The hidden value of AI in customer service: More conversations, not fewer

Most companies invest in AI-powered customer service to reduce costs and handle fewer enquiries. But what if the real opportunity is exactly the opposite? Our data suggests that when customer service becomes easier to access, customers start asking more questions — creating new opportunities to improve both customer experience and business performance.

When customer service gets easier, customers ask more

Across our customers in tourism, education, e-commerce, hospitality and retail, we've noticed a consistent pattern: inbound conversations increase after AI-powered customer service goes live.

At NKI, the number of customer dialogues increased by 200%. The explanation is surprisingly simple. The demand was already there. Customers just weren't willing to send an email, wait for a reply or contact support during office hours.

Remove those barriers, and more customers start engaging. This changes the way we should think about AI in customer service. Instead of asking how many enquiries we can eliminate, perhaps we should ask how many customers we can help.

The AI investment paradox

AI adoption is accelerating across the Nordics. Six out of ten Swedes now use AI services, compared with four out of ten last year (Svenskarna och internet, 2026). In Norway, nearly half the population uses generative AI, making it Europe's most AI-mature nation (Microsoft Global AI Diffusion, Q1 2026).

Businesses are moving too, but results aren't keeping pace.

According to the BCG Nordic AI Report 2026, 35% of Nordic companies have deployed AI in customer service, while only 4% report meaningful returns on their AI investments.

One possible explanation lies in what companies are trying to achieve. According to the Tieto Nordic AI Survey, 65% primarily use AI in customer service to improve efficiency, while only 27% prioritise improving the customer experience.

That difference in priorities matters. When the goal is simply to reduce costs, success is measured by how many enquiries never reach a human. But when the goal is to improve the customer experience, the focus shifts towards resolving issues, increasing availability and helping customers move forward.

Customers want instant answers. But only if they're useful.

Research shows that 82% of customers prefer an instant chatbot response over waiting for a human when asking straightforward questions. At the same time, 77% say a poor self-service experience is worse than having no self-service option at all.

The message is clear: speed matters, but not at the expense of quality. A chatbot that delivers irrelevant answers or traps customers in an endless loop doesn't improve customer service. It creates frustration.

The goal should therefore be to automate what AI can handle well, while making it easy for customers to reach a human when needed.

What 25,000 customer cases have taught us

In Q3, konvers.ai handled 25,000 customer cases, with a platform-wide automation rate of 73%. In October, that rate increased to 77%.

But the most interesting insights aren't necessarily found in the automation figures.

Availability creates demand. For e-commerce and education businesses in particular, evenings and weekends are important periods for customer interaction. AI makes it possible to provide immediate assistance even when traditional support is closed.

Better information leads to better automation. Customers typically start with automation rates of 60–70%. Through continuous improvements to their knowledge bases, individual customers have reached 80–90%.

Every question creates an opportunity to improve. When the assistant can't answer a question, it reveals a potential gap in the company's information. Addressing that gap doesn't just improve the AI assistant. It can also make the website, support content and overall customer experience better.

Together, these insights point towards something bigger than automation: customer service that continuously improves as a business learns more about its customers.

From cost centre to business opportunity

The next step is already taking shape. According to BCG, 54% of Nordic companies are experimenting with AI agents that go beyond answering questions.

Instead of simply providing information, AI can increasingly help customers complete tasks: making bookings, managing orders, identifying relevant products and resolving issues before they escalate.

For businesses, this creates opportunities that go beyond operational efficiency. Customer service can help reduce friction in the buying journey, improve customer satisfaction and support revenue growth.

Yet the Nordic market still has some catching up to do. While Swedish consumers are among Europe's fastest adopters of AI, only 27% of Swedish businesses report using AI at work, compared with a global average of 36% (EY, 2026). In Norway, the corresponding figure is 23%.

Consumers have already changed their behaviour. The opportunity for businesses is to respond.

The goal was never automation for its own sake

The companies getting the most out of AI-powered customer service aren't necessarily those with the largest budgets or the most sophisticated technology. They're the ones asking the right question.

Not simply: How much can we automate? But: How can we help more customers, more effectively?

Because the real value of AI isn't just handling the questions customers already ask. It's making it easier for more customers to ask them in the first place.