Conversational AI for Customer Service
The Pain
What is the secret to customer loyalty?
The answer straight from ~50,000 consumers, according to a massive survey conducted by Gartner, was, make it easy to get service.
To find the recipe for “ease,” Forrester asked 5,000 consumers about their biggest pain points in getting good service.
- Consumers pointed to lack of contact center agent knowledge and inconsistency of answers, followed by ineffective websites.
When we asked over 500 agents about their biggest hassle in being able to deliver good service,
- Agents pointed, not surprisingly, to the difficulty in finding the right answers.
The Cause: Customer Queries are Getting More Complex
Agents and self-service systems need to have conversations with customers to resolve their issues and provide advice. This would mean asking the next best question and performing the next best action, based on historical and real-time context, situational knowhow, and compliance requirements—which would require agents to have a 20-pound brain!
Contact centers have attempted to address this through non-stop training, something today’s millennial and Gen Z agents despise. But, stay-at-home orders from COVID-19 have made training and collaboration very difficult.
The Cure: Conversational AI and Knowledge
The answer lies in leveraging conversational AI and knowledge to guide agents and self-service systems through their dialog with the customer
Using conversational AI is similar to employing a GPS to get from Place A to Place B, that uses the best route, while complying with requirements such as avoiding the freeway and not turning the wrong way into one-way streets.
Conversational AI enables contact centers to scale effective, efficient, and compliant customer service and sales across in-house, gig, and outsourced agents, a real challenge amid the COVID-19 crisis.
The Method
When done right, conversational AI can transform customer service. Here are some best practices for success.
- It is about the content and knowledge that customers need and not what you have or think they need. Digital customer interactions can provide an accurate picture of what is needed.
- Don’t try to boil the ocean. Use the 80-20 rule to prioritize where conversational AI is needed.
- No one AI-technology hammer works for all business-need nails. For instance, while it’s better to use machine learning for use cases with low business risk (e.g., making contextual promotional offers on an eCommerce site), you are better off using supervised or curated learning when the risk for the business or the customer is high, such as for management of high-value assets or a life-and-death treatment decision for a patient.
- Go with a vendor who has a proven track record of success and best practice expertise in this domain (you don’t want a vendor to learn AI at your expense).
- Make sure to use a common AI and knowledge engine across all customer touchpoints so that repeat contacts don’t flood your contact center—you can ill-afford it in this crisis!
Conversational AI at Work
Here are sample metrics and real-world examples from our clientele. Note that the business value of conversational AI goes way beyond these metrics.
First-Contact Resolution (FCR)
FCR (First-Contact Resolution) is a key customer-focused contact center metric that significantly reduces consumer effort. While FAQs, search, topic-tree browsing help with simple queries, more sophisticated technologies like conversational AI are essential to resolve issues of medium to high complexity at first contact.
- A premier telco client improved FCR by 37% with eGain’s conversational AI. In fact, now, any agent is able to handle any call, the “holy grail” in contact center customer service!
Average Handle Time (AHT)
AHT (Average Handle Time) without FCR increases customer effort and defection. Happily, conversational AI can transform both seemingly conflicting metrics.
- A premier banking client reduced AHT by 67% while improving FCR by 36% by leveraging conversational AI to guide customers to answers, while complying with industry regulations!
Annual Training Hours (ATH)
- A global bank secured the #1 spot in customer service NPS and reduced ATH by 50% even as it expanded to 11 countries with mostly novice agents in its workforce!
Some technologies improve customer service on the margins, some enable incremental improvement, but only a handful actually transform it. Conversational AI clearly falls into the last category.
Related resources
- Conversational AI Explained—What is it and Why is it Important
- 12 Reasons Why Customer Service Chatbots Fail
- Conversational AI for Sales: The Missed Opportunity to Boost Revenue
- The Essential Complement to Agent Training in the Modern Contact Center: Knowledge and AI
- Cost Avoidance with Artificial Intelligence
- Digital Experience Transformation Needs AI and Knowledge
- Whitepaper: AI in Customer Service Contact Center