Over the past two years, HubSpot has undergone one of the most significant transformations in its history. The rise of artificial intelligence in mainstream business applications has pushed every software provider to rethink how technology fits into daily work. HubSpot has responded quickly, moving from incremental CRM improvements to positioning itself as a platform for human/AI hybrid teams. The company has invested heavily in conversational AI, data enrichment, and intelligent automation, while also redefining marketing strategy for an era when buyers often interact first with AI assistants rather than with websites or sales representatives.
The scale of this year’s INBOUND event reflected the magnitude of that shift. For the first time, HubSpot hosted the conference in San Francisco, a deliberate nod to its position at the center of the global AI conversation. And placed it provocatively just two blocks away from Salesforce’s headquarters. The event drew more than 12,000 customers, partners, and industry analysts.
There was some irony this year in the title of the event. At an event called INBOUND, much of the messaging focused on how the traditional inbound marketing model has broken down. For more than a decade, inbound marketing has been synonymous with drawing prospects through search, blogs, social, email and web content. That playbook enabled a steady flow of inbound clicks. HubSpot leaders acknowledged at the event the disruption that AI has brought to the inbound marketing model they helped pioneer and responded with a new framework, a growing ecosystem of AI partnerships, and a steady expansion of its features for the new AI-centric world.
HubSpot emphasized that AI-driven summaries and answer engines are intercepting traffic before it reaches company websites. Prospects are increasingly getting answers directly from AI assistants, often without clicking through to the original content. The reduction in clicks is dramatic, and it undermines the foundation of traditional inbound campaigns.
AI is also drawing information from a wide variety of sources beyond company-controlled channels. Large language models frequently surface content from Reddit, industry forums, analysts, and influencers, placing as much weight on third-party voices as on corporate websites or official social media accounts.
The implication is clear: marketers must adapt to a new environment where visibility depends on being represented accurately inside AI-generated responses, not just in search results or owned properties. Success will require a broader approach, ensuring the brand is present and credible across the ecosystem of sources that AI consults.
While the announcements at INBOUND highlight HubSpot’s vision for the platform, it is important to recognize that the capabilities discussed here will not all appear immediately. HubSpot introduces new features gradually. They often begin in private beta, available only to selected customers for early testing. From there, they progress to public beta, where a broader audience can use them with the understanding that changes may still occur. Only after this cycle do they reach general availability, meaning that many of this year’s announcements will take months before they are fully released. Even then, adoption typically trails availability. Most organizations implement only a portion of the functionality at first, often due to training requirements or competing priorities. The result is that the platform’s full potential is usually realized over time rather than immediately after launch.
The expanding range of AI capabilities in HubSpot offers clear potential but also creates some hesitation. Many customers will find it difficult to know which AI assistants or agents to focus on first, given the wide range of options. Concerns about losing control are equally real, since these tools can act with autonomy in drafting communications, enriching data, or handling service requests. As a result, most organizations are likely to begin cautiously, piloting a small number of agents and assistants before expanding their use more broadly.
Overall, INBOUND 2025 reinforced HubSpot’s ability to adapt quickly to a changing landscape. The event demonstrated not only product innovation but also HubSpot’s willingness to break and replace the models that have been their foundation.
- The Loop: HubSpot’s New Growth Framework
At the center of this year’s announcements was the unveiling of The Loop, a growth playbook designed to replace traditional funnel and flywheel models that no longer reflect modern buyer behavior. HubSpot framed The Loop as both a philosophy and a practical set of tools to help companies adapt to AI-driven, non-linear customer journeys.
The framework is structured around four stages:
- Express: This step emphasizes establishing a clear and consistent identity. Marketers are encouraged to define brand tone, values, and positioning in a way that translates across both human and AI-mediated channels. The goal is to ensure that whether a customer reads a blog post, interacts with a chatbot, or sees the brand mentioned in an AI-generated answer, the company is represented with the same language, credibility, and intent.
