Concept design by Larry Hall · EdgeReport — preview only, not affiliated

SHINEMedical Aesthetics

Performance Audit

The same business — measured on two foundations.

A side-by-side Lighthouse comparison of the current SHINE Medical Aesthetics website and a modern rebuild on production-grade infrastructure. Measured May 23, 2026.

Current Platform

WordPress with the Uncode theme and Visual Composer (WPBakery) on GoDaddy shared hosting. Multi-purpose ThemeForest themes prioritize visual flexibility for non-technical editors over page performance. Heavy bundles, render-blocking scripts, slow server response — mobile Lighthouse scores 27 with a 22-second largest contentful paint.

The same stack also blocks every major AI crawler — GPTBot, ClaudeBot, PerplexityBot, Google-Extended — in robots.txt. This is almost certainly an automatic setting from a WordPress security plugin, applied without awareness of the visibility consequence.

Proposed Platform

Next.js on Vercel edge infrastructure. Images served as WebP with intelligent lazy loading. AI crawlers properly allowed. Schema markup, llms.txt, and metadata built for the language models now answering commercial queries — and correctly accessible to them. Fully managed. The discipline is ours; the business is yours.

Lighthouse — lab measurements

Two devices, two foundations.

Lighthouse measures performance, accessibility, and best practices. Mobile and desktop are scored independently. The same URL is measured for both — only the connection profile changes.

Current Site

shinemedstudios.com

Measured May 23, 2026

27
Mobile
42
Desktop

Lighthouse Performance

LCP — Mobile22 s
LCP — Desktop3.0 s
Accessibility / Best Practices89 / 100
Modern Rebuild

shine-medical-demo.vercel.app

Measured May 23, 2026

98
Mobile
100
Desktop

Lighthouse Performance

LCP — Mobileunder 2 s
LCP — Desktopunder 1 s
Accessibility / Best Practices100 / 100

How language models see your business

Field data — measured from actual LLM queries.

The site's robots.txt file currently contains the following directives:

User-agent: GPTBot           Disallow: /
User-agent: ClaudeBot        Disallow: /
User-agent: PerplexityBot    Disallow: /
User-agent: Google-Extended  Disallow: /

ChatGPT, Claude, Perplexity, and Google AI Overviews are explicitly told to stay out. The visibility consequence is measurable.

We asked Claude — the language model behind ChatGPT's most direct competitor — three commercial queries a prospect would actually ask:

  1. 01

    Best med spas in Alpharetta

    Seven competing med spas named. SHINE mentioned (general category query).

  2. 02

    Botox and aesthetic injectors near Alpharetta

    Three competing providers named. SHINE not mentioned.

  3. 03

    Medical weight loss programs Alpharetta

    Three competing programs named. SHINE not mentioned.

For the high-intent specific queries — Botox injectors, weight loss programs — where conversion intent is highest and ad spend per click is most expensive, the business does not appear. SHINE is recognized for the general category but invisible for the specific commercial intents that drive revenue.

This is the compound mistake: a security setting applied without awareness of its commercial cost. None of the competitors named above have this block. The visibility asymmetry compounds monthly as more aesthetic medicine research shifts to AI-mediated queries.

The second audience now arriving at every business website is the language models answering questions about your services. A modern foundation serves both audiences. A foundation that blocks one fails both.

The honest read

What is working — and worth preserving.

SHINE has clear positioning in the Avalon district and sophisticated team credentials — Allergan-trained injectors, board-certified providers, PatientFi financing integration. The brand work is correct for the market and the price point. Open Graph and Twitter Card metadata are properly implemented. The business appears for the general “best med spas in Alpharetta” query, indicating the brand has earned its way into Claude's training data. The strategy is sound.

A rebuild is not a rejection of any of this. It is the infrastructure underneath, replaced — and the robots.txt block, lifted — so that the brand work and the team's credentials finally render fast enough to be seen by the prospects already being paid for, and visible to the audiences the brand work was built for.

The bottom line

A brand built for prospects — running on a foundation that hides it from them.

This page, the rebuilt homepage, and every measurement above were built and deployed in under an hour. The same discipline applied to the live property is a two-to-three-week engagement.

EdgeReport is the engagement behind this page: a fixed-fee rebuild of the site, the lead infrastructure, and the programmatic SEO pages on production-grade hosting — fully managed. Performance and GEO measured monthly. The discipline is ours; the business is yours. If the gap above is worth closing, the next step is a thirty-minute conversation.

larry@edgereport.ai