Editorial
Building a chatbot can be a tricky affair. How playful should the chatbot be? How many hundreds of emoji-filled responses and whimsical replies should you build in? Should you personalize responses based on the weather in the end-user’s location?
If you find yourself asking these questions — stop. You’re asking the wrong questions.
According to DigitasLBi, 73 percent of US adults are unlikely to use a chatbot again after just one bad experience. This is further supported by a Mindshare report which found 61 percent of people judge poorly performing chatbots more annoying than human representatives.
The secret to building a truly good chatbot experience is all about performance. According to responses in the Mindshare study, a poorly performing chatbot is what ticks consumers off, not a rigid but efficient one.
But what goes into the kind of bad experience that drives consumers to choose waiting on hold to speak to a customer service agent over interacting with your chatbot? We turned to thought leaders and practitioners to gain insight.
Related Article: A Good Chatbot Is Hard to Find
Why Do Most Chatbots Miss the Mark?
Studies show that 69 percent of customers would prefer to speak to a chatbot rather than a human sales or support representative. That means customers are convinced by the concept of a chatbot in theory — and yet most of them are not convinced by the “cuteness” factor.
We know that because 48 percent of consumers say that it “feels creepy” when a chatbot pretends to be human, 60 percent find it “patronizing” when a chatbot asks how their day is going.
“The primary objective of a chatbot is to resolve queries and get the customer from point A (the initial interaction) to point B (a quick reply button or CTA) as fast and efficiently as possible,” explained Devin Picknell, software marketing specialist at G2 Crowd. In other words, Picknell is in favor of less small talk, and more progress.
“The biggest mistake companies make with chatbots is trying to deploy them as a one-to-one replacement for live representatives. Companies usually try to 'stretch' the functionality of chatbots [to make them seem human, and by adding extra functionality],” added Derek Gleason, content lead at ConversionXL.
Learning Opportunities
Webinar
Jun
6
The Future of Social Media: Emerging Trends and How to Stay Ahead of the Curve
The impact of emerging technologies like AI on social media
Webinar
Jun
7
The Building Blocks of Personalization
This webinar will explore how personalization can create powerful ecommerce shopping experience.
Webinar
Jun
8
Attracting—and Keeping—Your High-Net-Worth Investors
Join PwC to dissect their recent survey findings on a specific customer subset — high-net-worth investors.
Webinar
Jun
13
The Future of Contact Centers with AI at the Helm
Learn how to balance human intuition and interaction with the potential uses of AI to create delightful and lasting CX
Webinar
Jun
20
3 Steps to Unlock Your Customer Data and Drive Revenue
Discover how Twilio Segment’s #1 Customer Data Platform helps implement data-driven strategies.
Webinar
On demand
AI for Marketers: Optimize Personalization Across the Funnel
Understand the benefits of utilizing AI across your entire marketing stack
Related Article: Learning From Our Chatbot Implementation Mistakes
Going Beyond Cuteness: How to Ramp Up Chatbot Engagement
Picknell advises brands to rigorously test their bot on a regular basis to “flush out any instances where users can get sucked into endless, frustrating loops.”
Ryan Lester, director of customer engagement technologies at LogMeIn, recommends brands make use of context and Natural Language Processing (NLP) to gain a better understanding of consumer intent. “The key for implementing a customer-centric chatbot is to make sure it’s conversational and can engage with customers in the way they naturally talk [while] understanding [their] intent,” explained Lester.
But while NLP can make chatbot interactions more personal via free-text and NLP, Philip Say, VP of innovation product management at Sutherland Labs, said having a better overall user experience (UX) can encourage users to use the quick reply buttons. “Ultimately, it comes down to being less about whether to use quick replies vs. NLP processing, and more about improving the UX of an old process into something better or more effective. When the UX is better and the utility relevant, users actually appreciate quick reply buttons.”
Say added that in some cases, removing the free-text and NLP option for certain areas can boost engagement. He explained how his firm adopted this approach. “At Sutherland Labs, we’ve customized user experiences where certain steps are 100-percent mediated by quick replies, and others where NLP capabilities or message composer is dynamically presented to the user for specific steps.”
Finally, Lester advises brands to be aware of “the next big trend” in the customer service chatbot sector, which is proactive artificial intelligence (AI). “Proactive AI anticipates the needs of customers based on known information [such as] location or past purchases, and where they are in their current journey, to provide them with relevant content at the appropriate time.”
Lester added that integrating proactive AI into chatbots will help “businesses increase conversion rates and overall customer satisfaction” while making the customer appreciate what modern chatbots can do for them.
Learn how you can join our contributor community.
