Businesses move to Drupal from many places: another content management system, a site builder that served its purpose, a custom application, a collection of static pages, sometimes a spreadsheet and a folder of documents. Whatever the starting point, the concern is the same: years of content, URLs that search engines know, and editors who have work to do tomorrow. This article explains how we move content into Drupal so that everything that matters carries over, nothing valuable is lost, and the team keeps working throughout.
First, an inventory of what exists
Before anything moves, we need to know what there is. We export or crawl the existing site and produce a complete list: every page, post, product, file and image, with its URL, its type, its language, its author, its dates and its relationships to other items. This list is usually longer than anyone expected and includes content nobody has looked at in years.
The inventory drives a conversation about what should come across. Not everything deserves to. Outdated pages, duplicates, drafts that were never published and content for products that no longer exist are candidates for archiving rather than migrating. The best migrations are also a content cleanup, and the inventory is where that decision is made deliberately rather than by default.
Mapping the old shape onto the new
The heart of a migration is a mapping: for each kind of content in the old system, what it becomes in Drupal, field by field. A blog post becomes an article, its title to the title, its body to the body, its categories to tags, its featured image to the image field, its author to a user account, its publication date preserved. A product becomes a product, with prices, specifications and images landing in the fields that hold them.
This is where content modeling earns its keep. Because Drupal stores content as structured fields, the mapping is precise, repeatable and testable. It is also the moment to improve the structure: a field that was a free-text label in the old system can become a proper taxonomy; information that was buried in a page body can be extracted into its own field. Content often arrives in Drupal better organized than it left.
Migrating with a pipeline, not by hand
Drupal includes a migration framework built for exactly this work. A migration is defined as a set of rules: where the source data comes from, how each field is transformed, and where it lands. The rules are code, committed with the site, and they can be run, checked, corrected and run again as many times as needed. This matters because the first run is never perfect. A date in an unexpected format, a character encoding issue, an image that was linked rather than uploaded: each is found on a test run, fixed in the rules, and the migration is rerun from scratch.
Because the pipeline is repeatable, the live site keeps working right up to the switch. We migrate into a staging environment, review, refine, and rerun. On launch day, a final run picks up anything created or changed since the last one, and the new site opens with content that is complete to the minute.
Preserving what search engines know
Every URL on the old site that has ever been linked to or indexed is an asset. A migration that changes URLs without redirecting them throws that asset away. We treat the URL map as part of the migration: where a URL can stay the same, it does; where it changes, a permanent redirect from old to new is created automatically as part of the pipeline. Search engines follow the redirects and transfer the standing of the old page to the new one. We covered the mechanics when we wrote about changing URLs without losing traffic; a migration is simply that discipline applied to every page at once.
Files, images and formatting
Content is more than text. Images, documents and embedded media need to move too, into Drupal's media library, with alternative text preserved where it existed and flagged where it did not. Body text carries formatting and internal links that reference the old site's structure; we rewrite those links to point at the new locations and clean formatting so it fits the new site's text formats. This step is where migrations accumulate small details, and the pipeline's repeatability is what makes those details manageable: fix the rule once, rerun, and every instance is corrected.
Editors keep working
A migration is not a content freeze. Editors continue publishing on the old site until the switch, and the final migration run brings their work across. In parallel, they get access to the staging site to learn the new editorial interface with their own content in it, which is a far better training environment than sample data. By launch day, the team has already used the new site for weeks, and the switch is a change of address rather than a change of habit.
Verification before the switch
We do not launch on trust. Counts are compared: this many articles in the source, this many in Drupal. Samples are read side by side. Every redirect is tested. Search is checked for the terms people actually use. Forms, integrations and any logged-in functionality are exercised on staging with real workflows. The pre-launch checklist runs in full. Only then does the domain point at the new site, and the old one stays available, privately, for a period afterward in case anything needs to be checked against it.
If you are considering a move to Drupal and the volume of existing content is what gives you pause, that is exactly the part with the most mature tooling and the most predictable process. Tell us what you have, and we will tell you what moving it would involve.
