AI in customer success: why intelligent adoption is crucial
(This blog expands on themes from our podcast series with Mint, The Long Game Dialogues. Read on to know more.)
If you work in a corporate setup today, chances are you've been asked to use AI to make your work easier—maybe even urged to use it before you've had a chance to think of a solution yourself.
From automating routine tasks to analyzing customer data or churning out content, organizations are investing heavily in AI infrastructure. This is especially true in customer service, which is often the petri dish for new tech initiatives.
According to a report by Bain & Company, a 5% increase in customer retention can raise profits between 25% and 95%. That alone explains why companies are eager to pour their resources into improving customer experience. The logic is simple: Keep existing customers happy and they'll stick around (and keep spending more).
As AI becomes more deeply woven into how companies serve their customers, it raises an important question: Can we make our systems smarter without losing the genuine human connection that customers still value?
The corporate itch to adopt AI
For businesses, it's obvious; automating customer service helps them keep up with rapidly evolving customer expectations, reduce the burden on their support teams, and most importantly, control costs by reducing the need to hire new staff.
But to understand why AI is being adopted at such a large scale, it helps to look at how dramatically customer expectations have shifted in recent years.
The Intercom Customer Service Trends Report 2024 found that customer expectations around response time have increased by a whopping 63% and resolution by 57%. With customers' growing impatience and pressure to meet their expectations, it's no surprise that companies are leaning heavily on AI to meet these increasing demands.
Take Telstra for example. The Australian telecom giant's CFO Michael Ackland recently said:
"We spend over $2bn per annum in operating costs across activities from sales to contact centres, activation, billing, and customer management. And we think AI will revolutionise these activities."
The value gained is evident:
Reduced costs
Quicker resolutions
Happier customers
Less burnt-out employees.
When all of these are top-priorities, the appeal of AI is understandable—especially when the systems in place benefit everyone involved. Customers get quicker answers to simpler problems, and employees can focus on more complex and meaningful work.
Efficiency isn't empathy
While AI has it's clear advantages, there are still a few things to keep in mind when it comes to how customers feel when interacting with your brand.
How many times have you found yourself talking to a chatbot that genuinely couldn't understand your question, likely because you're picking options from a limited checklist that it throws at you instead of having a conversation? And if your concern doesn't fit within the choices listed, too bad.
What really makes it frustrating is the digital wall that it erects, making it nearly impossible for you to reach a human who might actually be able to help. You're left running around in circles with half-baked answers.
A friend recently had a run-in with one. They'd ordered three cans of soda from an Indian quick-commerce app, only to receive two damaged ones. When they tried reaching out for help, they were promptly met with a chatbot. The chatbot requested a photo of the damaged items and offered a refund—and annoyingly, it only issued a refund for one can of soda.
The response was quick, yes, but also dismissive. There was no way to explain the issue properly or escalate it to a human support agent. This is a good example of how speed alone can't make up for the lack of nuance in customer care.
Of course not all AI-led support experiences are the same. But a tool stops being useful when all it does is create more problems rather than help solve the problem you went to it to begin with.
And that raises some concerning questions: Are we solving the right problems or are we just adding complexity where there was none?
The price of shortcuts
Sure, it's possible to refine LLMs (large language models) and train chatbots to handle these problems more smoothly. But the real problem lies in companies rushing to roll out these features—often long before they've reached any level of AI maturity. Choosing to automate first and course-correct later is a gamble at the expense of customer trust and brand loyalty.
Think of it this way: You have a broken tap and water flooding your entire bathroom.
Would you rather have someone who's simply read a manual on plumbing to fix your tap or someone who's been under the sink before?
You want someone who's been under the sink before! Someone who not only knows how to fix your tap, but understands why it's broken in the first place. That's what an experienced support team can bring: not just a scripted solution that's used to patch up a situation, but instincts and knowledge that come with years of experience.
Somewhere along the way, we may have swung too far in favor of automation—treating human support agents as easily replaceable instead of essential.
In 2022, Klarna, a Swedish fintech company, fired 700 employees and replaced their roles with AI-powered systems. And by 2023, they'd stopped hiring human employees altogether. But earlier this month, their CEO, Sebastian Siemiakowski, admitted that the AI agents had failed to meet their expectations and prioritizing cost over quality had negatively impacted Klarna's service.
Reflecting on the company's decision, he went on to say that, from a brand perspective, it's critical to be clear to customers that a human will always be available if they ever require them.
Finding balance
Of course, with the right checks in place, AI can be transformative; but without thoughtful integration, we run the risk of undermining the very experience we're aiming to enhance.
That's why we decided to talk to a few industry leaders to understand how businesses can strike the right balance between leveraging technology and nurturing customer relationships in a tech-driven world. So we teamed up with Mint for a chat with Vidya Vasudevan, Zoho's global head of community, and Amit Arora, CIO at SHR Lifestyles.
Here are some key takeaways from the conversation:
Don't shy away from new technology. Waiting too long can hold your business back, especially with customers' growing demands.
Use AI only when you need to. Not everything requires automation.
Introduce AI with clear goals. Technology should support your teams, not complicate processes.
Find out what your teams spend the most time on. If it's on repetitive or low-value tasks, consider using AI to reduce the load.
Before deploying AI, ensure that you have a solid understanding of your own customers. It's your responsibility to train your AI tools properly.
Human connection still drives customer trust. It's important to realize when a human needs to intervene in automated processes.
For all the details, you can watch the full conversation here.
