Completed work
Recovering a legacy content archive
There was no neat export. The archive still mattered. Discover the old content, identify what is usable and reconstruct it in a workable format.
The problem behind the job
If years of useful content are trapped in an old or broken site, the job is to recover what can be recovered without publishing duplicates, dead assets or material you cannot verify.
What I was working with
A publishing archive needed to be recovered from an older website structure. The source was not a tidy collection of modern CMS records. I worked through possible old URL patterns, identified valid pages and reconstructed the content while preserving the relevant article details.
The first problem was finding the pages
Before the articles could be rebuilt, the valid old URLs had to be discovered. That required an investigation of the source structure rather than assuming the visible navigation was a complete inventory.
The discovery work was part of the migration, not something that could be skipped before the import.
Keep the article information together
The recovered material included the article content and associated dates, images and captions. Those relationships mattered because the task was to rebuild an archive, not merely collect paragraphs of text.
An image copied without its connection to the article would leave another repair job behind.
Reconstruct the records
The valid content was turned into structured records for the new site. The sourcebook records the archive reconstruction as delivered.
This is an anonymised account of the method and the work. I'm not reproducing the original publisher's URLs, articles or identifying project totals.
Is there useful content trapped in an old website?
If your problem looks similar, send me the current setup and the outcome you need. The implementation may be different, but the same kind of thinking usually applies.
What the result does and does not say
The result was recovered and rebuilt content. That is distinct from claiming every old asset was recoverable or that the move produced a particular traffic increase.
The source needs to support those wider claims separately.
What I'd establish first for another archive
I'd look at the surviving source, available access, known URL patterns and the information that needs preserving. A migration inventory gives that work a clear target.
For missing media relationships after a move, see image recovery. The wider service is website migration and recovery.
HAVE SOMETHING AWKWARD?
Is there useful content trapped in an old website?
Send me the URL, what is happening, what you have already tried and what a useful result would look like. A small anonymised sample is enough to start - do not send passwords, credentials or sensitive customer data.