01 Context
Advanced Hormone Solutions is a hormone therapy clinic in Paramus, New Jersey, led by Dr. Susan Matos-Cloke. The practice specializes in bioidentical hormone replacement, SottoPelle pellet therapy, menopause and andropause care, and related treatments. It is a high-trust category where patients research carefully before they ever pick up the phone.
The clinic had a strong real-world reputation: a 4.8-star rating across 26 reviews and a loyal patient base across Bergen County. But the website that was supposed to represent all of that had been live since 2015 without a structural overhaul, and it showed. Organic search was underperforming, the site was nearly invisible inside AI-generated answers, and the clinic’s own Google Business Profile listed none of the services it provides.
The brief was deliberately scoped: a one-time technical SEO and Google Business Profile foundation. Fix what was broken, make the site fully crawlable, and give patients and AI engines a clean, accurate signal to work from. Not a growth campaign. The groundwork a growth campaign would later stand on.
02 The challenge
The first Semrush crawl made the scale of the problem concrete. Of 326 pages crawled, 301 (92.3%) had at least one issue. There were 15 critical errors, nearly 1,800 warnings, and hundreds of notices. The site wasn’t penalized so much as quietly working against itself.
The specifics compounded each other. Six pages shared duplicate title tags and two pages were 85%+ identical, so Google couldn’t tell which page to rank. Dozens of pages sat in the sitemap with no internal links pointing to them, leaving them orphaned and effectively unreachable. There were broken images, dead outbound links, blocked pages, and 275 images with no alt text at all. More than 1,000 pages shipped unminified CSS and JavaScript, dragging on load speed.
Underneath the technical debt sat a second, newer problem: AI visibility. The site scored 18/100 on LLM visibility and was rarely surfaced by ChatGPT, Gemini, or Google’s AI Overviews. It carried no structured data: no MedicalOrganization, Physician, or FAQ schema, which is exactly the machine-readable context AI engines and answer boxes rely on to cite a source confidently.
And the clinic’s most important local asset was underbuilt. The Google Business Profile had no services listed, no service area configured, no booking URL, and a generic description, leaving it invisible for the non-branded “near me” searches that drive new patients.
03 Approach
We plotted the engagement against the AXIS Model and led with Insight, the work of being found and understood, supported by Impact, the measurement and local-presence layer that proves it. Identity-level work (homepage and design) was deliberately held for a later phase. The thesis was simple: there’s no point pouring content, ads, or backlinks into a site that search engines can’t cleanly read. Foundation first.
That meant a strict sequence rather than a scattershot fix list. We took a baseline Semrush crawl on March 9, 2026 to document the starting state, worked through the issues in priority order (technical errors before on-page optimization before AI readiness), and then re-crawled on April 9, 2026 to validate every change against the same yardstick. Tracking was installed early so that, from this point forward, the clinic could measure its own results instead of taking them on faith.
04 What we shipped
A clean technical foundation. We cleared all 15 critical errors. Duplicate title tags, duplicate content, broken internal images, and 4xx errors were all resolved to zero. Thirty orphan pages were reconnected or removed, broken links repaired, and blocked pages reviewed. Total issues fell from 2,528 to 1,090, a 56.9% reduction, with the warning count more than halved.
On-page optimization at scale. We added descriptive, keyword-relevant alt text to all 275 images that were missing it, took missing meta descriptions from 106 pages down to 2, fixed duplicate meta descriptions, resolved a multiple-H1 page, and rewrote hundreds of non-descriptive anchor links (453 down to 222). Unminified CSS/JS pages were cut from 1,088 to 498 and thin text-to-HTML pages from 247 to 162, for faster pages and cleaner code.
An AI-ready layer. We deployed structured data across key pages and implemented an llm.txt file so AI crawlers can understand the site’s structure, alongside AI-oriented content optimization. The result moved AI Search Health from 86% to 89% and gave ChatGPT, Gemini, and Google’s AI Overviews the machine-readable signals they need to cite the clinic.
A fully optimized Google Business Profile. We rebuilt the clinic’s local front door: a high-resolution logo, a keyword-rich description (BHRT, menopause, pellet therapy), synced hours, on-site parking and appointment attributes, and UTM-tracked website and booking URLs. We added 13 treatment services with individual descriptions and configured a 20-city service area across Bergen County, from Paramus and Hackensack to Ridgewood, Teaneck, and Franklin Lakes.
A measurement stack that actually reports. We configured Google Search Console and GA4, set up Google Tag Manager and Microsoft Clarity, and built key event tracking for contact-form submissions, consultation requests, phone clicks, and email clicks, plus clickable phone and address links sitewide. Baseline reporting was documented so every future gain is measured, not marketed.
05 Results
Validated against a like-for-like Semrush re-crawl one month after baseline:
85 → 94
Site Health Score (+9 points)
15 → 0
Critical errors cleared (100%)
1,124
Warnings fixed (62.8%)
Beyond the headline numbers: 275 images fixed for alt text and accessibility, missing meta descriptions cut from 106 to 2, AI Search Health lifted from 86% to 89%, and structured data deployed across key pages. On the local side, the Google Business Profile went from zero listed services to 13, with a 20-city service area and full UTM tracking in place, turning a passive listing into an active acquisition channel.
| Metric | Before | After | Change |
| Site Health Score | 85% | 94% | +9 pts |
| AI Search Health | 86% | 89% | +3 pts |
| Total errors | 15 | 0 | 100% cleared |
| Total warnings | 1,789 | 665 | 1,124 fixed |
| Total notices | 724 | 425 | 299 fixed |
| Total issues | 2,528 | 1,090 | 1,438 fixed |
| Duplicate title tags | 6 | 0 | Fixed |
| Orphan pages | 30 | 0 | Fixed |
| Images missing alt text | 275 | 0 | Fixed |
| Missing meta descriptions | 106 | 2 | 104 fixed |
| Schema markup | None | Deployed | 5 types |
| GBP services listed | 0 | 13 | Fully built |
| GBP service area | Not set | 20 cities | Configured |
06 Reflection
The honest framing matters here, because it’s what makes the work credible. This was a one-time foundation engagement, and the largest unlock wasn’t a traffic spike. It was removing every reason a search engine or AI model had to distrust the site. A clean crawl, accurate structured data, and a fully described local profile are the prerequisites for everything that comes after.
That’s also why the story doesn’t end at handoff. Organic traffic is still down year over year, and the clinic’s domain authority (12) remains well below competitors like Biote (47) and HerKare (25). Those gaps don’t close with technical fixes alone. They close with the next phase: consistent content, authoritative backlinks, and the Identity-level design work the foundation was built to support. The site is now clean, fast, readable, and AI-ready. The runway is clear for growth to compound on top of it.