I built the funnel ICANotes runs on
I designed ICANotes' marketing end to end and manage it day to day. Instead of stitching together off-the-shelf tools, I built custom AI-native systems for each part of it: advertising, landing pages, content and SEO, lead generation, and outbound. I set the strategy, built the infrastructure, and direct the AI tools and small team that run it. Everything points toward one of two conversions, a free trial or a booked live demo.
AI Advertising
This is where AI does the most work. I run paid acquisition across Google, Microsoft, Meta, Reddit, LinkedIn, and ChatGPT Ads, plus the CRM, analytics, and CMS behind them, from one system I built. AI creates the campaigns, runs split tests across far more variants than anyone could manage by hand, and monitors and adjusts performance around the clock. The real win is efficiency and speed: one person doing the work of an entire ad team. On the creative side, I make the Meta image and video ads with my own AI studio, host them on our own infrastructure, and build custom and lookalike audiences from first-party data. Server-side conversion tracking keeps the numbers honest even against ad-blockers.
Three of my best-performing Meta ads, made with my AI studio:
The AI behind it. A custom MCP server connects Claude to every ad platform plus the CRM, analytics, and CMS, so it can build campaigns, run and read split tests, and monitor performance continuously, with a human confirming every change that goes live.
Landing Pages
Every campaign gets a page built for its exact offer. I built a React visual builder with A/B testing baked in, so there's no third-party testing tool in the loop. A theme system lets me reskin a page without touching its structure, components are reusable across pages, and the form widgets ship with bot protection and email validation. Domains are decoupled, so I can stand up and iterate on a page fast.
The AI behind it. AI generates and tunes the page copy per campaign so the ad-to-page message stays tight, and the built-in testing finds what actually converts.
Content & SEO
The engine that builds the top of the funnel and compounds. A Claude-powered blog system turns a topic or a source document into structured, WordPress-ready posts with images, no manual publishing bottleneck. On the SEO side, keyword research, technical audits, and opportunity discovery run in one workflow, with automated scanning for behavioral-health RFPs and a bias toward buying-intent keywords.
The AI behind it. Claude drafts full articles from a topic or document and delivers them publish-ready, while the SEO workflow surfaces the keywords and contracts worth chasing.
Lead Generation
Free interactive tools clinicians actually want, an MSE assessment, a SOAP-note tool, a progress-note stress test, each gated behind an email. Sign up and you're automatically enrolled into a nurture sequence matched to what you came for, so educational traffic turns into segmented pipeline instead of a one-time visit.
The AI behind it. AI builds the tools and the capture flow, and the segmentation that routes each new lead into the right sequence.
Outbound Email
A pipeline that sources practices straight from the national NPI provider registry, then enriches each one: mechanical data joins first, then Claude works the web to fill in practice details and verify the contact. The list runs through deliverability filtering before anything sends, and replies flow back into the CRM.
The AI behind it. Claude runs the research pass on every lead, filling in the details and verifying the contacts that make outbound land instead of bounce.
