How to Use Windsor.ai for CRM

Marketing can see which campaign generated the most clicks, and sales can see which deals closed, but there was no way to see the line between them without a manual pull: which channel’s leads became pipeline, and which just inflated a lead count.

GoHighLevel, HubSpot and Salesforce track what happens inside the CRM; Meta Ads and Google Ads report their own cost per lead. Neither shows what a lead actually cost once it closed, so answering “which channel produces revenue” used to mean exporting both and joining them by hand.

SEO is where I spend most of my time, but I want to understand how every channel in the business works together, not just search. Connecting the CRM stack and ad accounts to Windsor.ai meant I could finally separate the channels that close deals from the ones that only add leads that never convert. Now I ask AI directly: which source, how much, what stage, the same questions I see other Windsor teams ask too. I use Claude daily for search work already, so that’s where I do this too, though ChatGPT, Gemini and Perplexity work just as well.

🚀 Ready to see it live? Try Windsor.ai free, free forever plan, no credit card.

Already using Claude? Add the Windsor.ai for CRM connector directly from the Claude directory. On ChatGPT, Gemini or Perplexity, add Windsor the same way from that assistant’s own connector or app directory.

Connecting your CRM data to your AI assistant with Windsor: 2 steps

Windsor.ai is a single connector that reads from 350+ data sources, including GoHighLevel, HubSpot, Salesforce and every major ad platform, and sends it wherever you work. Setup is two steps: connect your sources in Windsor, then add Windsor to your assistant.

HubSpot also ships its own official MCP server, built for working inside HubSpot itself: creating, updating and searching records. Windsor.ai does a different job. It blends your CRM alongside ad spend and 350+ other sources in the same conversation, for the revenue question rather than record management.

Before you start

• A GoHighLevel, HubSpot or Salesforce account
• A Windsor.ai account (free for the first 30 days to test; paid plans start at $19/month for 3 sources, $99/month for 7)
• A Claude, ChatGPT, Gemini or Perplexity account

Step 1: Connect your CRM and ad accounts to Windsor

Go to onboard.windsor.ai/app/gohighlevel and sign in to authorize your account. Then add whatever else feeds the revenue picture:

• HubSpot or Salesforce, if you run either alongside GoHighLevel
• Meta Ads and Google Ads, for the spend side of the picture

The free plan lets you test multiple sources for the first 30 days, then drops to a single source with no credit card required. Keeping your CRMs and ad accounts connected long-term takes a paid plan: $19/month for 3 sources, $99/month for 7, with pipeline history going back further.

Step 2: Add Windsor to your AI assistant

• Claude: Customize → Connectors → search Windsor → Add
• ChatGPT: install Windsor’s native app from the ChatGPT apps directory
• Gemini and Perplexity: add Windsor from that assistant’s own connector or app directory

Sign in to your Windsor.ai account when prompted, then confirm it works with a simple question: “how many new deals did we create last week.”

The numbers I track once it’s connected

Metric	What it tells you
New leads / contacts (GoHighLevel, HubSpot, Salesforce)	Volume entering the pipeline
Deal stage and pipeline value	How much of that volume is qualified
Closed-won revenue	What converted
Cost per lead, by ad platform	What it cost to generate that volume
Lead-to-close rate, by source	Which channel’s leads close

The reports CRM teams run most

These come straight from how CRM teams use Windsor day to day, not a hypothetical list:

Lead source and revenue

• Which contact source brought in the most new leads this month, and how are they tagged?
• Which opportunity source is producing the highest-value deals, and what stage are they at?
• Which lead source has closed the most deals, and what’s the total deal value by source?

Pipeline

• Show me our pipeline stages and how many opportunities are sitting in each one.
• Pull this week’s new contacts with their source and tags.

Bonus: analyze CRM alongside your full stack

Because Windsor connects 350+ sources through the same install, I look past the CRM as well. I’ll ask things like:

• “Which Google Ads campaigns produced deals that closed in the CRM, versus campaigns that only generated leads?”
• “Compare cost per lead across Meta Ads and Google Ads against CRM win rate.”
• “Did the campaign we launched in March produce any closed-won revenue yet?”

Conclusion

Lead volume and pipeline value shouldn’t need a manual export to compare. I connected my CRM and ad accounts to Windsor once, added Windsor to Claude, and now I just ask which channel produces revenue.

Connectors mentioned in this guide

• GoHighLevel
• HubSpot
• Salesforce
• Meta Ads
• Google Ads

🚀 Ready to get started? Try Windsor.ai free. Free forever plan, no credit card. Paid plans from $19/month.