AI, HubSpot Implementation, HubSpot Consulting

Unlocking the Power of HubSpot's INBOUND 25 AI for Your Business

Setting the Scene

HubSpot’s INBOUND 25 announcements have been described as the most transformational yet. AI assistants, AI agents, smarter data handling, and even AI-powered CPQ took centre stage. At Avidly, our role isn’t to repeat the headlines – it’s to help clients separate signal from noise.
We both know these tools won’t transform your business on their own. They only create impact when set up properly, trained with the right context, and fuelled by accurate data.

Here’s a rundown of what’s been released with some commentary alongside.

INBOUND 25 Quick Catch-Up: Product and Pricing

HubSpot CRM is now a Standalone Product: You previously got access to the HubSpot CRM tools by purchasing any Hub (Marketing, Content, Service, Commerce, Sales). Now, new customers can purchase Smart CRM Professional or Enterprise on its own, on a seat-per user basis.
Breeze Assistant (formerly ‘Copilot’) is now more durable and contextual: The (putting it politely) rather limited Breeze Copilot is now relaunched as Breeze Assistant. It’s more durable and better connected, described by HubSpot’s Head of AI as, “Chat GPT with your HubSpot Context.” It can now be your actual helpful assistant and can start doing tasks too - without needing to hop over into ChatGPT externally.
Breeze Intelligence Pricing Updates: Previously, creating or enriching a Company, Contact or Property record was 10 Credits each time, respectively. But now,
- Creating or Enriching a Company or Contact record is now included as part of being in a paid Core Seat
- Enriching a single Property on request is now part of Data Agent using 10 credits per request
- Enabling intent signals and monitoring is 10 credits per month per company.
- Enabling Breeze actions in workflows costs 10 credits per action.
Breeze Prospecting Agent Pricing & Feature Updates: Enabling monthly monitoring of one contact for outreaching opportunities now costs 100 credits. Generating results from deep research on a company by the Agent costs 10 credits. In terms of functionality, this Agent can now autonomously update the CRM based on your notes, monitor for hiring/investment signals at companies and –very excitingly– hand-off to other Agents inside HubSpot.
Breeze Data Agent [Beta] Pricing Updates: Generating responses to one prompt for one record using this Agent comes at a cost of 10 credits. This new Agent “operates a bit like Clay” and can pull customer pain points from conversations, sentiment analysis, hiring status, location changes and more… All useful use cases that start making the most of your unstructured data inside HubSpot.
Operations Hub has become Data Hub: Data Hub now also includes new features such as the flagship ‘Data Studio’ (which consumes Breeze credits, by the way). Existing Ops Hub customers will be migrated during September 2025. Purchasing Data Hub Pro increases your included credits to 5,000 and Enterprise to 10,000.

Round-up of New Features Just Released

Besides the above headlines, there are over 100 Breeze feature updates embedded around HubSpot, each designed to save clicks and brainpower in micro-instances (for example, I, for one, am never drafting a meta description again). But there are another hundred-or-so bona fide features, updates and betas. Here are a few:

Flexible CRM Views: Display the same data in a table, board, report, and calendar style - one source of truth, many perspectives to best fit your needs.
Self-Generating CRM Data: Automatically enriches contacts and companies from calls, emails, and external sources.
Smart CRM Insights: Surfaces changes, risks, and next-best actions directly in the CRM, making it easier to be prepared for sales calls.
Data Studio: Creates unified, AI-ready datasets from scattered sources.
Improved Customer Agent: AI that resolves tickets, qualifies leads, and books meetings.
Data Agent: AI researcher that turns CRM and external data into actionable insights.
Breeze Studio & Marketplace: Build, configure, and deploy tailored AI agents.
Segments & Personalisation: Smarter targeting and customised campaigns.
Marketing Studio: AI-powered drag-and-drop campaign builder.
AI-Powered CPQ: Auto-generates quotes and proposals – promising but narrow in scope without customisation.
AEO Strategy Feature: Optimise for AI-driven search visibility.

Why Did HubSpot Release These Features and Where is the Platform Heading?

