Six use cases from real customers — what’s yours?
When you connect Claude to HubSpot, your business and customer context becomes actionable in new ways. Ask Claude a question and get an answer grounded in what HubSpot knows about your business and customers. Claude brings the thinking. HubSpot brings the context that makes it useful for you specifically.
Our team at HubSpot has analyzed hundreds of thousands of interactions from early adopters of the HubSpot connector for Claude. While people are taking this tool in countless directions, a handful of the same use cases keep surfacing across different industries, different team sizes, and different phrasings of the same underlying need.
That’s what this post is about. Not what’s theoretically possible. What people are actually doing — and coming back to do again.
1. “Help me find and unlock the revenue hiding in my long tail.”
If something like 20% of your customers and partners drive 80% of your revenue, this one is for you. The other 80% exist in your CRM, receive your emails, and largely stay dormant. You know they’re an opportunity. You’ve never had the bandwidth to act on it.
The traditional answer is headcount. More coverage. More field visits. The math rarely works — once your business reaches a certain volume, adequate coverage becomes a staffing problem most companies never fully solve.
One company managing a large distributed network decided to solve it differently. Their goal wasn’t to cut costs. It was what their leadership called “surgical precision in engagement,” or the ability to act on every relationship, not just the ones their team had bandwidth for. With the HubSpot connector for Claude, their teams now manage multiples of what they could before. The long tail is no longer out of reach.
Here’s how it works. Their performance data lives in one system. The relationship context — emails, calls, notes, support tickets — lives in HubSpot. Claude connects both in the same session. For each underperforming account, it reads the full engagement history, cross-references it against performance data, and produces a diagnosis: what’s driving the gap, what to do about it, which channel to use, and what to say.
In one case, the answer had been buried in HubSpot all along: competitive pressure, a pattern of disengagement, and a personal note from months earlier that no one had followed up on. Claude read all of it and recommended exactly what to say and how to reopen the conversation.
High-priority accounts still get the human attention they warrant. What changed is everything else. The dormant 80% now gets precise, personalized engagement that would have been impossible to deliver at scale — not because the team grew, but because each person on it can do more.
Try this: Pull your lowest-performing partner or customer accounts, then ask — “Read the full HubSpot engagement history for each of these accounts and cross-reference against their performance data. For each one, diagnose what’s driving underperformance and recommend the next action: channel, key message, and what context from our history I should lead with.”
2. “Show me what’s working across our campaigns, as well as what to cut.”
If you’re a marketer, this one’s for you. Campaign performance data lives in HubSpot. So do your landing pages, blog posts, and website pages. What used to require an export, a spreadsheet, and a meeting now happens in a single conversation.
With the HubSpot connector for Claude, you can pull campaign performance directly into your workflow (read only) — reach, engagement, influenced pipeline — and ask the questions that actually matter: What’s driving results? What’s stalling? Where should we put the next dollar?
The marketers getting the most out of this aren’t using it to generate reports. They’re using it to think. Claude reads the data, spots the patterns across campaigns and content, and helps them plan next steps without switching tools or waiting on a RevOps request.
Try this: Ask — “Pull performance data for all active HubSpot campaigns from the last 90 days. For each one, show me reach, click-through rate, and influenced deals. Identify the two with the best ROI signals and the two that are underperforming. Tell me what they have in common and what I should change or cut.”
3. “Brief me before this call.”
You have a call in 20 minutes. The account has been in your CRM for two years. Somewhere in there is the full history — the deals, the conversations, the last time someone from your team reached out and what happened. You don’t have time to piece it together manually.
This is the job that pre-call prep has always been. HubSpot holds the full picture. Now Claude surfaces it in the moment you need it.
One VP of Sales described the shift: he could pull pipeline and account context on demand, without building a report or waiting on RevOps. The information was already in HubSpot. Claude made it accessible. Pre-call prep was one of the most consistently mentioned use cases across many conversations we had with customers.
Try this: Before your next call, ask — “Pull everything HubSpot has on [company name]. Summarize the relationship history, current deal stage, last activity, and any open issues. Tell me what I should know before this call.”
4. “Log my call, update the deal stage, and create a follow-up task.”
You just got off a call. You have notes — maybe typed, maybe messy, maybe a mix of both. What you don’t have is time to open HubSpot, find the record, log the engagement, update the deal stage, and create a follow-up task before your next meeting starts.
This is the part of CRM hygiene that everyone agrees matters and almost no one does consistently. Not because people don’t care, but because the gap between finishing a call and updating the record is just wide enough that it doesn’t happen.
