Contact Centres as the AI Frontline

Upland Software and Metric Sherpa and surveyed more than 300 business professionals across a wide range of industries to explore how is AI really playing out in knowledge management? Spoiler: there’s a gap. And it’s big.

AI is everywhere right now. It’s in the headlines, it’s in your boardroom slides, and it’s definitely already crept into your tech stack.

We’re seeing a shift: Businesses are moving past skepticism to some serious FOMO because success stories with AI as a primary character are beginning to materialize.

After you dig through the details, you see that AI isn’t magic. Its real power comes from delivering trustworthy knowledge, the ultimate result of structured, well-organized knowledge management.

Think of knowledge as the invisible, but super rock-solid layer, that makes contact centre AI reliable. Without it, even the most advanced AI can give the wrong answer at the wrong time, frustrating frontline agents and customers alike.

Leaders who understand how important the knowledge layer is and who are willing to strengthen it can guide AI initiatives that actually work for their business, teams, and customers.

But how well are business leaders really doing at building their knowledge foundation?

Who’s Behind the Research?

For nearly 25 years, RightAnswers has been helping organizations deliver faster, smarter support by making knowledge easy to find, use, and trust.

With RightAnswers, knowledge becomes a living two-way link, flowing into conversation, workspaces, AI tools, and the next wave of channels.

Every interaction builds agent confidence, improves accuracy, and keeps your enterprise knowledge ecosystem learning.

Metric Sherpa is a research and advisory firm that bridges the gap between strategy and execution. Its team brings years of hands-on operational experience and backs it up with data-driven analysis, translating the numbers into keen storytelling insights to move the market forward.

TL; DR:

Short on time? No problem, here’s the quick hitters from our research:

• Contact centres are the AI frontline. It’s where success or failure with AI is the most visible to customers.
• Most companies aren’t ready for AI. Only 18.6% of organizations have structured knowledge content, and just 29% believe they’re AI-ready.
• The knowledge activation gap is real. 84% of leaders admit agents still waste time searching for answers.
• Industries vary widely in readiness. Healthcare lags due to the high compliance demands, while tech is moving a bit faster.
• Leaders can act. Now. Start small with AI pilots, enforce governance, and design the full knowledge cycle.

Why Contact Centres Matter Most in the AI Era

Really, it’s simple: agents are the face of your business, and contact centres are the frontline where AI can handle a range of tasks. It’s where big promises either stick or fall flat.

Every single customer interaction is a real-time test of whether your AI investment was worth it. And the uncomfortable truth is, if AI fails, customers notice immediately.

Think about it this way: your CTO might celebrate the potential of AI in earnings calls, but for your agents, they get two realities.

A) AI works, and agents become knowledge superheroes. Or B) AI fails, and they become apologists, scrambling to appease frustrated customers with ad-hoc answers that may not solve the issue at hand.

Option B isn’t a good look, but the blame lies with your knowledge, not your agents. The problem is, if your knowledge foundation isn’t ready, even the best AI platform can’t act as a silver bullet to fix it. In fact, it can actually make things worse.

84% of contact centre leaders say agents still struggle to find answers efficiently.

This statistic reveals a critical disconnect. While organizations pour millions into AI initiatives, agents (the people who actually interact with customers) are still drowning in information chaos.

Are Contact Centres Ready for AI?

According to our findings, the short answer is…sort of, but organizations are getting there. Every industry is at a different stage of adoption:

• Healthcare: slow, compliance-heavy, and very cautious.
• Financial Services: testing the waters but struggling to get around rigid governance.
• Tech: quick to test, but rollout isn’t always smooth.
• Telecom: lots of momentum, with a patchy foundation.

Only 18.6% of organizations have structured knowledge content.

The Knowledge Activation Gap

This is the big idea from our research: The Knowledge Activation Gap. It’s the difference between AI in theory and actual agent experiences with AI.

Here’s what we’ve seen it look like:

• An agent gets a tricky question from a customer.
• They bounce between all their systems searching for the answer.
• They dig through outdated files.
• They Slack their teammates for help.
• By the time AI offers a suggestion, trust is already gone.

Let’s get back to that knowledge foundation. AI systems are only smart because they’re pulling contextual answers built on trusted knowledge.

Fixing that gap takes work; you need formatted content for AI to read, proper content groups, and robust taxonomy and metadata to provide the foundational accuracy AI needs to make smarter decisions and deliver fast, relevant resolutions.

Unless you’ve taken comprehensive steps to ensure knowledge content is governed, structured, and designed for frontline use, the gap between knowledge and action will bottleneck progress, and the confidence in AI will crumble.

What Leaders Can Do Now

So, where do you start? Our research makes it clear that a lot of contact centres aren’t ready for AI just yet. But don’t worry, we have good news! You don’t have to overhaul everything at once. We have a few practical steps you can take today:

• Start with AI pilot programs – Most organizations are testing AI in small, siloed ways that are impossible to scale. Leaders need to start thinking about how AI can integrate into the broader
• knowledge lifecycle – It’s about moving from “proof of concept” to “part of how we work.”
  Start small, think big. Begin with low-effort, high-impact areas like surfacing articles or summarizing content. These are simple wins that can get your frontline teams behind AI.
• Design the full knowledge lifecycle – Let’s be clear, if your knowledge is a mess, AI can’t fix it (trust us, you don’t want to try). Leaders need to look at the process holistically, from how content is created, to how to enrich and deliver it.
• Make governance non-negotiable – Governance is an absolute must. Delegate ownership to the right people to ensure that knowledge content stays accurate and up to date.

At the end of the day, this foundational approach will get you closer to those AI goals. But it’s so much more than getting the proof point on a successful AI pilot.

When knowledge works seamlessly for agents, it goes beyond solving problems. It builds confidence in your entire brand; you have happier, stress-free expert agents, and satisfied customers who recognize incredible service. That’s a win.

To download the full report click here.

Reviewed by: Jo Robinson