Create a record of the website that exists
Inventory the important pages, media, metadata, navigation and live functions. Record current URLs and any useful search or enquiry data available. A screenshot of the home page is not a sufficient backup of an operating website.
Ask who controls the domain, hosting, source code, CMS and external integrations. Access problems are easier to resolve before the old environment is dismantled. Keep backups and agree who can authorise the switch.
Map content and functions to the intended replacement
For each important page, decide whether to retain, improve, merge or retire it. If the URL changes, identify the most relevant destination. Avoid sending every old address to the new home page merely because that is easy.
List the purpose of each plugin or integration. A platform change does not transfer a booking system or form backend automatically. Decide what stays external, what is rebuilt and what is deliberately removed, with a clear owner for the decision.
Build a testable review version
Compare the proposed website with the agreed inventory. Read the actual service information, check images and downloads, follow internal links and review a mobile enquiry journey. Verify language routes and canonical settings when multiple markets are involved.
Keep the review environment from being mistaken for a second production website. The technical approach to indexing and access should be deliberate, and those settings must be checked again for the intended public environment at launch.
Treat delivery as an end-to-end function
Submit a permitted test enquiry and confirm what the recipient actually receives. Distinguish a frontend confirmation, backend acceptance and inbox delivery. Check error behaviour as well as the successful route.
Review analytics or other measurement only where it is intentionally configured. A new design should not silently discard useful consent handling, event definitions or access to the accounts needed for follow-up. Do not invent historical data to fill a gap.
Agree the switch, recovery and follow-up
The launch plan should name the people responsible for the domain or routing change, content freeze, deployment and immediate checks. Retain a usable backup and define what would trigger a recovery action.
After the move, review important URLs, redirects, enquiry routes and available search signals. Search engines need time to process changes, so fluctuations should be investigated in context. No migration plan can honestly guarantee that every ranking remains unchanged.
- Retain a source and content backup.
- Review page-by-page URL mappings.
- Test key functions and failed submissions.
- Check production indexing and canonical settings.
- Confirm editorial access and publication.
- Record the actual launch version and responsible owners.
Make URL decisions a separate delivery record
A migration plan becomes more useful when old addresses have documented outcomes. Before the whole move, choose a small sample: a service keeping its URL, a guide changing address and a page being removed. Record why each destination suits the visitor and who makes the content decision. Test the rules in the environment that will actually apply them. A working new page does not show what happens to an old link. Ask a second person to follow the sample from its original address to the content and record any chains. Use acceptance testing for approval and the redirect map for the address decisions.