Scaling a FinTech Media Platform from 200K to 3M+ Monthly Visitors
A 2.5-year engagement where I built a 30-person organic growth department from scratch, scaled monthly revenue from $3K to a $4M peak, and pioneered AEO strategy across the property before the category had a name.
Context
A FinTech media platform brought me on as acting Head of SEO in late 2023. The site had real traffic (200K monthly visitors) but monetization was broken — $3,000 in monthly revenue against a property that had the audience to generate orders of magnitude more. The organic growth function didn’t exist as a department. There was no shared tooling, no attribution framework, no cross-functional alignment between editorial and engineering, and no strategy for the emerging shift to AI-driven search.
Challenge
Three problems, layered:
- Revenue was under-indexed to audience. The site had traffic but wasn’t capturing commercial intent efficiently.
- The operating model wasn’t built for scale. Growth depended on individual heroics, not systems.
- The search landscape was shifting. Traditional SEO was beginning to share share-of-voice with AI Overviews, Perplexity, ChatGPT, and similar answer engines. No one in the industry had a clear playbook yet.
Strategy
I organized the work into five parallel tracks:
1. Build the team as a system, not a hiring plan
Instead of hiring SEO specialists one at a time, I designed the department as five interconnected functions: Engineering, Editorial, Design, Social Media, and a small Analytics function that served all four. Each function had clear ownership, clear KPIs, and clear interfaces with the others. By the time the team reached 30 people, it operated more like a product organization than a marketing team.
2. Pioneer AEO and GEO strategy
I aligned the technical infrastructure with how AI discovery patterns actually work — entity recognition, structured data, extractable content, and source authority signals. The site became one of the first in its vertical to systematically optimize for SGE, Perplexity, Gemini, and ChatGPT. AI citations increased 25% month-over-month through the engagement.
3. Build the data infrastructure before scaling content
Before publishing at scale, we built the measurement and attribution layer. I partnered with Data Engineering to build custom Looker dashboards that improved attribution accuracy by 30% for affiliate click and conversion tracking. This let us make content investment decisions based on expected revenue, not guesswork.
4. Use AI operationally, not experimentally
I architected an agentic content planning pipeline using Claude Code that automated the research-to-brief workflow, increasing content production velocity by 40% while maintaining a 95% source accuracy rate before human-in-the-loop editorial review. I also built a scalable OpenAI-powered data auditing pipeline that improved audit efficiency by 65% and reduced manual review cycles.
Separately, I led the development of a centralized data-driven templating system that synchronized real-time financial market data across thousands of pages, using conditional logic to automate narrative accuracy. This reduced editorial maintenance by 80% while ensuring 100% factual consistency across the domain.
5. Restore brand integrity
The property inherited four active Google Manual Actions from legacy content and link strategies. I led the remediation: overhauling affected content, enforcing E-E-A-T standards across the domain, and coordinating reconsideration requests. All four manual actions were resolved, with 100% of lost domain authority restored within an average of six months.
Results
Over 2.5 years:
- Monthly traffic: 200K → 3M+
- Monthly revenue: $3K → $4M peak, $2M/mo average
- AI citations: 25% increase month-over-month across AI Overviews, Perplexity, Gemini, and ChatGPT
- Rich snippet CTR: +18% through structured data implementation and dynamic widget deployment
- Attribution accuracy: +30% for AI-influenced engagement and affiliate conversion tracking
- Content production velocity: +40% via agentic content pipelines
- Audit efficiency: +65% via OpenAI-powered automation
- Editorial maintenance: –80% via dynamic data-driven templating
- Revenue per user: +22% via CRO and A/B testing frameworks
- Google Manual Actions: 4 resolved, 100% of lost domain authority restored
What I’d Do Differently
If I started the engagement today, I would invest in AEO-specific measurement infrastructure earlier. When we started, there were no mature tools for tracking AI citation share or LLM response inclusion. We built internal tracking manually. If I were doing this from scratch now, I’d formalize that measurement layer in month one, not month twelve — it would have let us optimize for LLM visibility a year sooner.