Help Desk Chapter 1: Why and How to Migrate to HubSpot’s Help Desk Workspace
Help Desk Chapter 2: AI and Automation
You know the feeling. Your support team is drowning in repetitive tickets. Response times keep creeping up. Your best agents are burning out answering the same question for the hundredth time.
And that shiny “AI-powered support” your current tool promised? Turns out it’s about as useful as an umbrella made of tissue paper.
Here’s what’s different about Help Desk Workspace: the AI actually works. We’re talking 48% faster ticket resolution and 22% higher resolution rates. Real numbers, real teams, real results. See a real case study here.
The Foundation: How Help Desk Makes AI Work Better
Quick reminder: Help Desk tickets give AI the structure it needs to actually work. Consistent data in predictable places and full conversation context means smart suggestions instead of random guesses.
That same information also powers your automation and reporting—so you get AI intelligence and workflow efficiency working together.
The AI Features That’ll Change Your Day
1. Reply Recommendations: AI That Writes With You
What it does: AI analyzes thread context, relevant knowledge sources, and website pages and suggests complete, contextual responses. You see the suggestion right in the reply editor, ready to send or customize- all powered by Customer Agent.
Why it matters: Your agents aren’t starting from scratch or tab hopping to find answers. AI pulls the exact response your customer needs and hands it over as a ready to send draft. It’s like your best agent is sitting next to every rep, whispering the perfect answer for every ticket.
Pro tip: The quality of reply recommendations is directly tied to the content you give it access to. Make sure you have selected the knowledge sources, website pages, and blog posts you want referenced—those become the source material for AI suggestions. The better your content, the better your recommendations.
2. Ticket Summaries: Context at a Glance
What it does: Breeze Assistant automatically generates concise summaries of long ticket threads. One look, and you get the key points without scrolling through 47 back and forth messages.
Why it matters: No more scrolling through messages to figure out what’s happening. Manager reviewing performance? Team member covering for someone out sick? New rep getting up to speed? One click gives everyone the context they need—the customer’s issue, what’s been tried, what’s next.
Pro tip: Use ticket summaries for warm handoffs. When transferring a complex issue to a specialist or escalating to a manager, the summary ensures nothing gets lost in translation. Your customers stop repeating themselves, and your team stops asking redundant questions.
3. Support Macros: Automation That Complements AI
Quick sidebar: while we’re talking AI, let’s not forget about good old fashioned automation.
What it does: Support macros let you automate multiple ticket actions with a single click. Need to update the ticket status, change the owner, ticket category, or update the ticket priority? One macro handles it all.
Why it matters: AI handles the smart stuff (what to say, which article to share). Macros handle the fast stuff (routing, status updates, workflow steps). Together, they’re your efficiency dream team—intelligence meets speed.
Pro tip: Build macros for your most common workflows: escalation paths, status transitions, and team handoffs. Then let AI suggestions handle the customer facing communication. Your reps get the best of both worlds—smart content and fast execution.
FYI: This is in private beta. You can request access in your portal by clicking your profile in the top right corner > Product updates > Go to “All” > Search “ Support Macros”
4. Breeze Customer Agent: Welcome to Autonomous Support
What it does: Breeze Customer Agent handles complete customer interactions from start to finish. It reads incoming requests, pulls from your knowledge base, formulates responses, takes action, and only escalates to humans when it genuinely needs help.
Why it matters: This is where it all comes together. Customer Agent uses structured ticket data, pulls from your knowledge base, and works with the same context your team has. It’s not reading from a script—it’s an autonomous agent trained on your content, understanding your customers, working within your processes. On average, it resolves over 65% of conversations completely on its own (top teams hit 90%).
And here’s the best part: when Customer Agent can’t fully answer something, it flags knowledge gaps. Those gaps become your roadmap for what content to create or update next using Knowledge Base Agent.
Pro tip: Start with common, straightforward requests. Let Customer Agent own the repetitive questions your team could answer in their sleep—account status, basic troubleshooting, simple how-tos. Your team keeps the complex stuff. Build trust first, then expand coverage.
