What Firmbase's MCP integration means for your sales workflow

By Dave Curran, Co-Founder, Firmbase | March 2026 | 6 min read

If you've been reading about AI assistants and Model Context Protocol (MCP), you've probably seen it mentioned as a big deal. But if you're not deep in AI infrastructure, it's not immediately clear why you should care.

Here's why: MCP lets you talk to your sales tools in plain English, and have an AI assistant query them directly - without switching tabs, without manual data entry, without the friction that usually comes with integrations.

What MCP actually is (in one paragraph)

MCP is an open standard that lets AI assistants connect directly to tools and data sources. Instead of you having to grab data from five different tabs and paste it into a document, you ask your AI assistant a question in plain English, and it queries those tools directly to answer it. Think of it like giving an AI assistant the ability to use your software stack. It reads what it needs, synthesises it, and gives you the answer.

What this means for Firmbase users

With MCP integration, you can ask Claude (or any MCP-compatible AI) questions about your Firmbase data directly:

"Which of my target accounts have shown a buying signal in the last 30 days?"

The AI queries Firmbase, pulls back accounts matching that criteria, and gives you a list with context.

"Find 20 IT consultancies in Manchester with revenue between £2M and £10M that have appointed a new director this year."

The AI searches Firmbase using your exact criteria, pulls back matching accounts, and surfaces them. No manual filtering. No exporting to spreadsheet. No copy-paste.

"Show me the top 10 accounts in my universe that are showing three or more buying signals this month."

The AI applies signal weighting, ranks accounts by signal intensity, and surfaces your priority list.

This is different from using Firmbase directly. You're not logging into the interface and clicking through dashboards. You're having a conversation with an AI that has access to your Firmbase data.

Why this matters for sales workflows

Most sales workflows involve a lot of tab-switching and copy-pasting:

That's six tabs. That's friction.

With MCP integration, it's one conversation:

"I'm about to reach out to Example Ltd. Pull their latest Companies House data, show me the buying signals we've identified, tell me who their leadership team is, and suggest an outreach angle."

The AI does the work. It pulls Companies House data (revenue trajectory, director appointments, cash position). It shows you the signals Firmbase has identified (job postings, funding, growth rate). It synthesises that into a recommendation.

You get a one-page brief. You write your email. Done.

Real examples of what this enables

Example 1: Building a weekly priority list

"Give me the top 20 accounts in my target universe that are showing buying signals this week, sorted by signal strength. For each one, tell me what signals they're showing and suggest the next action."

The AI pulls your universe from Firmbase, identifies accounts with recent signals, ranks them by signal convergence, and gives you your weekly hit list with context and next steps. That's a 30-minute task if you do it manually. It's five seconds with MCP.

Example 2: Researching before a cold call

"I'm calling the Managing Director of Example Ltd this afternoon. What do I need to know? Pull their recent Companies House filing, show me if they've hired anyone recently, tell me what job postings they've got open, and suggest what pain points they might have based on their growth pattern."

The AI pulls all of that, synthesises it into a one-page research brief, and you have real context before the call.

Example 3: Building a strategic account plan

"Show me all accounts in my universe that have appointed a new finance director in the last six months AND have revenue growth >30% year-on-year. For each one, pull their latest financial position, show me if they're taking on debt, and tell me whether they look like acquisition targets or bootstrapped growth."

The AI applies those multiple criteria, surfaces matching accounts, and gives you insight into which ones might be in capital-raise mode (and therefore more open to new tools).

Why MCP integration is inevitable

The next generation of sales software won't be "best-in-class individual tools." It'll be open ecosystems where best-in-class tools connect directly to AI assistants.

Instead of "Firmbase is good, but you have to use it" - it's "Firmbase is part of your sales stack, and you can talk to it through your AI."

That's radically more powerful because it means:

How to set this up

If you're using Claude (which has MCP support built in), it's straightforward:

No engineering required. No API configuration. Just conversation.

MCP integration is available now for Firmbase users

If you're using Claude or another MCP-compatible AI assistant, you can connect directly to your Firmbase data and build your sales workflow around natural language queries.

Start your free trial to access MCP integration, or get in touch if you're already a customer and want to enable it.

FAQ

Q: Do I need to be technical to set up MCP?

A: No. If you can connect an app to your email, you can set up MCP integration. It's a few clicks and you're done.

Q: Will this work with other AI assistants besides Claude?

A: MCP is an open standard, so yes - any AI assistant that supports MCP can integrate with Firmbase. Claude is the most developed right now, but others are coming.

Q: Is my data secure if I'm querying it through an AI assistant?

A: Yes. Your data stays within your Firmbase instance. The AI queries it, reads the response, and that's it. We don't log queries or share your data with the AI provider.

Q: Can I combine Firmbase data with other tools in one query?

A: If those tools also have MCP integration, yes. You could theoretically ask: "Show me accounts from Firmbase that have mentions in our CRM marked as 'hot', have emails sent in the last week, and show up in our LinkedIn searches for [keyword]." The AI pulls from all three.

Q: What if I want to automate outreach based on MCP queries?

A: The current integration is read-only, so you can't trigger outreach directly. But you can pull a list from MCP and then action it manually, or integrate with an automation tool that watches Firmbase for signal changes.

Q: How is this different from just exporting data from Firmbase?

A: Exports are static - you export once and the data gets old. MCP queries are live - you ask a question and get real-time data. You can also ask complex questions that require synthesis (multiple criteria, signal weighting, etc.) that export + manual analysis can't handle.

Author Bio

Dave Curran is the co-founder of Firmbase, a UK B2B sales intelligence tool that helps sales teams find, prioritise, and reach the right accounts without needing a RevOps team to make it work. Before Firmbase, Dave co-founded Love Mondays (acquired by Glassdoor, where he went on to serve as VP of Product) and Openvolt. He writes about UK B2B sales, prospecting, and go-to-market strategy.

Firmbase helps UK B2B sales teams discover their complete account universe, prioritise based on real buying signals, and reach out with genuine relevance - without the complexity of enterprise tools. Start your free trial