GTM Engineering · Marketing Operations
Twenty years building and scaling systems for revenue teams. Applying experience and judgment in an AI-native era. Trust me, I've seen it all.
Where I can help
The handoff between marketing and sales is where most pipeline goes quiet. Leads sit, ownership blurs, and both sides end up with their own version of the process. I build the architecture and workflows that close that gap, designed for how teams actually work now. The old playbook is five years stale.
Tracking UTMs and conversions is the easy part. The friction is in how teams interpret the data when marketing is trying to measure campaign effectiveness and sales is tracking comp plans. Both can coexist. We will work toward one set of numbers everyone recognizes, and a reporting layer people stop arguing with.
Most organizations use a fraction of what they've bought. Whether you're onboarding a new platform or consolidating for efficiency, I've seen where these systems fail and can help get the most out of your investments.
Things are moving fast in an agentic era. Agents and automations can only work from what's been written down. We turn tribal knowledge and undocumented process into clean, current knowledge bases your systems can actually use, so the outputs are repeatable and worth trusting.
I brought Ronnie in to make sense of our GTM data and he moved fast. He audited HubSpot and BigQuery and built a knowledgebase my team and our AI agents pull reports from. A week later we were rethinking our signup flow based on what actually converts.
The POV
Most GTM deployments don't fail on the technology. They fail on the foundation.
Standing something up has never been easier. An agent in an afternoon, a hundred landing pages in a week, a new platform live by the end of the month. AI natives can ship fast, but often lack the judgment that determines whether an initiative will last in six months. The expensive part is everything underneath the build: data clean enough to trust, a process worth automating in the first place, and enough governance that it doesn't drift or break at scale. This applies across the board whether it's an agent, attribution model, or a CRM migration. So the work starts with an honest read on what's actually working, and what to fix first so it's still standing a year from now.
Frameworks and methodology
For teams already committed to building with AI, this framework sorts out what to take on first and where the foundations need work before you start. It's free and ungated.
Run the diagnosticDiagnose, plan, build, iterate. Every engagement runs the same arc: an honest read first, then a sequence, then working systems your team owns.
Full version on servicesProjects
The fastest way to see how I work is to look at what I've shipped on my own.
AI signal-intelligence platform that builds buying-group intelligence for PLG. Connects the committee across your systems, then tells you who's missing and how to activate them.
Platform for local poker rooms to manage their website and games for members.
A skill that helps automate job search, company research and interview prep.
Writing
Field notes and hot takes on recent trends on AI and GTM.
AI made building cheap, so the thin stuff might be in trouble. The complex, the regulated, and the things nobody wants to own are not. Most internal builds skip the question that actually matters.
Why most AI projects fail on the foundation, and the calibration that separates AI worth building from AI worth skipping.
About
Two decades building the systems that connect go-to-market strategy to what actually ships. Lately that means doing it with AI, carefully.
More about meSystems and processes that last
I've designed and implemented multi-million dollar martech stacks and marketing data infrastructure that kept running long after the leadership that scoped it moved on.
Sprawl into something runnable
Turned half-integrated, overlapping stacks into systems a team could actually operate.
Automation that holds up in production
Built AI and automation into live GTM workflows that kept running past the proof-of-concept stage.
Signal architecture, end to end
Took PLG and PQL signal models from idea to instrumented pipeline.