5. Knowledge Base Agent: Your AI Content Writer
What it does: Knowledge Base Agent automatically transforms successful support interactions into comprehensive knowledge base articles. It analyzes your ticket data, identifies gaps flagged by Customer Agent, and creates article drafts that you review and approve before publishing.
Why it matters: Here’s where the magic happens. Customer Agent surfaces knowledge gaps when it can’t fully answer questions. Knowledge Base Agent automatically steps in, analyzing those gaps alongside your resolved tickets to create comprehensive article drafts. You review, edit to match your brand voice, and publish. That new content immediately feeds back into Customer Agent and reply recommendations, making your entire AI system smarter with every article you add. It’s a continuous improvement loop that actually works.
The workflow: Customer Agent handles live support → flags knowledge gaps → KB Agent drafts articles to fill those gaps → your team reviews and publishes → improved content powers better AI responses. Your knowledge base grows smarter based on actual customer needs, not guesswork.
Pro tip: Don’t worry about having a perfect knowledge base before you start. KB Agent works best when you let Customer Agent run for a bit first—it’ll show you exactly which articles you need. Focus on filling the gaps Customer Agent identifies, and you’ll build a knowledge base that actually matches what your customers are asking for.
How It All Works Together: The Compounding Effect
Here’s where things get interesting. These features aren’t just a collection of tools—they’re a system that gets better the more you use it.
The continuous improvement loop: Customer Agent handles support and identifies knowledge gaps → Knowledge Base Agent analyzes those gaps and ticket data to draft comprehensive articles → your team reviews and publishes → improved content immediately powers better Customer Agent responses and reply recommendations. The system literally teaches itself what content it needs, then helps you create it.
The best part? Each piece makes the others better. Better knowledge content leads to better AI suggestions. More accurate AI means more autonomous resolutions and clearer gap identification.
It’s not just tools working together—it’s a system that gets smarter as you use it.
Making It Work for Your Team: Getting Started Right
Ready to put these AI features to work? Here’s the practical path forward:
Start with your foundation: You don’t need a perfect knowledge base to start—KB Agent works better when you let Customer Agent run first. Here’s the smart approach: enable Customer Agent on low-stakes channels, let it handle tickets and flag knowledge gaps, then use KB Agent to draft articles for those specific gaps. Customer Agent shows you exactly what content you need, KB Agent helps you create it, and you review everything before it goes live. This way, you’re building documentation that directly addresses real customer questions, not guessing at what might be useful.
Take the assisted to autonomous journey: Start using reply recommendations and ticket summaries. Let your team get comfortable with AI suggestions before introducing autonomous interactions. Once they trust the AI’s accuracy, roll out Customer Agent for low-stakes scenarios. As Customer Agent surfaces knowledge gaps, use KB Agent to address them—you’ll see AI performance improve in real time.
One thing to get right from the start: Review everything before you publish. KB Agent creates comprehensive drafts based on ticket data and knowledge gaps, but you’re the expert on your brand voice and specific product details. Think of KB Agent as a skilled first-draft writer—it does the heavy lifting of structuring content and pulling from resolved tickets, but your review ensures accuracy and maintains your standards.
The Bottom Line
Help Desk’s AI works because tickets provide a reliable structure alongside human context. That combination enables AI to understand what’s happening, learn from patterns, and deliver accurate help. Whether it’s suggesting replies to your team or autonomously resolving customer issues, AI has both the framework and the context it needs to be genuinely useful.
And here’s what makes it sustainable: the system identifies its own blind spots through knowledge gaps, then gives you the tools to fix them with Knowledge Base Agent. You’re not just using AI—you’re building a smarter support operation every single day.
Questions about implementation? Join us for our community AMA from November 17-21 where our product experts will be answering questions about these AI features.
Whether you’re wondering how to prioritize knowledge gaps, curious about KB Agent best practices, or trying to figure out which feature to tackle first, bring your questions. This is your chance to get direct guidance from the people who built these tools.
Questions about any of these AI features? Drop them in the comments, or save them for the AMA November 17-21 where our product experts can dive deep into your specific scenarios.