We Audited 7 Puerto Rico Law Firm Websites. Here's What We Found.
By Archie Cortés, Founder of AutoPilotPR. We build AI-native marketing systems for Puerto Rico businesses. Every audit we publish uses real tools on live sites. Methodology is at the bottom.
Every week, a Puerto Rico law firm pays for Google Ads while their website fails the first impression test. A potential client clicks the ad on their phone, waits 11 seconds for the page to load, and calls the next firm on the list. The first firm never knows.
We wanted to know how widespread this problem is. So we audited 7 Puerto Rico law firm websites in September 2026. Not subjective opinions. Real Lighthouse scores, real schema checks, real response times, run from a server in Puerto Rico.
Here's what we found.
Table of Contents
- How We Ran the Audit
- The Full Results
- What the Data Shows
- The Schema Problem
- What a High-Scoring Law Firm Site Looks Like
- The Self-Audit Checklist
- Resumen en Español
How We Ran the Audit
Lighthouse (mobile performance): We ran npx lighthouse v13.4.1 using a headless Chromium browser with mobile preset. This is the same tool Google uses to calculate Core Web Vitals. Scores range from 0 to 100; Google labels 0 to 49 "Poor," 50 to 89 "Needs Improvement," 90 to 100 "Good."
Schema markup check: We fetched each site's HTML source and checked for application/ld+json structured data blocks and any LegalService, Attorney, or LawFirm schema types. This tells AI systems and search engines what the site is and what it does.
Response time: We measured TTFB (Time to First Byte) and total load time using curl from our Puerto Rico server. This is raw server response speed, separate from the browser rendering time Lighthouse captures.
Sites audited: 7 publicly-accessible Puerto Rico law firm websites, selected from Chambers & Partners, Legal 500, and Martindale-Hubbell Puerto Rico rankings. Two additional firms were excluded: one was unreachable at audit time, one was a global firm site that rate-limited our request. Audit date: September 4, 2026.
Raw data is in the repo: data/teardowns/law-firms-2026-09/audit-results.json: every score, timestamp, and methodology note is there for verification.
The Full Results
| Firm | City | Mobile Score | LCP | Layout Shift | ld+json | Legal Schema |
|---|---|---|---|---|---|---|
| Firm A | Hato Rey | 69 | 8.5s | 0.10 | No | No |
| Firm B | San Juan | 57 | 9.1s | 0.00 | Yes | Yes |
| Firm C | San Juan | 56 | 6.6s | 0.33 | Yes | No |
| Firm D | San Juan | 60 | 10.9s | 0.23 | Yes | Yes |
| Firm E | San Juan | 66 | 13.5s | 0.00 | Yes | Yes |
| Firm F | San Juan | 56 | 6.3s | 0.00 | No | No |
| Firm G | San Juan | 59 | 11.5s | 0.00 | No | No |
Naming convention: We audit only publicly observable facts. We do not name a firm alongside a negative finding. The identifiers above are consistent across all references in this post, so a reader who wants to look up Firm A can do so by cross-referencing the raw data file.
AutoPilotPR reference: Our own homepage (/web-design-puerto-rico) scored 83 in the same Lighthouse run configuration. That's the benchmark we're measuring against.
What the Data Shows
Median mobile Lighthouse score: 60. Range: 56 to 69.
Every single site we audited falls in Google's "Needs Improvement" category. None scored above 75. The highest score in the group, 69, still has an 8.5-second LCP. Google's threshold for "Good" LCP is 2.5 seconds.
LCP (Largest Contentful Paint) is the common failure point. This measures how long it takes for the main content of the page to appear on screen. Four of seven sites had LCP above 10 seconds on mobile. The fastest LCP in the group was 6.3 seconds. Google classifies anything above 4 seconds as "Poor."
What does a 10-second LCP mean in practice? A potential client in Ponce, searching for an attorney on their phone during lunch, waits more than 10 seconds before they can even read the firm's name. Most won't.
53% of mobile users abandon a site that takes longer than 3 seconds to load (Google, 2018). That number gets worse every year as user expectations rise.
Layout shift is a real problem on two sites. CLS (Cumulative Layout Shift) measures whether page elements move around while loading. A score above 0.1 is considered poor. Two firms scored 0.23 and 0.33. Content is content is visibly jumping around as the page loads. On a legal site, where the first impression of professionalism matters, this is a credibility problem as much as a technical one.
The best performer in this group (score 69) still has an LCP of 8.5 seconds (three times Google's "Good" threshold times Google's "Good" threshold). There is no firm in this audit that would pass a Core Web Vitals assessment.
The Schema Problem
4 of 7 firms have no LegalService or Attorney structured data on their homepage.
Schema markup is how your website tells Google, ChatGPT, and other AI systems: "This is a law firm. This is the practice area. This is the location. This is the phone number." Without it, AI systems have to guess, and they often guess wrong or leave you out entirely.
