A one-stop EHS shop the machines can't read yet.

Your site isn't broken, most aren't this clean. Across 12 rendered loads: zero console errors, zero failed requests (one harmless autoplay warning aside). Proper meta description, full Open Graph, published sitemap.xml.

The substance is real: 35 years, a roster from NASA and the US Mint to PG&E, Stanford, Kaiser, Tesla and Intel, and director Jay Jamali, CSP/CHMM/CHCM, 40+ years, a national ASSP presenter.

So the findings below aren't about what the site does badly. They're about what a Weebly brochure can't do for a firm this deep.

Machine-readable credentials → a 35-year firm becomes the AI's first answer for California EHS, not off the shortlist. Procurement is moving to AI-assisted search, and with nothing structured to read, a buyer's assistant can't surface Enviro Safetech even though it's the deepest option, the fix puts you in the answer.

One Services hub → "we do a lot" becomes a catalog a buyer or an AI can actually list. The 25 scattered service pages have nothing to anchor on today, so the breadth that wins work stays invisible; one legible page makes the full range findable.

Lead with 2026, archive the COVID posts → diligence sees the active firm you are. A national talk this year proves the firm is thriving, but a buyer or AI doing homework sees content frozen around 2020 and reads "paused", freshening the front page stops that misread from costing a deal.

A short form on the staffing and consulting pages → your biggest deals stop funneling through one inbox. The highest-margin lanes route to a single mailto today, so they arrive unqualified and bottlenecked; a two-minute intake captures site, dates, and certs and lets the top deals scale past one person.

Ask an AI to "find a California consultant who can write a Cal/OSHA respiratory program, run fit-testing, and staff a site safety officer", and it finds no structured answer here. Zero JSON-LD blocks on the homepage and every inner page: no Organization, LocalBusiness, Service, Person, or Course markup. No llms.txt.

$ curl -s envirosafetech.com | grep -c 'application/ld+json'
0
$ playwright: document.querySelectorAll('script[type="application/ld+json"]')
[]   # on all 6 pages
$ curl -s -o /dev/null -w '%{http_code}' envirosafetech.com/llms.txt
404

Procurement is moving to AI-assisted search. A firm this deep should be the machine's first answer for California EHS, but with nothing machine-readable, it's off the shortlist. Cheapest to fix, and it compounds.

The nav labels ABOUT US / SERVICES / ONLINE TOOLS are dropdown-only, no landing page. Their old hub pages now 404, yet robots.txt spends its only rules blocking those four dead URLs. So your ~25 service pages have no single page saying "here's everything Enviro Safetech does", nothing for an AI to anchor on.

$ curl -s envirosafetech.com/robots.txt
User-agent: *
Disallow: /about-us.html   Disallow: /services.html
Disallow: /training.html   Disallow: /online-tools.html
$ for p in about-us services training online-tools; do
    curl -s -o /dev/null -w '%{http_code} ' envirosafetech.com/$p.html; done
404 404 404 404

One Services hub, plus the Agent Kit's schema, turns scattered pages into one legible catalog for Google and AI. The difference between "we do a lot" and a machine being able to list it.

The Safety Blog stops at 2015, every blog URL in the sitemap is a 2013–2015 topic. The homepage "Company Updates" leads with COVID-era N95/KN95 tips, a March-2020 simulator announcement, the 2022 "Goes Green" post, a June-2023 LOTO event. The one current item, Jay at #Safety2026, is buried among them.

sitemap safety-blog/* newest topics: GHS Dec-1 deadline, 2014 Cal-OSHA
   heat advisory, ASSE 2014/2015, 2015 NFPA 70E ...  (nothing after 2015)
homepage updates: "N95 /KN95 COVID", "March 9, 2020 ... Mobile Confined
   Space Training Simulator", "10/12/22 ... Goes Green", "June 3, 2023 LOTO"

You're active; a national talk this year proves it. But a buyer or an AI doing diligence sees a firm that looks paused. Lead with 2026, archive the pandemic content, and the firm looks as current as it is.

Credit where due: the training and quote lanes have real forms (get-a-quote, safety-services-quote, training-quote, quick-safety-audit). But the highest-value lanes, Safety Staffing, On-Site Safety Officer, Consulting, Safety Programs, have no structured intake. The only CTA is mailto:[email protected]. Nothing captures what dispatch needs: site, dates, certs (EM 385, Cal/OSHA), industry, wage class.

$ for p in staffing site-safety our-expertise programs; do
    curl -s envirosafetech.com/$p.html | grep -c wsite-form-container; done
0 0 0 0
staffing / site-safety / consulting CTA:  mailto:[email protected]
(training-quote, get-a-quote, quick-safety-audit DO have Weebly forms)

A two-minute form on the staffing and consulting pages turns "email Jay and wait" into "we already know your site, dates, and certs", and stops routing your biggest deals through one inbox.

No CSP, HSTS, X-Frame-Options, Referrer-Policy or X-Content-Type-Options headers (Weebly-on-Cloudflare default; read from public headers only, nothing tested, not a vulnerability claim). The hero video autoplay throws a NotAllowedError every load. And analytics loads from ssl.google-analytics.com, the legacy Universal Analytics host Google retired in July 2023, so measurement may be silently off.

$ curl -sSI envirosafetech.com | grep -iE 'strict-transport|content-security|x-frame|referrer-policy'
(empty)
homepage pageerror: NotAllowedError: play() failed because the user didn't interact...
third-party host: ssl.google-analytics.com  (legacy UA endpoint)