How Zendesk AI can start supporting your teams and customers today

Whatever tools we use to make our lives easier, more productive, or more profitable, the message coming out of Relate 2023 was clear: success depends entirely on whether your technology actually helps humans get the job done.

Zendesk AI is the CX professional’s latest and greatest toolkit, designed with the support experience top of mind. It’s tailor-made for agents navigating lengthy ticket queues and chat windows, emotional days (we all have them), and more.

An AI solution made for CX teams is just the beginning. But event attendees couldn’t wait to learn more, flooding the product lounge and demo stations. In the days since, the Zendesk Community has been busy with questions about unique use cases and more advice users could implement today. While we encourage everyone who couldn’t attend to experience the full sessions and breakouts on demand at your leisure, here are a few tidbits we pulled to share with your team.

Intelligence panel: The “home” of agent productivity

Hot off the keynote, Zendesk’s head of AI Cristina Fonseca continued the momentum in “Harness the possibilities of AI for service,” the first breakout session in the product lounge.

The Intelligence panel is the center of the agent experience, enabling teams to solve issues faster with AI-powered insights and recommendations. Boasting capabilities to gauge intent, language and sentiment, provide a summary of conversation, and make suggestions for macros, it represents the future of AI-assisted customer service.

“We see the Intelligence panel as the home of our agent productivity suite,” Fonseca said, adding that we can expect to see new features added to the panel within the next couple of months.

Through our partnership with OpenAI, generative AI helps agents and customers arrive at a resolution faster. By generating support responses, agents can rest easier in those high-stress, high-stakes moments. Agents can also lean on Tone shift and Sentiment analysis to ensure they get a second pair of “eyes” to spot language that could land poorly or signal frustration.

Advanced bots have advanced a lot

While we all have stories about getting frustrated with a customer-facing bot as it misunderstood our question, our advanced bots demonstrate how far bots have come. Pre-trained in the top topics for specific industries and editable for customization, the bots automatically understand what the customer is asking for and how they feel. The bot is enabled to deliver the perfect answer in each moment and to escalate to a live agent when it doesn’t know.

If an issue is escalated, the bot can tag customer requests by intent to increase response accuracy and help agents resolve issues faster. And they can tackle high-volume requests: the repetitive questions that come in droves and can often be managed via self-service.

Speaker Ian Hunt, Director of Operations and Customer Services at UK retailer Liberty London, knows high-volume. He joined Zendesk product expert Claire Peracchio to discuss how intelligent triage is already showing results during spikes and volatile periods, contributing to a 73 percent decrease in first reply time. At the helm of his organization for “calendarmaggedon”–the playful name Hunt and team call the frenzy over Liberty’s annual beauty product advent calendar–he expects AI to be a big help in this upcoming high season.

He reminded the crowd that while an agent has heard the same question 20 times that day, the customer is asking for the first time, and it’s the most important thing to them at that moment. But even the most empathetic agent can use an AI assist to keep those requests managed promptly and their queues freer for more complex issues.

One AI solution does not fit all

Zendesk AI tools for CX are adaptable for different companies’ needs and comfort levels with automations. On one end of the spectrum, a company can leverage a full AI-supported operation today, enabling intelligent triage, macros, and bots at the same time. On the other, as one attendee explained, they may want to start small with one bot that suggests responses for a live agent.

As Fonseca said: “Automation is a journey.” And we’re just getting started.