If you are still using a generic chatbot that says “Hi there! How can I help?” on every single page of your website, you are leaving money on the table.
Customers today expect more. They want answers relevant to what they are currently reviewing. They don’t want to explain themselves to a robot that doesn’t understand context.
HubSpot just rolled out a major update that fixes this exact problem.
They have overhauled their Customer Agent with “Prompt Recommendations” and opened the door to “Agentic AI” with the Model Context Protocol (MCP).
This isn’t just a technical update. It is a fundamental shift in how you will automate your sales and marketing.
Let’s break down what this means for your business and how you can use it to optimise your revenue and margins.
Key Takeaways
- Context is King: New “Prompt Recommendations” let your AI agents start conversations based on the specific page a visitor is viewing, boosting engagement.
- Agentic AI is Here: The Model Context Protocol (MCP) enables you to connect HubSpot data with external AI tools such as Claude, enabling autonomous workflows.
- Higher Conversion: Personalised, context-aware interactions lead to better lead capture and higher conversion rates.
- Operational Efficiency: By allowing external AI to read and act on CRM data, you reduce manual data entry and streamline complex processes.
- Developer Freedom: Your development team can now build custom agents that operate outside the native HubSpot interface while still syncing seamlessly with your CRM.
The End of the “One-Size-Fits-All” Bot
We have all seen them. The chatbots that pop up with the same generic greeting, whether you are reading a blog post about SEO or staring at a pricing page.
It is lazy marketing.
HubSpot’s new Prompt Recommendations feature changes the game.
Instead of a static greeting, the AI agent now analyses the webpage the visitor is viewing and suggests context-aware conversation starters.
Think about the difference this makes.
If a prospect is on your pricing page, the bot shouldn’t ask, “Do you want to sign up for our newsletter?” It should ask, “Do you have questions about the Enterprise plan features?”
If they are reading a technical blog post, the bot should offer a related case study or white paper.
This relevance signals to the user that you understand their intent. It builds trust instantly.
When you align your engagement strategy with user intent, you see a direct lift in engagement rates. More engagement means more leads. More leads mean more potential revenue.
Opening the Door to “Agentic AI”
The second major update is the public beta of the Model Context Protocol (MCP) Server.
This may sound technical, but the business implication is significant.
Historically, your CRM data was trapped inside your CRM. If you wanted to use an external AI tool to analyse that data or take action on it, you had to build clunky integrations or manually move data back and forth.
MCP changes that. It acts like a universal translator or a standard plug (think USB-C) for AI tools.
It allows developers to securely connect HubSpot data with external AI tools and workflows, such as Anthropic’s Claude desktop app.
This enables what we call Agentic AI.
Traditional AI answers questions. Agentic AI does things.
With MCP, you can build custom AI agents that can read your HubSpot CRM data and take actions outside of the native interface.
Imagine an AI agent that monitors your deal pipeline. When a deal stalls, it doesn’t just notify you. It automatically drafts a re-engagement email based on the prospect’s last interaction, updates the deal stage in HubSpot, and schedules a follow-up task for your sales rep.
That is the power of Agentic AI. It turns your CRM from a database into an active participant in your business processes.
How This Optimises Your Business
You might be wondering, “How does this actually help my bottom line?”
Here is the breakdown:
1. Increased Revenue through Better Conversion
The Prompt Recommendations feature directly addresses the friction at the top of your funnel. By making that first interaction relevant, you hook the visitor.
If you can increase your visitor-to-lead conversion rate by 1%-2%, the revenue impact over a year is significant. You aren’t spending more on ads to get traffic; you are just getting more value from the traffic you already have.
2. Improved Margins through Automation
Agentic AI and the MCP server enable you to automate complex, multi-step workflows that previously required human intervention.
Every minute your sales team spends manually entering data or drafting generic follow-up emails is a minute they aren’t selling. By offloading these tasks to AI agents that can act autonomously across systems, you reduce your operational costs.
Better margins come from doing more with the same resources. This technology allows you to scale your output without linear increases in headcount.
3. Smarter Data Utilisation
Data silos kill efficiency. When your AI tools can’t talk to your CRM, you get fragmented insights.
MCP breaks down these walls. It allows your external AI models to access the rich context stored in HubSpot. This means your AI generates better insights, more accurate forecasts, and more personalised content because it has the full picture.
What You Should Do Now
These updates are powerful, but only if you use them.
Here is your action plan:
- Audit Your Chatbots: Look at your high-traffic pages. Are your bots sending generic greetings? Switch to context-aware prompts immediately. Test different hooks for your pricing, product, and blog pages.
- Talk to Your Developers: If you have a dev team, ask them about the Model Context Protocol. Challenge them to find one manual workflow that involves data outside of HubSpot and see if an AI agent can now handle it.
- Review Your Customer Journey: Map out where users drop off. Is it because they can’t find answers quickly? Use the new AI capabilities to plug those holes with relevant, instant information.
Conclusion
The future of marketing isn’t just about having AI; it is about having AI that understands context and can take action.
HubSpot’s move toward Agentic AI and context-aware interactions is a clear signal that the industry is shifting from static automation to dynamic, intelligent engagement.
You have the tools. Now you need to execute.
Start small. Optimise your high-value pages first. Experiment with custom agents for your internal workflows.
The businesses that adopt these tools early will build a significant efficiency advantage. Those that don’t will continue to pay for manual work that their competitors have already automated.
