A practical guide to HubSpot's Breeze Agent Behavior Settings

AI agents are popping up everywhere, and if you're in the HubSpot world, you've probably heard of Breeze. It promises to be a helpful assistant, handling everything from customer support chats to sales prospecting, all inside an ecosystem you already use.

But here’s the thing: its real-world usefulness boils down to how you tweak its "behavior settings." Get it right, and you’ve got a solid automated teammate. Get it wrong, and you’re stuck with a frustrating, and surprisingly expensive, robot.

This guide will walk you through what HubSpot's Breeze agents are and how to set them up. More importantly, we'll take an honest look at the limitations and hidden costs that come with customizing them and show you a more flexible alternative for teams who need to be in the driver’s seat.

What are HubSpot Breeze agents and their Breeze Agent Behavior Settings?

First off, let's get the names straight. "Breeze" is HubSpot's umbrella term for all its AI features. "Breeze Agents" are the specific tools that handle workflows on their own. Think of them as AI-powered team members working behind the scenes in marketing, sales, or support. You’ll see a few common ones, like the Customer Agent for support, the Knowledge Base Agent for writing help articles, and the Prospecting Agent for outreach.

So, what exactly are Breeze Agent Behavior Settings? It’s not a single dashboard with a few toggles. It’s the sum of all the instructions you give the agent during setup. This includes the knowledge sources you hook it up to, the "role" and "personality" you assign it, and the rules you set for when it needs to call a human for help. All these pieces together shape how the agent thinks and responds, which makes the setup process the most important part of getting any real value out of it.

How to configure Breeze Agent Behavior Settings: A step-by-step guide

Setting up a Breeze agent has a few distinct stages. Even though it's built right into the HubSpot interface, the process can feel a little rigid and has some requirements that might catch you by surprise. Here’s how it usually goes.

Step 1: Prerequisites for Breeze Agent Behavior Settings

Right off the bat, you'll run into the first big catch: you need a HubSpot Professional or Enterprise plan. With Professional plans starting at around $450 a month, this immediately puts Breeze out of reach for a lot of smaller businesses. You’ll also need to be an admin with the right permissions and have the HubSpot tracking code on your website.

Step 2: Create the agent and define its personality

Once you’re on the right plan, you can head over to the Breeze AI menu to build your agent. The first thing it asks for is a "Role" and a "Personality" (Friendly, Professional, Witty, and so on). These are the first, most basic layers of control you have over the agent’s behavior, setting its general tone and purpose.

Step 3: Picking content sources (this is the big one)

This is easily the most critical step because it’s where you build the agent’s "brain." How the agent behaves is almost entirely based on the information it has access to. You can connect it to your HubSpot Knowledge Base, certain website pages, or even uploaded files.

But there’s a catch. The quality of your agent’s answers is completely dependent on your content being clean, well-structured, and always up-to-date. Some early users have found that getting good results means spending a ton of time rewriting and reformatting existing help docs just to make them work for the AI.

Step 4: Setting up handoff rules

Next, you have to decide what happens when the agent gets stuck. Your handoff rules determine when and how a conversation gets passed to a person. You can route the chat to a specific team, leave it in an unassigned inbox for anyone to grab, or have it automatically create a support ticket. This is a vital behavior setting that defines the agent’s limits and hopefully prevents a customer from having a bad time when it can’t find an answer.

Step 5: Test and go live

Finally, HubSpot gives you a preview tool to test the agent's answers before you unleash it on your customers. This step is a must, but it often just shines a light on all the gaps in your knowledge base, which sends you right back to Step 3. It’s a repetitive loop that can take a lot longer than you might think to get right.

The hidden costs and limitations of Breeze Agent Behavior Settings

While having an AI agent built into HubSpot sounds convenient, a closer look shows some pretty big limitations and costs that can make it tough to work with, especially for teams that need control and predictable spending.

Limitation 1: Pricey plans and a confusing credit system

The first hurdle is the cost of entry. As we mentioned, Breeze is only available on the Professional or Enterprise plans. If you're upgrading just for this, you'll also get hit with a mandatory $1,500 onboarding fee.

But the real headache starts with the usage-based pricing. You get a tiny allowance of 30-50 free conversations a month. After that, you pay with credits, which comes out to about $1 per conversation. This pay-per-resolution model means your bill climbs as your support volume grows, creating unpredictable costs that punish you for growing your business. A few hundred chats could easily tack on hundreds of extra dollars to your monthly bill.