With write access, the HubSpot connector for Claude closes that gap. Paste your raw notes into the conversation. Claude reads them, identifies the relevant contact and deal, and handles the logging — engagement recorded, deal stage updated, follow-up task created — in one shot. It confirms what it’s about to do before touching anything, so nothing lands in HubSpot without your sign-off.
The output isn’t just a cleaner CRM. It’s a record accurate enough to be useful the next time someone needs to prep for a call with that account, bringing you back to use case #2.
Try this: “Here are my notes from a call I just finished: [paste notes]. Find the matching contact and deal record in HubSpot. Log the call as an engagement with a summary, update the deal stage to reflect where we landed, and create a follow-up task for me with the next step and a due date.”
5. “Tell me what HubSpot knows about these people.”
Someone hands you a list. Could be event attendees, a batch of inbound leads, a set of accounts a new hire brought over. Everyone on it is a potential opportunity — but without context, every name gets the same generic outreach.
Using the HubSpot connector for Claude, you can answer the real question before you write a single word: Who’s already a customer — and shouldn’t be cold-approached by your sales rep? Who’s a warm prospect that went quiet after a pricing conversation? Who has an open deal that’s been stalled for three months?
Getting this wrong is expensive. You pitch a long-time customer like a stranger. You send a prospecting sequence to someone mid-deal. Your team spends two days at a conference talking to the wrong people.
HubSpot already knows the difference. Claude makes that knowledge portable — pulling it out of your CRM and into whatever format you’re working in. A cold list becomes a prioritized one: who to reach out to first, what to say, and who should never receive a prospecting sequence.
This is the single highest-volume use case in our data — not because it’s the flashiest, but because it happens before every event, every campaign, every outbound push.
Then you can act on it directly. Claude can log tasks, update contact records, and adjust enrollment in sequences — all in the same session, and always with your confirmation before anything changes in HubSpot.
Try this:
Step 1, enrich: Upload your list as a CSV, then ask — “For each company in this list, check my HubSpot records and return: company name, primary contact name and email, lifecycle stage, current deal stage, and most recent activity date. Note whether they’re a current customer, an open opportunity, or no record found.”
Step 2, act: “For each warm prospect on this list who’s been inactive for 60+ days, create a follow-up task for their account owner. For current customers, update their contact record to exclude them from future outbound sequences. Leave anyone with an active deal untouched.”
6. “Based on my team’s current workload, where should I be making changes?”
Sales managers and RevOps leads deal with a version of this constantly. The pipeline is visible. What’s harder to see is whether it’s distributed in a way that makes sense — who’s overloaded, who has capacity, and which deals are sitting with the wrong rep.
That context now lives in the HubSpot connector for Claude. Team memberships, reporting lines, user roles, seat assignments — all of it accessible in the same session where you’re looking at pipeline. Instead of pulling rep-level data manually and cross-referencing it against deals, you can ask the question directly.
In practice, this is most useful at two moments: when pipeline velocity slows and you’re not sure if it’s a market signal or a coverage problem, and when someone new joins the team and you need to redistribute accounts without guessing at capacity.
The goal isn’t to automate the decision. It’s to give managers the full picture fast enough to make a call before a deal goes cold or a rep burns out.
Try this: “Pull all open deals in HubSpot. For each rep on my team, show me their current deal count, total pipeline value, and average deal age. Identify anyone who’s overloaded relative to the team average and suggest which deals could be reassigned and to whom.”
- Note: If you already have the HubSpot connector installed, you’ll need to update it to access these new capabilities. Learn how to update your connector.
One thing that determines whether any of this works well
HubSpot holds your context, but every instance is configured differently. Custom pipeline stages, non-standard lifecycle definitions, properties named things only your team would recognize. Claude doesn’t know your setup by default.
The customers getting the most consistent results have spent 20–30 minutes describing their instance to Claude before running queries: what their pipeline stages mean, how they define a qualified lead, which custom properties actually matter. Claude’s Skills feature is the most durable way to do this — write it once, apply it every session.
Try this: Install the Claude for Small Business plugin for Cowork and use the prebuilt onboarding skill by typing the command /smb-onboard. The whole process takes less than a half hour and ends with Claude knowing enough about your business to be immediately useful.
Think of it as onboarding Claude the same way you’d onboard a new team member. The context is already in HubSpot. This step makes sure Claude knows how to read it.
Your turn.
These use cases are a starting point, not a ceiling. Honestly, you’ve probably already gone further.
People are chaining HubSpot with other tools, automating workflows, and building things we haven’t seen yet. If you’ve found a prompt or workflow that actually works, drop it in the comments. Share what you were trying to do, what you landed on, and what it replaced. The more specific the better — a prompt that reliably works is worth ten descriptions of what’s theoretically possible.
The HubSpot connector for Claude is generally available to all users and plans. Setup takes under 10 minutes. Learn more