PatchOps has 11 detailed articles, including a 12-minute HubSpot + NetSuite guide, but the site sends their content only through JavaScript. A reader that does not run JavaScript gets the same document at every address: the homepage, each article, /pricing, /careers and even a made-up URL all return only the title "PatchOps | AI Enablement & Custom Tooling", with no body text, no canonical tag and no article-specific description or preview tags. Even after rendering, articles keep the homepage title, and the CPQ comparison article rendered empty. Link previews on LinkedIn, Slack and email, and most AI-assistant fetchers, read the raw page, so every shared article looks like the homepage and has nothing to quote. A search for an article's exact title finds nothing from PatchOps.
First action
Fetch one article with JavaScript off and list the tags it needs: title, description, canonical, og:title, og:description, og:image.
How to measure it
Articles indexed under their own title
- Now
- –
- Target
- 11 of 11
- By
- 30 days
Steps
Confirm the gap
Fetch the homepage, three articles and a made-up URL with JavaScript off; save the output as the before picture.
Head of Technical 1 hour
Pick the method
Build-time prerendering of the 12 content routes, or server rendering in the existing Express app.
Head of Technical half a day
Add per-page tags
Title, description, canonical, Open Graph and Twitter tags, author and date for every article.
Head of Technical 1 day
Add structured data
Article schema on posts and Organization schema on the homepage.
Head of Technical 2 hours
Return real 404s
Unknown addresses get a 404 status instead of the homepage shell.
Head of Technical 1 hour
Fix the empty article
Find why the CPQ comparison rendered empty and republish it.
Head of Technical 2 hours
Verify previews
LinkedIn Post Inspector and a non-JavaScript fetch for every URL.
Head of Delivery 2 hours
Resubmit and track
Resubmit the sitemap in Search Console and track indexed pages and impressions per article.
Head of Technical weekly
Why it works
Serve real HTML for every route from the existing Express app: prerender the homepage and the 11 articles at build time (or render them on the server), give each page its own title, meta description, canonical tag, Open Graph and Twitter tags, author, date and Article structured data, and return a real 404 for unknown addresses. Then check each URL with a non-JavaScript fetch, the LinkedIn Post Inspector and Google Search Console's URL inspection, and resubmit the sitemap.
PatchOps already has the expertise on the page; the fix makes it readable to every crawler, assistant and link preview. It is a one-time engineering change that improves every article, current and future, and turns each LinkedIn share into a proper card with the article's own headline.
Ready to use
ChecklistPer-page HTML checklist
[ ] Unique <title> per article (article headline | PatchOps) [ ] Meta description written for the article [ ] Self-referencing canonical [ ] og:title, og:description, og:image, og:url [ ] Author and published/updated dates in the HTML [ ] Article JSON-LD; Organization JSON-LD on the homepage [ ] Full article text present with JavaScript off [ ] Unknown URLs return 404 [ ] Sitemap with lastmod, resubmitted [ ] LinkedIn Post Inspector shows the article card
Risks and how to handle them
Prerendered and live pages drift apart
Generate the HTML in the build pipeline, so every publish refreshes it.
Old shares keep the cached preview
Refresh each URL once in LinkedIn's Post Inspector.
Evidence
- Your websitePage source for readers that do not run JavaScript (checked twice, November 19, 2026): the homepage, /blog/hubspot-netsuite-integration, /blog/ai-document-extraction, /blog/hubspot-cpq-comparison, /pricing, /careers and a made-up address all return the same document whose only text is the title "PatchOps | AI Enablement & Custom Tooling": no body text, no headings, no canonical tag and no article-specific description or Open Graph tags. Rendered with JavaScript, articles still carry the homepage title, and /blog/hubspot-cpq-comparison rendered empty.Source
- Your websiteSitemap: 14 URLs: the homepage, 11 articles, privacy and terms; no lastmod dates; no service, industry, case-study or about pages. Articles are tagged AI Enablement, Custom Tooling, Technical Guide, CPQ Deep Dive, Integration and Best Practices, run 3 to 12 minutes and have three authors; topics include HubSpot + NetSuite integration (12 min), HubSpot Data Hub vs ETL tools, QuotaPath to HubSpot, private apps, custom UI extensions, workflows at scale and when HubSpot's native CPQ isn't enough.Source
- Search resultsWeb search (November 19, 2026): the query "patchops" returned only developer documentation for similarly named code objects (Scala library classes, a 3ds Max scripting interface, AWS API Gateway's "PatchOperation"), with no result from patchops.io; "site:patchops.io" returned no pages from the site; neither the exact title of the newest article nor a search of HubSpot's ecosystem marketplace returned a PatchOps result.Source
- Your websiterobots.txt: "Content-Signal: search=yes,ai-train=no,use=reference" and "Allow: /" for all agents, with "Disallow: /" for named training crawlers such as GPTBot, ClaudeBot, Google-Extended, Applebot-Extended and CCBot; search and assistant fetchers are not singled out, so they are allowed.Source