How I went from $800M revenue projects to building open-source GTM tools, and what I learned

Where this started: enterprise at scale

I spent 15 years inside some of the largest transformation projects in Quebec. Seven major enterprise engagements. Over $50M in budgets managed. One assignment that contributed to $800M in revenue at the Quebecor retail network.

I know what it looks like when revenue operations work at full power: cross-functional alignment, real-time pipeline intelligence, ICP models that actually predict conversion, forecast accuracy within 5% of actuals.

It's genuinely impressive. It's also completely inaccessible to any company that isn't sitting on an eight-figure technology budget.

The problem I couldn't stop thinking about

On the SMB side of the market:

• Brilliant founders with strong products losing deals they should have won


• Sales and marketing teams working in parallel universes


• Revenue forecasts missing by 30-40% - not from bad execution, but from invisible structural constraints


• AI tools adopted enthusiastically but used as glorified autocomplete

The gap wasn't strategy. It wasn't effort. It was systems.

The methodology became the product

I developed the Artefact Method - a structured consulting framework to help SMBs build enterprise-calibre intelligence systems, faster and without the enterprise price tag.

It worked. Clients described getting results in 90 days that they'd been chasing for over a year.

But consulting doesn't scale. Every engagement was bespoke. Every diagnosis started from scratch.

That's when I started asking: what if the methodology itself could be encoded?

Enter MCP - the infrastructure I had been waiting for

When Anthropic published the Model Context Protocol specification, I spent three days reading almost nothing else.

MCP gave me a standard way to build structured tools that an AI could call using natural language - a framework for encoding analytical methodology as callable intelligence.

I rebuilt every core framework from 15 years of consulting into 7 MCP tools. Then I open-sourced it.

Why open-source

First: The GTM tooling ecosystem is dominated by data connectors. Everyone is building pipes. Nobody is building methodology. The fastest way to prove a new category is to make the entry cost zero.

Second: The developers building on top of AI will shape what business intelligence looks like for the next decade. I'd rather help them build it with analytical rigour baked in.

What's coming next

• Pro tier - live HubSpot data, custom RFM thresholds, custom ICP mapping ($149/month)


• Enterprise tier - multi-CRM support, dedicated onboarding ($499/month)


• Artefact CRO Platform - full SaaS revenue operating system (artefactcro.app)


• Ongoing open-source development - more tools, more signal types, community contributions welcome

Three things to take from this right now

1. 
   Diagnose before you optimise. Most revenue problems feel like execution problems but are structural. Run a constraint analysis before you add headcount or spend.


2. 
   AI is only as useful as the data it can see. If your AI tools are working from copy-pasted context, you're using a calculator to do brain surgery.


3. 
   Methodology is the moat. Data is everywhere. The companies that win are the ones with a repeatable, systematic approach to turning data into decisions.

Let's build this together: What's the biggest structural gap in your GTM right now?

Try Artefact MCP free:

pip install artefact-mcp 

GitHub: github.com/alexboissAV/artefact-mcp-server

Pro access:

Direct: [email protected]