A human-centric approach to intelligent service

At ServiceNow, we’ve always believed that customer service and support must remain fundamentally human. At its core, support is about people helping people—solving problems, building trust, and offering empathy during moments that matter. But like many support organizations, we’re challenging ourselves to do more with less—to control costs without compromising the quality of service our customers expect and deserve.

This ongoing tension sparked a critical question for us:
How do we adopt technologies like AI and automation to boost efficiency, while still delivering personalized, high-touch experiences?

The answer wasn’t immediately clear. Like many companies, we began by focusing on where AI could improve two key areas: self-service and agent productivity. As both customer zero and an innovation partner for our own AI-powered Customer Service Management (CSM) and Technology Provider Service Management (TPSM) products, we had early access to emerging capabilities like case summarization, routing, and knowledge article generation. Our IT team was also exploring AI across various internal use cases, and they needed real-world feedback. As a result, we found ourselves rapidly piloting AI features.

Some of these early use cases delivered quick wins. Others needed iteration. But each experiment taught us something valuable—not just about the technology, but about what it takes to make AI successful in a support context.

One of the biggest lessons? Without a clear strategy, maturity suffers. We’ve seen this trend industry-wide. Our 2025 Enterprise AI Maturity Index research shows that AI maturity has dropped nine points year over year, and we believe that’s largely due to rushed adoption without foundational planning. That’s why we’ve since paused to rethink our approach.

We’re now exploring what it truly means to build a support organization composed of both human and AI agents. We're aligning this exploration with our hospitality vision, where empathy and ease of experience remain paramount. To move forward with intent, we launched an internal idea-a-thon to crowdsource where Agentic AI could make the most meaningful impact. The response? Over 180 unique use cases submitted by our staff—proof that the best ideas often come from the people closest to the work.

If you’re asking yourself where to start, our advice is simple:

Start small. Learn fast. Scale what works.
Early use cases won’t yield massive savings, but they’ll teach you the principles that can unlock broader transformation.

Despite the growing pains, our investments in AI are already driving real results:

·       AI agents support 72% of customer self-service requests—allowing customers to get fast answers and freeing up human agents for more complex needs.

·       AI Agents are automating 37% of our customer support case workflow—handling tasks like routing, categorization, and summarization.

·       60% of knowledge articles are now AI-generated, accelerating time to publish by an average of 88%. This has enhanced the relevance and robustness of our knowledge base—ultimately strengthening our self-service channels.

These outcomes are encouraging, but we’re not stopping there. We’re now thinking beyond traditional workflows. What if AI wasn’t just inserted into existing support processes—but instead, we designed new workflows with AI as the foundation?

Imagine a world where, instead of choosing between chat, search, or case forms, a customer starts with a simple conversation—just like they would with a search engine. An AI agent listens, understands, and works to resolve the issue—looping in a human when it’s necessary or requested. This shift from reactive triage to proactive understanding reduces friction and accelerates resolution.

As we continue our journey, here are a few guiding takeaways:

·       A well-planned change management strategy is critical for successful AI adoption.

·       Knowledge Management is the linchpin—fueling both automation and customer self-service.

·       AI is best deployed fast, but its value takes time to measure.

·       Early wins often come from automating low-complexity tasks—but real transformation comes from reimagining the experience end-to-end.

At ServiceNow, we’re proud to be our own best customer. And we’re committed to helping our customers do what we’re doing every day:

Deliver world-class service with the perfect balance of intelligence and humanity.