Alongside the official announcements, HubSpot’s Head of AI Nicholas Holland gave a refreshingly candid rundown on Reddit — explaining how Breeze Assistant, Prospecting Agent, and Data Agent will actually work inside HubSpot day-to-day. Filling in some details between the lines of the Press Release materials circulating.
But more than that, he spent 40 minutes explaining to us why HubSpot has doubled down on Agents and contextual understanding with these updates and exploring what our day-to-day is going to look like working inside HubSpot in the (not to distant) future.
Check out this candid discussion and get an INBOUND 25 Cheat Sheet here…

What’s Exciting – and What Still Needs Work

Prospecting Agent, Customer Agent and Data Agent — and they can talk to each other

These are the most immediately useful of the new releases. They’re designed to take routine work off your team’s plate, whether that’s answering tickets, qualifying leads, or researching prospects.

What makes them especially exciting is the way they can hand off between each other. For example:

The Prospecting Agent can research and build account plans, personalise outreach, and even monitor signals like hiring or fundraising. If a contact engages, it can then hand off to the Customer Agent.
The Customer Agent handles tier-one interactions autonomously – resolving repetitive queries or qualifying a lead before escalation.
The Data Agent enriches everything in the background – pulling pain points from call transcripts, scanning sentiment, or surfacing firmographic changes from the open web.

Together, they create a multi-agent workflow: prospecting, service, and enrichment agents all reinforcing each other. That’s where the clearest short-term ROI lies… but only if adoption is structured and data foundations are strong.

Avidly’s role: We help configure these agents so they don’t just “exist” in your portal but actively improve outcomes, training them with your data, brand voice, and escalation logic.

Smart Insights and Flexible Views

These aren’t flashy features, but they’re quietly transformational. Instead of your teams digging through properties or activity logs, the CRM itself highlights what’s changed and what to do next, across whichever view suits the user (table, board, calendar, or dashboard).

The value here is in compounding productivity gains. Sales and service teams save minutes on every call prep, pipeline update, or follow-up task. Over time, those minutes turn into hours, and those hours turn into consistent pipeline movement.

Avidly’s role: We help clients surface the right insights by aligning HubSpot properties, data quality, and workflows so the recommendations are relevant, not noise.

AI-Powered CPQ

This one deserves a note of caution. It’s one of the headline announcements – but unlike Breeze Agents, which have had a year of iteration, CPQ is still early in its development.

It shows real promise: automated proposal generation, streamlined quote-to-cash, and tighter integrations. But usefulness depends entirely on product data and pricing rules being complete and accurate. For most enterprise or complex B2B clients, the out-of-the-box version will likely be too narrow to run standalone.

Avidly’s role: This is where our custom development and integration expertise matters. We can bridge the gap between HubSpot’s CPQ baseline and the bespoke logic most businesses need. Think of HubSpot’s CPQ as a foundation — and Avidly as the builder who makes it fit your business reality.

Why the Fundamentals Still Matter

As Joakim pointed out in his recent interview with HubSpot, it all starts with what some might call the “boring” work – cleaning data, building structures, and finding small, iterative wins. Without that, AI agents have nothing reliable to work from.

Before a Customer Agent can qualify leads properly, your CRM needs clean, enriched data. Before Smart Insights can flag risks, you need the right properties in place. Before CPQ can generate a proposal, your product library has to be complete and accurate.

AI does not erase the basics. It amplifies them. Which is why our starting point with clients is so often:

Cleaning and enriching CRM data.
Automating low-value manual tasks.
Building small, iterative use cases that show quick wins and build confidence.

Then AI Moving From Expensive Interns to Genuine Teammates

Nicholas Holland, HubSpot’s Head of AI, described the shift as moving from “expensive interns” to “high-value teammates.” We agree – but only if you treat agents as part of your team. They need onboarding, training, management, and iteration.

At Avidly, we take on that role early, acting as the first agent managers and trainers until clients are ready to take it on themselves. We also help them design an unstructured data strategy – most businesses don’t even realise how much value is trapped in meeting notes, call transcripts, and support tickets. And crucially, we support the change management needed to shift from AI curiosity to AI leadership.

Want to see how HubSpot’s new AI agents could work in your business? Book a session with Avidly and we’ll help you find the practical, iterative wins that matter most.

The Takeaway and How to Leverage These Tools Properly

INBOUND 25 confirms that HubSpot is embedding AI into the core of its platform – not as a bolt-on, but as a foundation. That’s a step forward. But announcements don’t create outcomes. Clean data, careful iteration, and hybrid human+AI teams do.
That’s where Avidly helps clients cut through the hype and achieve impact.