1. Start with the public surface
The first validation should behave like a real user and a real crawler. Open the home page, deep links, static assets, sitemap, robots, ads.txt and any page that search engines or ad platforms rely on.
This catches common SPA fallback, proxy and routing mistakes before they become indexing or monetization problems.
- Open the home page and key internal routes directly.
- Check sitemap.xml, robots.txt and ads.txt.
- Validate canonical and language alternates after major content changes.
- Confirm that protected admin or internal routes are not exposed.
- Use logs to verify which service handled the request.
2. Validate forms, APIs and email delivery
A site can look alive while its most important conversion path is broken. Contact forms, lead capture, API health and email notifications should be tested after every relevant deploy.
When campaigns are active, preserve source information such as UTM parameters from the first page view to the backend notification.
- Submit a real contact form test.
- Check backend health and API logs.
- Confirm SMTP sender, Reply-To and destination mailbox.
- Verify UTM capture when testing campaign landing pages.
- Review error logs immediately after the test.
3. Check rollback and neighboring systems
A VPS often hosts more than one system. A deploy for one domain should not break another project sharing the same Nginx, Docker host or network resources.
A professional post-deploy routine includes rollback awareness and a quick check of neighboring systems.
- Know which image, commit or package can be restored.
- Check that other domains on the VPS still answer correctly.
- Confirm container names, ports and networks did not collide.
- Review disk space after image pulls and logs.
- Record the validation result in a short deploy note.
How to use this article
Treat this page as a decision aid. Use it with the related hub, checklist or service route when the topic affects production, customer experience, deployment, security or business continuity.