Craig Campbell SEO Training & Consultancy

Tutorials

Migrating a website for SEO: preserve content and verify the new site

By Craig Campbell · 16 Nov 2026

7 min read

A website migration can look successful while older articles, photos or enquiry journeys have quietly broken. This guide shows how to carry an established site into a new platform and prove that the important content still works.

It complements the existing migration checklist with a content-preservation and sign-off workflow. The UK shop example is fictional and illustrative only; no client results or traffic-retention claims are invented.

1. Define the move and success

Migrating a website for SEO means carrying useful content and discovery paths into a new setup while controlling what changes. This guide focuses on replacing a content platform without casually changing established addresses. A domain move is a separate decision with additional redirect and verification work.

Write down the current and intended domain, page paths, platform, media location and tracking. Agree who can approve content removal, who owns technical release checks and who can stop the launch. Separate unrelated changes where practical so a problem is easier to diagnose.

  • Same addresses, new platform: corresponding pages should normally load directly at their existing URLs.
  • Changed paths or domain: review a destination for each useful old address.
  • Hosting only: check availability and infrastructure rather than treating it as a domain change.

Success starts with preserved access and working enquiries, not a promise of higher rankings. Use the existing website migration SEO checklist for the broader launch gates; this guide adds a practical content-handoff and acceptance workflow.

2. Capture the content baseline

Before importing anything, keep a recoverable export and a record of the live site. Combine the platform’s published-page list with crawl discoveries, sitemap URLs and available landing-page or server-log evidence. None of those sources alone is a complete historical inventory.

  • Record article and page URLs, titles, headings, text and publication dates.
  • Include categories, pagination, images, captions and downloads.
  • Record existing redirects, canonical signals and language alternatives.
  • List menus, related links and form destinations.
  • Keep observed traffic and enquiries separate from missing data.

Choose important page groups for acceptance checks: services, popular guides, older linked articles and pages needed to navigate the archive. A quiet article is not automatically disposable. The content audit guide explains reviewing useful material rather than deleting by traffic alone.

3. Specify what must survive

Turn the inventory into a handoff list with an expected result for each type of content. “Imported successfully” is too vague if an article has lost its images, date or useful links.

Fictional UK example — illustrative only

Kelvin Trail Supplies is a fictional Glasgow outdoor shop replacing its old publishing platform. The decisions below are invented teaching examples, not Craig’s client work or measured outcomes.

  • Boot care guide: keep its original address, article text and publication date; confirm all existing care photos and captions appear.
  • Training page: keep the address and enquiry wording; submit a test enquiry and confirm receipt through the normal owner workflow.
  • Historic size-chart download: keep the original file path and verify the actual file opens.
  • Superseded service page: require a manually reviewed equivalent before approving a permanent redirect.
  • Old event archive: retain useful information unless removal has been separately approved.

The project cannot be signed off merely because its homepage looks finished. Each retained item needs its own evidence.

Record source, intended outcome, owner and pass/fail evidence. If something cannot be recovered, flag it before release rather than substituting invented text or a misleading destination.

4. Test templates and real pages

Protect the development copy from unintended access. Test both representative templates and individual important pages: a working article template does not prove every imported article is complete.

  • Compare source and destination titles, headings, dates and main content.
  • Check photos load, captions remain attached and downloads open.
  • Follow article links, collection links and pagination.
  • Test long headings, tables and code examples on a phone.
  • Check meaningful content and ordinary links exist in the rendered page.
  • Complete the enquiry journey and confirm the submission reads back or reaches its intended recipient.

The internal linking guide helps check discovery paths. Keep approval records for deliberate differences; do not silently turn a preservation check into a content rewrite.

For search checks, review canonical destinations, indexing directives and structured data on actual pages. Use the schema markup guide to confirm data describes the visible content rather than carrying over inaccurate old output.

5. Plan address changes and release

Do not redirect addresses that can keep serving their corresponding content directly. Where a move is necessary, use a reviewed map to genuine replacements and test the full journey. The redirects for SEO guide explains permanent versus temporary responses, while the chain guide covers unnecessary hops.

Before launch, agree a release window, a content freeze or a reliable way to reconcile late edits, and a recovery plan. Identify which system will continue serving useful old redirects and files. Keep access to the infrastructure the migration still depends on.

  • Confirm public pages use the intended domain in links and canonical signals.
  • Remove development-only indexing restrictions from the public version.
  • Keep restricted or private areas protected.
  • Check sitemaps contain preferred public destinations.
  • Confirm the person approving release can see the acceptance evidence.

For a domain or subdomain move, check whether Google’s Change of Address process applies. Replacing the platform while keeping URLs is not itself a reason to use that tool. The older checklist retains its more detailed explanation.

6. Recheck on the public website

Checks against a development copy do not prove the public release is correct. After the switch, open the real addresses directly and through navigation, then repeat the critical page and enquiry checks.

  • Check retained URLs serve the expected pages.
  • Test old moved addresses through to the relevant final page.
  • Confirm photos and historical downloads remain accessible.
  • Recheck canonical signals, noindex rules and robots access.
  • Verify search measurement still records the intended public pages.

Stop or repair the release when there is a reproducible critical failure, such as missing key content or broken enquiries. Use the technical SEO audit guide to investigate access, rendering and indexing faults. A recovery decision should restore a coherent site and URL map, not leave competing versions half-switched.

7. Monitor without inventing a verdict

Record the actual release time and later significant changes. Review page-level evidence in Search Console, alongside genuine enquiries or sales where measured. Separate technical acceptance from Google’s later crawling and indexing.

Compare like-for-like periods and page groups. Note seasonality, content changes and shifts in search results. A site-wide total can hide a problem in an important service section. Ranking fluctuations can occur during moves, but they do not excuse a page that demonstrably fails to load.

Investigate a specific fault before repeatedly reversing the release. There is no universal timetable for full search processing and no guaranteed traffic-retention percentage. The monthly reporting guide helps turn evidence into the next decision without claiming that every change was caused by the migration.

Checklist and next step

  • Is the scope clear about which addresses actually change?
  • Is there a recoverable baseline and a historical content inventory?
  • Do acceptance checks include articles, images, downloads and enquiries?
  • Are every deliberate removal and redirect separately reviewed?
  • Have critical journeys been repeated on the public site?
  • Is there an owner for post-launch investigation and maintenance?

Start by agreeing the content that must survive and one representative page group to test. A careful migration reduces avoidable mistakes; it cannot guarantee crawling, rankings or future business results.

Sources and review notes

Previously checked Google guidance is reused with the original dates below. The acceptance workflow and fictional shop are editorial recommendations, not a prescribed Google timetable or a claim of firsthand testing by Craig.

Let's work together.

Whether you need an SEO partner, a consultant, a speaker for your event or training for your team — get in touch and let's chat.

Get in touch
Get in touch