- Tailor: Rather than relying on broad segments such as industry or company size, this stage focuses on aligning messages with individual buyer context. Using CRM data enriched by AI, marketers can adjust content, offers, and outreach based on real-time behavior and intent signals. This creates a level of personalization that moves beyond static audience definitions and responds to how buyers actually engage.
- Amplify: Visibility is no longer limited to search engines and owned social channels. In the current landscape, AI systems pull answers from a wide range of sources, including forums, review sites, and influencers. Amplify means ensuring that brand messaging is available and credible wherever prospects may encounter it. HubSpot’s new Answer Engine Optimization (AEO) tools are designed to help brands structure content so it appears in AI-generated summaries and responses, extending reach beyond traditional inbound clicks.
- Evolve: Campaigns can no longer be “set and forget.” This stage requires continuous monitoring of performance data (engagement, attribution, and outcomes) and using AI to surface insights that inform quick adjustments. Marketers are expected to treat campaigns as living systems, refining messages, channels, and tactics in response to feedback rather than waiting for the next planning cycle.
The Loop reflects a broader shift away from linear campaign planning toward continuous adaptation, with content, insights, and revenue operations feeding back into each other in real time.
The underlying message here is that inbound clicks can no longer be relied upon as the primary engine of growth. Instead, marketers must broaden their strategy to include AI as both a channel and a gatekeeper, ensuring the brand remains discoverable, credible, and consistent wherever prospects encounter information.
Putting The Loop into practice will require a shift in execution as well as mindset. Marketers will need to restructure content so it can be easily parsed by AI systems, with clear answers to common questions, concise summaries, and supporting schema markup. They will also need to diversify distribution, ensuring that brand messaging is present not just on owned websites, but in the wider ecosystem where AI pulls information: forums, review sites, analyst reports, and influencer commentary. Personalization will play a central role, with marketers using CRM data and AI tools to tailor outreach by vertical, role, and intent. Finally, The Loop calls for continuous iteration, where campaign performance is monitored in real time, insights are surfaced by tools like Breeze, and strategies are adjusted without waiting for quarterly reviews.
In practice, this means moving away from set-piece campaigns and toward a living system of content, data, and engagement that evolves alongside customer behavior. The Loop is less about one-time execution and more about building the organizational discipline to adapt constantly, guided by AI-enhanced insights.
- Marketing Hub Innovations
To support The Loop strategy, Marketing Hub introduced several beta features:
- Visual Campaign Canvas: A collaborative, AI-augmented environment for planning campaigns, mapping touchpoints, and coordinating activities across teams.
- Segments + Personalization: Enhanced targeting capabilities that adapt CTAs, landing pages, and content blocks dynamically based on real-time CRM data.
- AI-Powered Email: Enables generation of personalized, one-to-one messages at scale, with predictive send-time optimization and engagement scoring.
Together, these features illustrate how HubSpot is repositioning marketing execution around adaptability and precision. The Loop requires marketers to treat every interaction as part of a continuous cycle, not a one-off campaign. The visual campaign canvas ensures that activities are planned with this cycle in mind, showing how touchpoints connect and where gaps may exist. Dynamic segmentation ensures that as customer data shifts, the messaging automatically adjusts without requiring teams to rebuild entire campaigns. AI-powered email closes the loop by delivering one-to-one outreach at scale, aligned to both the individual’s behavior and the broader campaign strategy.
- Launch of the Data Hub
HubSpot introduced Data Hub, formerly known as Operations Hub, as a new foundation for managing customer information across the platform. It is intended to reduce fragmentation by pulling structured and unstructured data into one repository. HubSpot also announced that data enrichment is now free and will no longer consume Breeze credits. This will compete with third-party enrichment services from the likes of ZoomInfo and Apollo.
Key capabilities include:
- Automated Data Quality Management: Deduplication, standardization, and enrichment to keep records clean.
- Smart Trend Detection: AI-driven recognition of behavioral or market shifts within customer data.
- Cross-System Integration: Designed to connect external sources without heavy technical work, creating a single view of the customer.
