the question into a sequence of menu item selections.
Categorize your question according to our predefined options, then navigate your way through our content.
And that’s still the paradigm today. If you can’t find the answer, call our hotline and start from scratch with our IVR. The opposite would be good service: tell us your question to begin with, and we will figure out the answer for you. Web chat is a first step in the right direction, but it is still a clunky experience: visit our website, optimally on a big screen, start a chat, then wait for someone to connect. There it is again: the “wait on hold” that we all fear. And if you close the window, you can start anew.
Messaging: The Shortest Path to an Answer
A solution based on messaging takes the shortest path from the question to the answer and combines the advantages of communication channels that are asynchronous (like email) and synchronous/realtime (like phone calls).
Type the name of the business into something like Facebook Messenger (if it isn’t in your recent list of conversations yet), find it, send your question. What comes next is a clever mix of artificial intelligence and human support: if the question is simple enough to be handled through automated data lookup, the answer comes instantly. Optionally, you will lead a short dialog with a so-called chatbot to bring your question to conclusion (e.g., it would ask you for your order or claim number to check the status.)
If the algorithm determines the question is too complicated to be handled by a chatbot, you may be quietly transferred to an agent behind the scenes. No need to wait or accidentally “close the window” as is the case with web chat. You can always return to the messaging application when you get a response from the business, or when you have time to advance the conversation.
Chatbots with human backup will be the leading theme for customer service for the next few years to come.
Using APIs for Self-Service
As was the case with websites, mobile apps or IVR systems, the prerequisite for this type of self-service is that your business data and knowledge is accessible for a computer system.
The technical term for such access is API — Application Programming Interface. Only with standardized APIs can efficient systems be written that expose data and information without the need for a human in the contact center to step in.
Most enterprises today are either on the path or have already concluded laying the foundation with a robust web service layer. Documented APIs are being shared with vendors that help the enterprise build a mobile presence, a sophisticated IVR application or, now, a chatbot for messaging experiences.
But is this the end game? Are we in customer service heaven when we have exposed all our data through an internal API that power our own customer service front-ends like a website or a chatbot?
The Rise of Personal Assistants
Many technology companies and visionaries have a bolder vision for the end game: a virtual personal assistant.
A virtual personal assistant can communicate with businesses on your behalf, already knowing your preferences and desires, understanding your behavior and, ideally, predicting your needs. While such a system is probably still 5 to 10 years away, we need to plant the seeds now, as it will inevitably come, and no business wants to be left behind when it becomes mainstream.
Beyond robust natural language understanding capabilities and artificial intelligence that can learn our behavior over time and adapt to us, such systems need better access to data. While it was OK to limit enterprise APIs to the vendors that needed them to build websites, mobile apps, IVR systems or now chatbots, we next need standardized access for anybody and anything.
Customer Service Heaven
To understand why, let’s consider a typical scenario.
Say you lose a credit card you use for various services, such as paying your gas and electric bill, your car insurance and your TV streaming provider. Approaching each vendor individually is a pain. They all have their own ways to change the payment method. Some allow you to change with a phone call or do it online. Others expose it through the mobile app, and the exact steps you need to take are always proprietary.
To navigate the options requires human intelligence. There is no standardized API to “changing a payment method.”
While a chatbot lets you simply send “I lost my MasterCard and need to change my payment method” to the business, which is the least effort of all options available today, the personal assistant would do the job of approaching each business for you. You would tell it “I lost my MasterCard, can you inform all vendors that use it to switch to my Visa card instead” and it would report “done” a few seconds later. If a business needs clarification, you would message with it via your personal assistant, i.e. use a chatbot interface but use your assistant as a relay.
Customer service heaven, indeed.
A Bold Vision
That is the bold vision. Personal assistants will be the leading theme for 2020 and beyond. Chatbots are a milestone on the path towards that vision.
Back in the 90s, it took businesses small and large at least 10 years to understand why they, too, would want a homepage on the web. It will now take years to understand why they need a chatbot.
Once we have widespread coverage through chatbots, we are a step further towards a true personal assistant. The APIs to business data will eventually need to be made public and follow some kind of standard. The same way as we standardized the task of “changing a payment method” in our example above, we would need to digitize our entire lives.
Each individual facet of our lives needs to be meticulously designed as digital workflows.
It will take a massive effort to accomplish that and agree on a common standard, but the rewards will be such that we won’t be able to imagine life before we had a personal assistant — just like we cannot imagine a life without our smart phones anymore. It’s a journey worth embarking on.
Latest posts by Tobias Goebel (see all)
- New Ways to Customer Acquisition and Retention in the Contact Centre - March 19, 2018
- How We Built the Wine Bot Margot for Lidl - February 13, 2018
- The State of the Bot Going Into 2018 - December 27, 2017