Discussion about bots is everywhere. Some describe bots as the next best customer care solution, while some spread fear about displaced workers and robotic conversations. But customer service bots are already in wide use today, and they simplify interactions for millions of consumers every day.
Bots have been used by our favorite brands for years, and they come in many different forms. This post aims to demystify what bots are. It will also show how they’ve been used in the past, how consumers use them, how businesses use them and why customer service bots aren’t as intimidating as many people think.
This article is part of the Inside CX series from Intrado, through which we’re creating a detailed, focused and actionable library of CX content. Use this player to hear another version of this content designed for podcasts and subscribe to future episodes.
Difference Between In-Home and At-Work Bots
First off, millions of people have already adopted bots into their homes. The most obvious examples are smart speakers or home assistants, like Amazon’s Alexa, Google Home or Apple’s HomePod.
Developers program these at-home bots with skills to accept any input. Anything you can think of, the bots respond in some way. Whether you ask, “Hey, what’s the weather like,” or, “What’s the next movie at my preferred theater?” the bots empower users to use natural language to get information.
Bots for a business, on the other hand, are designed for more specific uses. Developers build customer service bots to serve a defined intent or complete a common action that end users want to achieve. Customer service bots are not built to respond to any input. There’s no reason to build it in that fashion. The bot simply completes the anticipated interaction with that business. These bots can exist in any form, including webchat, voice and SMS.
Decades of Bot Building
Customer service bots have been around for longer than people think. Although chatbots in a tool like Facebook Messenger often come to mind first, Intrado has been building voice bots for more than 30 years. Originally known as IVR, we refer to these bots as intelligent virtual assistants (IVA) today.
But over decades of development, the technology has changed. Previously, we defined static call flows and workflows as interacting with a bot. Though simple, these machines anticipated basic requests and provided some self-service to the end user. However, the primary intent of these older customer service bots was to discern why someone was calling and, often, route them to an agent to resolve it.
Over the years, we’ve developed more unique ways of interacting with customers. While customer service bots don’t need to answer every question like an in-home assistant, new technology allows for the same type of interaction in a customer service bot that people expect from their virtual assistant at home.
So if you call a 1-800 number and reach a customer service bot, you can now start a conversation by saying, “Hey, I’m calling because I need to make a late payment on my utility bill, and I also want to check my balance to make sure I don’t get a late fee.” Modern IVAs understand this natural language. They take all that information, understand what the customer is trying to do and immediately begin the task. No more call trees just to figure out what the user wants.
Beyond Voice: Omni-channel Customer Service Bots
Going beyond voice, the same technology is available for texting and chat channels, too. While voice is still a prominent channel for both in-home and customer service bots, omni-channel options have progressed a lot over the years. That’s because sharing data has become simpler than ever. No matter where the interaction starts, contact centers can see the customer’s entire experience within individual channels and across channels. That allows agents to respond more appropriately, and businesses can improve channels that make customers unhappy.
For example, if a customer can’t check out from their shopping cart on the web, they text the contact center to get support. That failed task online is shared with the SMS chatbot. With that context, the bot can predict the quickest path to resolution. Omni-channel isn’t about creating multiple customer service bots in multiple communication channels. It’s about making those channels work together to create the least impactful customer experience.
And some interactions are easier outside the voice channel. Lately, the market has introduced a new concept called visual IVR, which supplements a voice interaction with visual elements on the user’s smartphone. The bot supplies links, maps, buttons, forms and more to make the interaction easier to understand. At Intrado, we call this Visual Voice. Click here to learn more about our Visual Voice solution. This technology allows a bot to handle many requests that had been going to agents.
Bots Benefit Both Customers and Employees
Bots that accept natural language or work in multiple channels clearly make things easier for consumers. But customer service bots also make things more efficient for the employee.
There have been many conversations about how AI and bots will impact employees in a variety of industries. Some talk about bots doing back-office work to make businesses operate more efficiently. Yet others worry that bots could replace some human employees altogether. If that’s the case, then is increased efficiency worth it?
Remember, customer service bots today are designed to handle very specific requests. They aren’t be-all-end-all customer service solutions, so there will always be a need for a live person to enter some interactions. The goal of customer service bots is not to replace an agent. The goal is to take simple, low-value tasks off the hands of a contact center representative because those low-level tasks aren’t providing any value to the employee. They’re essentially white noise.
For example, when someone calls to check their bank balance because the mobile app or website is down, placing that call into a representative’s hands is a non-value add. You take what could have been a 10-second call and tie up an agent for two minutes because of back-office work or end-of-workflow processes they must follow.
But if customer service bots handle basic tasks, live representatives can handle difficult calls that truly support your customers. That’s the intent of the employee’s position, so employees feel more valuable and customers get quicker service.
Building a Bot: Where to Begin
Whether you work with Intrado or another company on building your first customer service bot, here are a few guiding principles to keep in mind as you get started:
1. Identify the Kind of Experience You Want to Provide
You could still set up a phone tree and ask customers to press “1” for “yes” or press “2” for “no,” but that probably won’t provide a very good customer experience. And a poor experience negatively impacts your business and impacts how your agents interact with customers. Think about how to make customers’ experiences as effortless as possible.
2. Decide How Far You Want to Go with Bot Technology
You can build one bot to complete a common request. Or you could build a network of customer service bots to automate as much as possible. You may find an acceptable DIY option to build and run a single bot yourself, but it may help to work with a larger cloud-based provider and learn from their lessons when building a more complex bot ecosystem.
3. Don’t Try to Boil the Ocean
Start with your easiest and most common customer service tasks. You probably have some idea of where to begin, but beyond that, create a process for identifying more automation options in the future. Find tasks that customers naturally tend to use, and add new features slowly in a natural progression. Start small, continue to iterate and always keep in mind the customer experience.
4. Don’t Create Barriers
This is probably the number-one rule when creating a bot solution. Customers will find a way to get what they want if you reduce the friction between them and the end experience. If the bot experience is easy, customers will adopt the solution faster and more frequently, instead of trying to find a way around it. The best bot experience is when you don’t know you’re having a bot experience. So keep it simple.
5. Keep Up with New Innovations
Over time, the technology will change. Don’t plan on creating your customer service bots and forgetting about them. Always be aware of new technologies and innovative uses, and be open to changing and updating your solutions.
6. Choose a Bot Partner with Experience
A partner can take on some of the responsibilities when building and maintaining customer service bots by providing best practices and keeping up with new innovations. Intrado, for example, has experience developing our own proprietary bot infrastructure, but we are also starting to work with industry leaders in cloud-based dialog technology. If you have questions, we’d love to discuss your needs, your current state, your current customer experience and whether now is the right time to go down the customer service bots path.
Bots: The Future of Customer Service
If trends continue, customer service bots will become more commonplace in every area of customer service. Whether it’s voice, mobile, web or another channel, bots will likely play a role everywhere we look, and that can pay dividends for both consumers and contact center employees.
So be on the lookout for examples of customer service bots in everyday life as inspiration for improving your own customer experience. And consider whether a bot could’ve made your task or interaction simpler.