CX Contact Center

The CX leader's checklist: Self-service, data, and the humans behind it all

How forward‑thinking teams are using AI to connect data, elevate agents, and deliver experiences that feel more personal at every touchpoint. 
6 min read

Published on May 20, 2026

Image Placeholder
CX leaders know the stakes. Seventy-six percent of contact center leaders say their agents are overwhelmed by systems and information. Agent turnover sits at 52% annually. And nearly half of all customers will switch brands after a single poor experience. The pressure to modernize is real, but so is the opportunity.

On May 14, CX expert Adrian Swinscoe joined Zoom's senior product marketing manager Chandler Galt for a 45-minute fireside chat that cut through the noise and delivered a clear, practical roadmap for building a CX operation where humans and AI work together, not against each other. Here are the biggest takeaways from the conversation.

The human-AI partnership: Steps to create connected, effortless CX

The CX landscape is in its third wave

Adrian opened by framing where the industry stands. The first wave of CX technology was on-premise and voice-centric: basic IVR and phone trees. The second wave brought cloud-based, omnichannel platforms, but stitched-together systems created new silos instead of solving old ones. Now we're in the third wave: agentic AI that's built into the platform, not bolted on, with connected data powering real-time intelligence.
 
The problem? Many organizations are still paying what Adrian called the "toggle tax," with agents juggling multiple tabs and disconnected systems, losing context with every click. Hundreds of data points can be created during each customer engagement. That fragmentation doesn't just frustrate agents, it drives customers away.

Three pillars of the human-AI partnership

The heart of the session was organized around three themes from Adrian's mini-guide series for CX leaders. Each one came with concrete actions attendees could take back to their teams immediately.

1. Design self-service that continuously learns

The self-service gap is massive: 81% of customers try to help themselves first, but only 14% fully resolve their issues (Gartner 2024). Adrian drew a critical distinction between generative AI, a "well-read librarian" that answers questions, and agentic AI, a "skilled personal assistant" that takes action like checking eligibility, initiating refunds, and updating systems. Gartner predicts agentic AI will autonomously resolve 80% of common issues by 2029.

The key takeaway: Start with high-volume, transactional use cases for quick wins, define escalation triggers based on sentiment analysis (not just complexity), and build feedback loops so your virtual agent learns from every human agent intervention. Adrian also emphasized the "Service Dividend," reinvesting automation savings into your people rather than cutting headcount.

"You've got to figure out how to build feedback loops into your whole system, so your virtual agents are learning from the best practice of your human agent interventions. So you end up with a symbiotic relationship between AI agents and human agents."

Adrian Swinscoe

A CX Leader’s Playbook for Intelligent Self‑Service

2. Connect data and teams for unified intelligence

Intelligent automation is only as good as the data behind it. Adrian and Chandler explored how organizations can move from hoarding "Big Data" to building shared intelligence that drives action: proactive outreach, predictive insights, and real-time decision-making powered by connected data.

The action steps here were direct: Conduct a technology stack audit, map your data flows to identify where customer context gets lost, clean and consolidate your data before layering on AI, and build cross-functional visibility between the contact center, product, and engineering teams. The companies seeing the biggest returns are the ones treating their contact center as a strategic data hub, not a cost center.

"The idea of unified intelligence is exciting because it lets us lean on technology to do what it's good at, and allows human beings to focus on what they're good at — the human connection, understanding context, empathy, and personal relationships."

Adrian Swinscoe

The CX Leader’s Guide to Breaking Down Data Silos

3. Give agents AI-driven confidence

Perhaps the most resonant theme was the shift in the agent's role, from "data entry clerk" to "relationship orchestrator." Adrian outlined three phases of the AI-augmented agent experience: pre-interaction (AI surfaces context, intent, and sentiment before the agent says hello), during interaction (real-time guidance as a "digital whisperer"), and post-interaction (automated summaries and action items that minimize after-call work).

The results speak for themselves: 37% of companies have reduced agent attrition with agent-assist tools (Metrigy), and McKinsey estimates 30–45% productivity gains are possible. But Adrian stressed that none of it works without trust. Leaders need to involve agents early, give them the ability to override AI recommendations, and celebrate wins powered by the partnership, not just the technology.
"Technology that is foisted on employees invariably is not adopted. You have to bring your agents into the conversation, ask for their views, get their input, and involve them in choosing what's going to work — and what isn't."

Adrian Swinscoe

The CX Leader’s Guide to Empowering Agents

The bottom line: Start with your people

When Chandler asked Adrian for the one thing attendees should walk away with, his answer was clear: this is not a one-and-done solution. The changing role of agents means continuously iterating. Don't get caught up in the hype cycle. Focus on outcomes for your people, and positive customer outcomes will follow. The age of AI is not the end of the human agent. It's a tool to give your people the confidence to do their best work.

Your leadership action plan from the session:
  1. Design — Build self-service that continuously learns from human expertise. Audit where customers drop off, start with high-volume use cases for quick wins, and build feedback loops so your virtual agent improves from every human agent intervention.
  2. Connect — Unify data and teams for shared intelligence. Map your data flows to find where customer context gets lost, consolidate before layering on AI, and build cross-functional visibility between the contact center, product, and engineering teams.
  3. Elevate — Give human agents AI-driven confidence and autonomy. Involve agents early, start with AI-powered summaries before moving to real-time guidance, and give agents the ability to override or provide feedback on AI recommendations.
  4. Adapt — Be agile and willing to change fast as your customers evolve. Continuously iterate on tools, workflows, and training. Treat CX transformation as an ongoing practice, not a project with a finish line.
Missed the live session? Watch the full fireside chat and download Adrian's three mini-guides to put these action steps into practice.

The human-AI partnership: Steps to create connected, effortless CX

Ready to get started? Explore Zoom CX or request a personalized demo to see how it all comes together for your team.

Our customers love us

Okta
Nasdaq
Rakuten
Logitech
Western Union
Autodesk
Dropbox
Okta
Nasdaq
Rakuten
Logitech
Western Union
Autodesk
Dropbox

Zoom - One Platform to Connect