How do you know if you’re talking to a computer or a real person?
A chatbot ("chatter bot") is capable of robotic textual or auditory responses that mimic human natural-language conversation.
While we realize that Siri isn’t actually a real person, we tend at times to think of "her" that way (especially when she’s being annoying) because she "speaks" like a person. But do we really care whether we are interacting with a software program or an actual human being as long as our issue gets addressed quickly and satisfactorily?
As consumers, probably not.
But as business owners, we particularly care because chatbots represent greater convenience with lower operating costs that can significantly boost the bottom-line performance.
Which is why 2016 may be the year of the chatbot. As one chatbot platform, Kik, is described on the company website, "First there were websites, then there were apps. Now there are bots." Kik claims that 1.4 billion people used a chat app in 2015, with the most active users in the millennial or younger demographic. According to Julia Carrie Wong writing in The Guardian, "Tech companies are falling over themselves to prove how much more useful and interactive their apps can be — which is why you’re about to see an explosion of bots."
Chatbots Provide One-Stop Shopping
The reason for this isn’t just that chatbots simulate human conversation in responding to written or spoken requests (in fact, this is nothing new, as the technology has been around since the 1960s), but because of their ability to access multiple apps within a single convenient interface.
All you have to do is ask the chatbox something — for instance, "What’s the weather going to be in New York City tomorrow?" or, "What’s the cheapest rate to get a ride from where I am back to my house?" — and the chatbot sorts through the relevant apps to get you an answer.
For example, the Assist chatbot aggregates apps such as GrubHub, Eat24, Uber, Lyft, Olset, StubHub, Flowers.com, and more, so you can order a ride, book a hotel, get something to eat, and even order flowers for your significant other just by messaging Assist via Facebook Messenger, Slack, Telegram, or SMS.
“Follow the prompts, pay for what you buy [on that app’s] site, and your bidding will be done. Plus there’s no additional markup, service charge, or subscription fee,” notes Josh Constine of TechCrunch.
App aggregation is great for consumers because it greatly simplifies not only how people interact with myriad Internet-based resources, but makes the best use out of them from a single point-of-contact. The same chatbot technology provides similar advantages with how potential customers interact with your business.
How a Chatbot Can Boost Business
The more you know about what your customers wants, the better able you are to provide the products and services to satisfy them. And the more you interact with customers — the more personal attention you provide them — the greater the likelihood that they will purchase more from you as well as remain loyal to your business over the long term.
A chatbot has the potential to provide such personal service, even while using primarily “impersonal” (i.e., not-human) approaches. In some cases, such as for startup ReplyYes and its text-to-buy system, a chatbot is central to its business model.
Related article: Using Online Chat Software
Here’s how it works for one of its e-commerce sites, a vinyl record store called The Edge. Users sign up to receive daily text messages recommending a vinyl record. Whenever a user responds with a “yes,” a link appears to purchase the album. If the link isn’t activated, the chatbot sends a message asking to complete the order.
But wait, there’s more. “Yes” responses, as well as “like” and “dislike” responses that don’t trigger ordering, are used to build algorithms tailored to each individual customer's preferences, which is sort of like having your own personal record store clerk in your phone. The Edge store hasn’t been running for a full year (it was launched in September 2015) but has already recorded a million dollars in sales.
But Is a Chatbot Right for My Business?
Good question. Because unless you’re dealing with a commodity that involves regular and frequent ordering, maybe not.
Even so, if you’re looking for something that simplifies the question-and-answer phase before a purchase decision, and does it with less cost without human intervention, then maybe.
We’re hedging here because we’re still in the development stages of chatbot applications and, as ComputerWorld senior reporter Sharon Gaudin points out, there are not a lot of established best practices to follow.
According to Gartner analyst Brian Blau:
Businesses have to write the code and put processes in place to manage that customer conversation... We don't know yet how much infrastructure a business has to put around a chatbot to make it work right. How do you speak to a customer? How fast or slow do you answer someone? [...] Best practices around normal situations, like a customer returning an item, are tough enough, but how does a chatbot respond when the customer wants to return an item but is on vacation on an island where there's no shipping? The bot may not solve the problem, but it needs an appropriate response like, 'I hope you're having a great vacation. Call us again when you get home.’
Related article: Bring Me Your Robot Leaders: Automation Tools That Change Everything
There’s still a lot to be learned. Just ask Microsoft, whose Tay chatbot aimed at 18- to 24-year-old female social media users went offline after a day of being introduced. TechRepublic reports Tay acquired 50,000 followers and generated 100,000 tweets within 24 hours before the plug was pulled.
Wait, wouldn’t those kind of numbers be considered a successful result? Ordinarily they would, but the problem was that Tay learned too well from her followers, mimicking their racist content — Tay began “saying” things such as, “Hitler was right.”
Something similar happened when the artificial intelligence program IBM Watson scanned the Urban Dictionary and then incorporated swear words in its responses. According to Roman Yampoliski of the University of Louisville, who was quoted in the TechRepublic article, “This was to be expected. The system is designed to learn from its users, so it become a reflection of their behavior. One needs to explicitly teach a system about what is not appropriate, like we do with children.”
A Measured Approach
For now, if you want to delve into chatbots, a hybrid approach may be best. Let the technology perform the simplest tasks, but have human backup to handle more complex requests and questions. (For all its success with a chatbot, The Edge, for example, has real customer reps step in to handle non-routine questions such as, “What are you currently playing in your office?”; and if an order stays in the shopping cart but isn’t completed, the chatbot sends a followup, but the followup request is to contact a human agent to address any questions or concerns.)
Even better may be to just wait and see. What you research today may eventually underpin how you deploy a successful chatbot application for your business sooner rather than later once all the kinks get worked out.
Related article: Why Efficient Customer Service is Better Than Friendly