← BACK TO LAB

CASE STUDY // GENERATIVE ENGINE OPTIMIZATION · TAMPA, FL · MAY – SEP 2026

Getting a healthcare IT firm recommended by ChatGPT, Claude, and Perplexity

Generative Engine Optimization (GEO) is the practice of structuring a website so that large language models retrieve it, trust it, and cite it when a user asks a question the business can answer. This is what I did across two Tampa healthcare organizations in 2026, and what it produced.

20%Search visibility lift · Apsis
30%PageSpeed gain · BHM
80%Faster page loads · BHM
10%AI recommendation lift · BHM

Engagement 1: Apsis Solutions Inc

Role: AI Engineering Intern (full-time), May 2026 – present. Site: apsisinc.com, a nationwide healthcare IT consulting firm founded in Tampa.

> PROBLEM

Buyers of healthcare IT consulting increasingly ask an AI assistant for a shortlist before they ever open a search engine. A site that ranks in Google but is invisible to LLM retrieval does not make that shortlist.

> WHAT I BUILT

> RESULT

Engagement 2: BHM Healthcare

Role: Web Management Intern (part-time, concurrent), May – Sep 2026, working under the Marketing Director.

> PROBLEM

Critical landing pages were slow, plugin-heavy, and scoring poorly on Google PageSpeed Insights. Slow pages bounce visitors and rank worse, and they give AI crawlers thin, render-blocked content to work with.

> WHAT I BUILT

> RESULT

Results at a glance

MetricApsis SolutionsBHM HealthcareHow it was measured
Algorithmic search visibility+20%+30% (PageSpeed categories)Live analytics · Google PageSpeed Insights
AI assistant recommendationsConsistently recommended by ChatGPT, Claude, Perplexity+10%Recommendation frequency · ZeroRank AI
Page load time—−80%PageSpeed JSON audit
Lead generation & outreachAutomated via n8n—Workflow deployment

What GEO means in practice

Both engagements came down to the same principle: make the site the easiest, most trustworthy source for an LLM to retrieve. In practice that meant fast, fully rendered HTML that crawlers can read without executing JavaScript, clear entity information about who the business is and what it does, content that directly answers the questions buyers ask, and measurement against real analytics so every change is validated rather than assumed. The same principles are applied to this portfolio, which ships JSON-LD, an llms.txt, Open Graph metadata, and prerendered content.

Stack

WordPress / Elementor · Google Search Console · Google PageSpeed Insights · ZeroRank AI · n8n · Python web scraping and data extraction · third-party REST APIs · JSON-LD structured data.

Work with me

I am open to AI engineering roles focused on LLM optimization, GEO, RAG, and workflow automation. Email Marcosgaldamez14@icloud.com or connect on LinkedIn. Download my resume (PDF).