娇色导航

Q&A: AI and customer experience strategies for staying ahead of the curve

Maintaining a unified customer experience (CX) across websites, mobile apps, and social media is time-consuming and expensive for any company. So, it’s understandable that organizations are turning to artificial intelligence (AI) and machine learning (ML) to augment their CX with chatbots and real-time data analysis.

Research shows that IT and business leaders are bullish on using AI to connect with customers. CIO’s State of the CIO report for 2024 revealed that 40% of IT decision-makers believe AI is the most important technology for improving the customer experience.

But while AI and customer experience are slowly merging, AI is still just one part of an overarching CX strategy. When AI technology for CX is deployed too aggressively without a clear plan, it may do more harm to customer relationships than good.

With coming up on May 14, we asked Mike Egli – a seasoned customer experience consultant and CX Client Principal at RingCentral – to share his thoughts on how companies can be more strategic when integrating AI to improve the customer experience.

How do you envision AI changing the customer journey, and what challenges do you foresee with AI adoption?

A strategy for AI and customer experience starts with first understanding your business challenges. The challenges could be revenue loss, lack of efficiency, and too much overhead. But let’s figure out where AI can make the biggest business impact and start there.

For instance, contact centers want to reduce overhead because the human cost of labor is high. Because of GenAI, self-service chatbots can now quickly evaluate customer data and give responses in a conversational way that wasn’t possible a few years ago. So they’re taking low-hanging fruit customer support tasks and switching those to a chatbot virtual agent.

That will lead to positive customer experiences – especially among Generation Z. The majority of Gen Z only want to be serviced through chat. It’s a compelling combination – the preferences of a young demographic intersecting with AI innovation for a better CX. And you’re also reducing your overhead by millions of dollars by taking 30% of call center activities that used to go to a human being and handing it over to AI.

Nevertheless, most companies do not only serve Gen Z; older customers may prefer a live human agent. So, the element of choice is important.

But even if the preference is a live person, the AI doesn’t just stop. AI will support a human agent by listing the customer’s previous purchases and providing recommendations in real time, effectively making a human agent superhuman.

This can turn a 10-minute customer call into a five-minute call – which, over time, will have a positive impact on the bottom line.

What are the metrics companies should use to prove that AI/automation is worth the investment?

My suggestion is always to find the problems first. Then, rank the problems based on how much they affect the business. And that will create the success metrics that AI will help address.

For example, the success metrics could be reducing overhead by 20% or increasing conversions by 25%.

AI is getting cheaper to implement, but it’s not cheap. You don’t want to just throw money at AI without defining the business goals it will help you achieve.

If your average revenue per customer call is $400 … after an AI implementation, it should be $410 or $420. The metrics vary by industry, but it should come down to this: find out the CFO’s most important business metric, and let’s see how we can target that with AI.

Can you share an example of a successful AI implementation that improved customer satisfaction?

A gaming company I work with has a rule that customer support agents are not allowed to transfer a call. This is unique and goes against the conventional guidance of having multiple tiers within contact centers.

But they want the personal touch – human agents only. They didn’t want to use AI for chatbots; they wanted AI to enhance human agents. The company was willing to spend on the overhead of labor, but they wanted quicker answers and lower handle times. This is a case of AI being used as a data-mining virtual assistant for humans. The approach worked great for this company that prioritizes human-centered CX. They were able to get 6-minute calls down to 3 minutes, a significant gain in efficiency.

How about an example of an organization mishandling an AI and customer experience strategy?

I worked with a large US bank that built its own AI from the ground up rather than purchasing an off-the-shelf solution. Their goal with AI was to eliminate over 90% of human interactions with customers. That’s very ambitious, and it has led to some major challenges.

They stuffed their AI solution into the bank’s mobile app as a customer service bot – but this approach was unsuccessful. Customers were frustrated that requests could not get escalated to a human being if needed. Over 40% of customers uninstalled the mobile app!

AI is simply not built yet to take on 90% of customer support cases. I asked the bank where the 90% metric came from. Did they partner with Forrester, McKinsey, or another external consultancy? They said no. They sourced it internally. It was a strategy set by finance to reduce contact center staff while maintaining the same revenue.

You can take this as a cautionary tale about being too aggressive with AI and customer experience without doing enough research first. 

Mike Egli will discuss strategies for using AI to enhance customer journeys at on May 14.

for FutureIT Boston.