Proof of concept
This website is our first case study.
Performance
100
Accessibility
100
Best practices
100
SEO
100
Lighthouse mobile scores from our pre-launch build (October 2026): LCP 1.2 s, CLS 0, TBT 0 ms. Measure the live site yourself: run PageSpeed Insights on omglaunch.com. We’ll add our rankings and AI share-of-voice timeline here as data comes in.
Technical foundation
-
✓ Static HTML, zero client JavaScript
Every word is in the initial HTML. AI crawlers that don’t run JavaScript (most of them) see 100% of the content.
-
✓ Inlined CSS & one preloaded font
No render-blocking requests. A single 27 KB variable font with a metric-matched fallback for near-zero layout shift.
-
✓ Immutable asset caching
Hashed assets are cached for a year; HTML revalidates on every visit so updates appear instantly.
-
✓ Clean, consistent URLs
Trailing-slash canonical URLs, one hreflang (en-MY), XML sitemap and RSS feed.
Answer & AI readiness
-
✓ Direct-answer blocks
Every service page opens with a 40–60 word definition: the passage most likely to be quoted.
-
✓ Question-led headings
H2s and FAQs phrased as the questions buyers ask, answered immediately underneath.
-
✓ Extractable formats
Comparison tables, numbered processes and definition lists that LLMs can lift cleanly.
-
✓ llms.txt & llms-full.txt
A curated plain-text map of the site for AI systems, generated from the same data as the pages.
-
✓ Open AI crawler policy
robots.txt explicitly welcomes GPTBot, OAI-SearchBot, ChatGPT-User, PerplexityBot, ClaudeBot, Google-Extended and more.
Entity & trust signals
-
✓ Connected JSON-LD graph
Organization, ProfessionalService, WebSite, WebPage, Person, Service, FAQPage, HowTo, OfferCatalog, Article, BreadcrumbList and DefinedTermSet — linked by stable @ids.
-
✓ Consistent NAP
Name, address and phone come from one data file, so they’re identical on every page, in schema and in llms.txt.
-
✓ Named methodology
The LAUNCH Method gives people and machines a clear, repeatable description of what we do.
-
✓ Honest claims only
No invented statistics, testimonials or guarantees. AI systems increasingly cross-check — so do buyers.
Inspect it yourself
The same approach, applied to a person rather than a company: our founder’s own site, benyeoh.com, is built on one canonical person entity that this site and ProfileArchitect both reference.
Want the same for your site?
Get a free 10-prompt AI visibility snapshot across ChatGPT, Gemini, Perplexity and Google AI Overviews — plus three fixes you can make this week.