When someone asks ChatGPT "who are the best employment attorneys in San Juan," it pulls from a combination of review platforms, directory listings, and websites that have clear structured data. A firm without schema is invisible to that query, no matter how good its reputation is.
3 of 7 firms had ld+json blocks but without legal-specific schema types. Having generic structured data is better than nothing, but it does not communicate the practice-area specificity that AI systems use for recommendation queries.
Only 31% of websites have any structured data at all (Web Almanac, 2022). In a legal market where most firms are starting from zero, implementing LegalService schema correctly is a real differentiator.
What a High-Scoring Law Firm Site Looks Like
For context, here's the performance profile of an AutoPilotPR-built site:
- Mobile Lighthouse score: 83
- LCP: under 3 seconds
- CLS: 0.00
- Schema:
LegalService,LocalBusiness,Person(attorney profiles), FAQ schema on practice area pages - Response time (TTFB): under 400ms
The gap is not a mystery. High-scoring sites use modern frameworks (Next.js or similar) with static generation so pages load from CDN edge nodes rather than origin servers. Images are compressed and served in WebP format. Fonts load in parallel rather than blocking render. None of this requires a custom-built site from scratch. These are standard practices that any competent web developer can implement.
The reason most law firm sites in Puerto Rico don't meet these standards is that they were built on WordPress or legacy CMSes in 2018 to 2021 and have never been performance-audited. The design may have been updated, but the underlying architecture is slow.
The Self-Audit Checklist
You can run a version of this audit on your own firm's website in 20 minutes. Here's how:
Step 1: Lighthouse Score Go to PageSpeed Insights, enter your URL, and run the mobile test. Look at: Performance score, LCP, CLS, Total Blocking Time.
Step 2: Schema Check
Go to Google's Rich Results Test, enter your URL. If you see no results, you have no schema. If you see results, check whether LegalService or Attorney appears.
Step 3: Core Web Vitals in GSC If you have Google Search Console access, go to Experience > Core Web Vitals. Any pages marked "Poor" or "Needs Improvement" are hurting your rankings.
Step 4: Mobile Load Test Open your website on your own phone on a 4G connection (not WiFi). Count the seconds until the main content appears. If it takes more than 4 seconds, so does it for every potential client.
Step 5: AI Visibility Check Open ChatGPT or Perplexity and type: "best [practice area] attorney in [your city] Puerto Rico." Does your firm appear? If not, you are invisible to AI-referred clients: a segment growing every month.
What to do with the results:
- Score below 60: Architecture is broken. WordPress plus slow hosting will not fix with theme changes. A rebuild is likely necessary.
- Score 60 to 75: Addressable with performance optimization (image compression, font loading, server upgrade) without a full rebuild.
- Score 75+: Schema and AI visibility are probably the bigger gaps. Add
LegalServiceschema and claim all directory listings. - Not cited in ChatGPT: Structured data, directory presence, and review recency are the three levers. All three are achievable in 30 to 60 days.
Resumen en Español
Auditamos 7 sitios web de bufetes de abogados en Puerto Rico en septiembre de 2026 utilizando Google Lighthouse (rendimiento móvil), análisis de schema markup y tiempos de respuesta del servidor.
Hallazgos principales:
- La puntuación mediana de Lighthouse móvil fue 60 de 100 (rango: 56 to 69). AutoPilotPR obtiene 83 con la misma metodología.
- Todos los sitios auditados caen en la categoría "Necesita mejora" de Google. Ninguno aprobó el umbral de rendimiento "Bueno".
- El LCP (tiempo hasta que aparece el contenido principal en pantalla) superó los 10 segundos en 4 de los 7 sitios. El umbral "Bueno" de Google es 2.5 segundos.
- 4 de 7 firmas no tienen schema markup de
LegalServiceoAttorney. Esto significa que los sistemas de IA como ChatGPT o Perplexity tienen dificultades para identificar y recomendar estas firmas en búsquedas locales.
Lo que puede hacer ahora:
- Pruebe su sitio en PageSpeed Insights desde un dispositivo móvil.
- Verifique si tiene schema markup en Google Rich Results Test.
- Busque su firma en ChatGPT y Perplexity para verificar visibilidad en búsquedas de IA.
Si su puntuación está por debajo de 70 o no aparece en resultados de IA, contáctenos para una auditoría gratuita. En 90 días, le garantizamos citaciones en búsquedas de IA para consultas relevantes en Puerto Rico, o el servicio Autopilot es gratuito hasta que lo logremos.
Methodology note: Lighthouse scores were run using npx lighthouse v13.4.1, mobile preset, headless Chromium, September 4, 2026. Scores can vary by ±5 points between runs due to network conditions and CDN caching. Schema analysis used source HTML inspection. Raw audit data: data/teardowns/law-firms-2026-09/audit-results.json. We accept no payment from any firm audited. No firm was contacted before or after publication.
Want to see where your firm stands? Get a free website audit and we'll run the same Lighthouse and schema analysis on your site, plus check your AI citation status across ChatGPT, Perplexity, and Google AI Overviews.