- Support for Advanced Reporting: Improved segmentation and analytics, especially for organizations adopting multi-touch attribution models.
This signals HubSpot’s intent to compete more directly with enterprise-grade data management and customer intelligence solutions.
- Expansion of Breeze Agents and Assistants
HubSpot significantly broadened its AI agent ecosystem, positioning these as digital team members that handle both routine and complex tasks.
- 18 New Breeze Agents, joining the 15 already launched. Highlights include:
- Customer Agent, designed to resolve a majority of common service tickets without human intervention.
- Prospecting Agent, which automates outreach and qualification, using CRM data and external sources.
- Data Agent, capable of answering queries based on CRM and third-party datasets.
- Breeze Assistant, a general-purpose, context-aware helper that adapts to user behavior, making it more personalized over time.
- Custom Assistants, trainable with proprietary documentation and workflows, ensuring companies can embed their own processes directly into HubSpot’s AI ecosystem.
- Breeze Studio and Marketplace
Two related platforms were launched in public beta:
- Breeze Studio: Provides tools to design, train, and customize AI agents with company-specific knowledge and rules.
- Breeze Marketplace: A catalogue where businesses can browse and deploy pre-built agents and assistants, accelerating adoption without requiring technical development.
Together, these tools position HubSpot as not only a CRM provider but also a platform for building and managing AI-powered digital employees.
- AI-Powered CPQ in Commerce Hub
HubSpot unveiled an AI-driven Configure-Price-Quote system, reinforcing its investment in B2B commerce:
- Smart Quote Generation: Automated proposals, branded templates, and drag-and-drop customization.
- Buyer Engagement Tracking: Notifications when prospects open, view, or share proposals.
- Closing Agent: An AI assistant that helps buyers answer pricing or product configuration questions.
- Flexible Approvals: Workflow support for multi-level pricing approvals and structured product bundles.
This places HubSpot directly in competition with dedicated CPQ vendors, while keeping CPQ fully integrated within its CRM.
- Sales Hub Enhancements
Several new features in Sales Hub were showcased:
- Meeting Assistant: Provides reps with relevant customer context before meetings, generates real-time insights, and sends personalized follow-up summaries.
- Deal Risk Detection: Uses transcript analysis from sales calls to flag stalled or at-risk opportunities.
- Pipeline Intelligence: Deeper integration of AI-driven scoring and prioritization, giving managers clearer views of forecast health.
- Integration with Google Gemini
HubSpot announced a Gemini CRM Connector, complementing existing integrations with ChatGPT and Claude. This allows data stored in HubSpot to be accessed by Gemini for analysis or content generation, with results written back into HubSpot records. The broader message was openness to multiple large language models, giving businesses flexibility.
- Strategic Vision: Human-AI Hybrid Teams
A recurring theme throughout INBOUND 2025 was HubSpot’s belief in hybrid teams, where AI supports rather than replaces people. HubSpot emphasized:
- AI’s role in removing repetitive work while humans retain responsibility for authenticity, creativity, and trust.
- A shift away from vanity metrics like clicks and toward meaningful engagement outcomes.
- The importance of transparency in AI outputs, with businesses maintaining control over how AI agents represent their brand.
Summary Table
| Area | Key Announcement |
| Strategic Framework | The Loop playbook for continuous, AI-driven growth |
| Marketing | Campaign canvas, dynamic personalization, AI-generated email |
| Data Infrastructure | Launch of the Data Hub for clean, unified, actionable data |
| AI Agents | 18 new Breeze Agents, expanded Assistants, custom training |
| AI Development Tools | Breeze Studio for custom agents, Breeze Marketplace for adoption |
| Sales Enablement | AI Meeting Assistant, deal risk detection, pipeline insights |
| Commerce | AI-powered CPQ with quotes, tracking, and closing support |
| Integrations | New Gemini connector, alongside ChatGPT and Claude |
| Vision | Human-AI hybrid approach focused on authenticity and trust |