In contrast, platforms like eesel AI offer straightforward pricing. With flat-rate monthly plans that don't charge you for every ticket or resolution, you know exactly what your costs will be and can scale your support without dreading the end-of-month invoice.

The eesel AI pricing page, showing transparent, flat-rate plans as an alternative to usage-based models.

Limitation 2: You can't see what's going on under the hood

Beyond picking content sources and a basic personality, you have almost no direct control over the AI’s logic. You can't write custom instructions for specific situations, fine-tune the prompts it uses, or set up complex business rules. The agent is a bit of a "black box" that just goes by the content you give it, which is why some people have reported that the quality of its responses is often "mostly poor."

This is where a dedicated AI platform really shines. For instance, eesel AI gives you a powerful workflow engine you can manage entirely on your own. Its prompt editor lets you define the AI's persona, tone, and reasoning with total control. You can also build custom actions that let the agent do things like look up order details in Shopify or update ticket information in Zendesk. Those are things Breeze just can't do without a lot of complex, custom work.

The eesel AI interface, where users can set up custom rules and fine-tune the AI's logic, unlike the limited Breeze Agent Behavior Settings.

Limitation 3: A long and painful setup process

Getting good results from Breeze isn’t something you can do in an afternoon. HubSpot’s own case studies admit it can take weeks of tweaking your knowledge base to get the agent working reliably. The AI’s behavior is reactive, meaning you're always playing catch-up, fixing its mistakes by adjusting your content instead of just telling it how to act in the first place.

This is a huge difference with eesel AI, which is built to get you up and running in minutes, not months. The whole approach is designed for speed and confidence:

• One-click helpdesk connection lets the AI learn from thousands of your past tickets in an instant. It gets your brand voice and common solutions right from day one.

• A powerful simulation mode lets you test the AI on your real, historical tickets before it ever talks to a customer. You can see exactly how it would have answered, giving you a clear picture of its performance and ROI.

• It connects to all your knowledge sources, like Confluence, Google Docs, and Slack, without making you manually reformat everything.

The eesel AI simulation mode, which allows users to test the AI on historical data before deployment, overcoming the limitations of Breeze Agent Behavior Settings.

eesel AI: A better alternative to HubSpot's Breeze Agent Behavior Settings

HubSpot's Breeze AI offers a clear trade-off: you get the convenience of native integration, but you pay for it with high costs, limited control, and a slow, reactive setup. For teams that care about speed, flexibility, and a predictable budget, that's a tough pill to swallow.

eesel AI was built to solve these exact frustrations. It’s a specialist tool that gives you complete control over your AI agent’s behavior without the downsides of a closed system.

Here’s how eesel AI puts you back in control:

• Truly Self-Serve: Forget about mandatory sales calls or pricey onboarding fees. You can get everything set up in minutes, all by yourself.

• Total Control: Use a visual prompt editor and custom actions to define exactly how your AI behaves, what it can do, and when it should pass a conversation to a human.

• Test with Confidence: The unique simulation mode lets you see how the agent will perform on your real historical tickets. It takes the guesswork out of the process and gives you a clear idea of its impact before you launch.

• Unify All Your Knowledge: Instantly connect to your help desks, internal wikis, and chat tools. eesel AI works with your knowledge right where it lives.

• Predictable Pricing: Our flat-rate plans mean no surprise bills. You can scale your support operations without watching your costs balloon.

Final thoughts: Are HubSpot's Breeze Agent Behavior Settings the right choice for you?

So, when does HubSpot Breeze actually make sense? It can be a decent starting point if your company is already all-in on the HubSpot ecosystem (on a Professional or Enterprise plan) and your support needs are fairly simple and low-volume. If you just need to answer a few basic FAQs and the pricing model doesn't scare you, the native integration is a nice perk.

However, for most teams that need fine-grained control over their Breeze Agent Behavior Settings, a budget-friendly solution that can grow with them, and a setup that takes minutes instead of weeks, the limitations of Breeze become a major roadblock. In those cases, a dedicated, specialist platform will almost always give you more power, flexibility, and a much better return on your investment.

Take back control of your support automation

Stop wrestling with rigid, expensive AI tools that tell you how to work. See for yourself how easy it is to build, test, and manage a powerful AI support agent with eesel AI. You can simulate its performance on your own data and be live in minutes.

Start your free trial or book a demo today.