HubSpot Expands AI Offerings With New Agents to Help SMBs ‘Play Big’

An AI-generated image of a human agent using a chatbot application on a computer. Credit: Adobe Firefly

HubSpot has introduced new and updated AI-powered tools to help small and medium-sized businesses grow like their enterprise counterparts. Among the 200 features announced, the company unveiled four agents built on its Breeze AI platform: a customer assistant, a knowledge base tool, a prospecting assistant, and a content generator. While designed to help organizations scale their go-to-market efforts, these agents align with HubSpot’s mission to empower SMBs to leverage AI—without the hefty budgets and resources expended by larger firms.

HubSpot's Head of AI and Senior Vice President of Product, Nicholas Holland. Image credit: HubSpot
HubSpot’s Head of AI and Senior Vice President of Product, Nicholas Holland. Image credit: HubSpot

“SMBs are constantly struggling with, ‘I don’t have enough people. I don’t have enough money. I don’t have enough expertise,” Nicholas Holland, HubSpot’s head of AI and its senior vice president of product, explained in an interview. “We’re bringing AI to the table to help solve these issues in a way that…a large company would throw money or people at it. We want to bring this to something where it’s not the size of the money that determines success.”

The product releases HubSpot is making today leverage the foundation the company established nearly a year ago. At the time, it unveiled a litany of AI-powered features, including upgrades to its Service Hub and Content Hub. Holland remarked that the new agents continue HubSpot’s strategy to refine and improve AI tools for SMBs. He contended that the company has committed to a multi-year arc of making these agents necessary and valuable.

Making AI ‘Easy, Fast, and Unified’

But what exactly does HubSpot consider an agent? According to Holland, an agent is an AI system equipped with multiple capabilities, contextual awareness, and the ability to handle complex tasks beyond simple one-off interactions. He adds that there are different types, ranging from workflow agents to reasoning and custom chat agents. However, Holland calls a bot that handles tasks like generating an email subject line a “pre-made prompt that’s well-thought-out” and wouldn’t be an agent under HubSpot’s definition.

That being said, Holland stated HubSpot is pursuing AI with an “easy, fast, and unified” philosophy. Specifically, its customers want it “to be easy to bring into their organization. They want it to be fast to bring those use cases across a variety of the needs that they have, and then unified…they want to unify the unstructured, the structured, and the world data all together.”

It’s the basis for HubSpot’s Breeze platform, which launched in September 2024. With its suite of tools, businesses can transform themselves into AI-first organizations, leveraging the technology to work more efficiently and uncover new growth opportunities.

Subscribe to The AI Economy

Four New and Updated HubSpot Agents

That leads us to the four agents being introduced at HubSpot’s 2025 Spring Spotlight event. The first is a customer agent, a quick-to-set-up bot that works across multiple channels (e.g., email, website, WhatsApp, etc.) and answers customer questions. Holland told me the agent could be trained on any knowledge base, from the website to the help desk, local files, and more. He claimed that this AI assistant helped resolve over 50 percent of customer support tickets.

The next one is a knowledge-based agent. As its name implies, it helps businesses enhance their support resources in real time based on support tickets. Its mission is to close any knowledge gaps by looking at previous customer interactions around a subject, issue, or query and generating an answer. This knowledge base agent is powered by Frame AI, the conversation intelligence startup HubSpot acquired in December. Its technology helps identify unstructured data, parsing it from conversations, support tickets, emails, and calls.

Holland pointed out that the customer and knowledge base agents working together are examples of the multi-agent orchestration capabilities available through Breeze.

“So this is now where, on one platform, using the shared data and context, you got a customer agent working with all your customers, it figures out when it doesn’t know something, and then the knowledge agent looks at how all your representatives answer those questions, and then it fills that knowledge gap,” he said.

The final two agents are not new to HubSpot but rather upgrades. They start with the content agent, one of the first Breeze bots introduced in 2024 that helps organizations generate case studies, blog posts, landing pages, and websites. HubSpot has infused the next-generation agent with a “deeper understanding of people’s brand and identity.” Additionally, the AI assistant leverages the content graph, enabling it to automatically insert deep links when a new blog post is published—drawing from its understanding of all previously published content.

The last one is a prospecting agent. Similar to the content agent, this one debuted in September. It gives businesses a sales tool to research potential customers, identify decision-makers, score leads, and develop better outreach strategies. The latest update enables it to provide deeper and “more robust research” on customer accounts. This means probing data across different sources, producing an entire account plan for sales representatives, and providing more personalized outreach.

Holland pointed out that the customer agent is generally available, but the other three are in public betas.

Helping SMBs Avoid the ‘Cobble Tax’

From a feature standpoint, it’s clear that HubSpot is in direct competition with Salesforce, Zendesk, Intercom, and others. However, that’s not the reality. HubSpot isn’t targeting the enterprise—it’s going after small and medium-sized businesses. “HubSpot’s kind of mission is to help SMBs grow by giving them everything that they need to…attract, sell to, and delight customers,” Holland asserted.

He later added, “We want to be the champion for the SMB. From an AI standpoint, we’re building an AI-first customer platform that really makes small teams mighty. It’s really helping the SMBs play big. And that’s a key part of what drives us.”

Organizations that want to execute programs similar to their enterprise counterparts must use multiple applications rather than a single best-of-breed platform. Holland references this as a “cobble tax,” a penalty SMBs face for having to authenticate across different tools, needing to teach employees complex procedures to navigate different systems, and not having unified data and context. He believed companies should be on this journey “from novel to necessary,” and having data “off stack” doesn’t fall into these categories.

HubSpot’s core message is that it simplifies AI adoption, making it seamless and accessible for SMBs by offering an integrated, user-friendly solution that doesn’t require technical expertise. It’s similar to how large enterprises might rely on Microsoft Project for complex project management, while smaller firms prefer the simplicity and flexibility of tools like Trello.

Other Spring Showcase Announcements

In addition to the four HubSpot agents, the company unveiled updates to its Marketing Hub Enterprise, which helps marketing teams identify leads and close prospects. Now, it supports lookalike lists, journey automation, and multi-account management.

HubSpot also introduced three new Workspaces, dedicated areas within the platform featuring tools and workflows tailored to a particular role within a business. There are now workspaces for sales, customer success, and support (help desk).

“SMBs don’t need more AI hype—they need technology that helps,” HubSpot’s Executive Vice President of Product, Andy Pitre, said in a statement. “The products we’re launching at the Spring 2025 Spotlight are helping teams move fast on AI and solve their go-to-market challenges.”

Featured Image: An AI-generated image of a human agent using a chatbot application on a computer. Credit: Adobe Firefly

Leave a Reply

Discover more from Ken Yeung

Subscribe now to keep reading and get access to the full archive.

Continue reading