Inside the Mind of a Modern Bot
Chatbots are becoming the cornerstone for many customer engagement strategies. The reason being is that chatbots are efficient, becoming more effective by the day, and help take some of the strain off the customer service team. It’s the effective part I’d like to dig into.
Chatbots have been around for some time, but their ability to truly enhance the customer experience has been called into question in the past. Mostly because the early technology could only handle very basic, specific tasks. “How do I change my shipping address” for example. But if a customer deviated from that language in any way, they would become significantly less effective.
All that is changing. Chatbots today are intelligent. They are contextual. And they are always learning. We are living in a world where chatbots are helping people not only make a purchase, for example, but help them decide on what purchase they want to make. Imagine a world where you are looking to book a vacation, but instead of knowing where you wanted to go, you start doing some research on your phone. A widget then pops up, recognizing you’ve been looking at a few European cities, and asks if you’d like to book a city-style vacation. The conversation then goes something like:
Bot: Would you like to book a city vacation?
Bot: Here are a few suggestions [Links to “Top 10 City Breaks”]
Customer: What’s the cheapest flight I could get to London?
In this example, the bot has seamlessly continued the conversation to help build a picture of the information it needs to provide the right answers. Understanding that you were looking for a city break, and planning a journey, it reads London and knows that you are asking for the least expensive flight, and that the next piece of information it needs is when he would like to go.
Bot: Which dates were you thinking of flying?
Customer: The 17th – 1st of next month or thereabouts
Natural Language Processing helps the bot understand the dates that you are giving, even though you haven’t mentioned the month by name. It also needs to process the word thereabouts to understand the dates are flexible, but want to stay in that range.
Bot: Would you like to fly out from Newark or JFK?
Customer: Either one [But appreciates the bot recognizes him as a New Yorker]
Bot: Take a look at these options
Customer: Do you have anything later than these I prefer night flights
The bot here understands that ‘later’ means later in the day, rather than the month, and can look for evening or overnight flights, even if they aren’t the cheapest option anymore. However, it still manages to hold onto the intent from the questions before. Keeping the context that the customer is still looking for a flight to London within 1 or 2 days of the dates he mentioned, now a new piece of information has been added to the list.
Bot: The only night flight that week is on the 15th. You could come back overnight too, on the 30th. The flight gets back to JFK at 7:15am on the 1st. Does that work for you?”
Customer: Yes, but I need to finalize plans. Can you save that for me?
Bot: Sure, no problem. Anything else I can help with?
So in this example, you leave the chat in your phone, confirm the plans work for you and revisit via a website. The expectation is that when you start a chat, it will hold the planned agenda and skip to the confirmation of the trip and payment.
This seamless, human experience across channels is what differentiates a today’s chatbot technology from its predecessors. It understands the intent of the customer, knows which information to hold onto and which to let go of, uses context from previous interactions, and knows the location and language of the user.
The future of customer experience is a hybrid of both automated and agent-assisted technology. Bots, which until recently have been designated for simple, quick questions and answers, are going to be elevated to help with more complex issues and be able to deliver round the clock service when human agents aren’t available. The customer experience is going through a transformative change, and smart chatbots are leading the way.