Recover Content From an Old CMS
How to recover useful pages, images and metadata from an abandoned, broken or inaccessible content management system.
Collect every available source
Use the live site if it still works, hosting files, database exports, XML exports, local developer copies, media folders, old sitemaps and web archives where permitted. No single source is guaranteed to be complete.
Prioritize public value
Recover pages that users can still find through search, bookmarks or backlinks first. Important service pages, guides, legal information, downloadable documents and evergreen articles generally outrank abandoned tags or admin-generated archives.
Reconstruct metadata carefully
Titles and descriptions may be stored separately from body content. Canonical URLs may come from plugins or templates. Capture this information before stripping the CMS so the new page does not lose important search context.
Avoid copying contamination
If the old site was compromised, treat executable code as untrusted. Text and media can be reviewed and reused; unknown PHP, JavaScript or server configuration should not be copied into the clean build without understanding it.
Document uncertainty
When a page cannot be fully recovered, mark what is missing. Do not silently invent old dates, authors or claims merely to fill gaps. A smaller accurate archive is more useful than a reconstructed history that cannot be trusted.
Use rendered pages to resolve broken templates
A database export may contain shortcodes, template tokens or raw IDs that are difficult to interpret without the old application. If a rendered version of the page is available, compare it with the stored source to understand what the visitor actually saw before rebuilding the content.
Preserve publication context
Dates, authors, categories and revision history can matter for reports, news and long-form guides. Recover them where they are meaningful, but do not carry them over automatically to ordinary evergreen service pages that never needed blog-style metadata.