Who should do the full site replacement and 301 redirect first when multiple site groups are revised together

📅 October 26, 2026 👁️ 0

When a group of multi-site versions is updated, the 301 redirect is performed first, followed by the full site replacement. The reason is not habit, but rather that the two actions modify different objects: the 301 redirect modifies the URL ownership relationship, which needs to be established within the time window when the old address can still be accessed and the new address has already gone live; the full site replacement modifies the keywords in the text and link text, which belongs to the content layer. If the order is reversed, the replacement will write the new path into the text, while the redirect table is still empty, and the search engine will see both the old address and the new address in the same round of crawling, causing the inclusion to fluctuate back and forth.

The two actions modify different objects.

The redirection solves the problem of 'who owns this address now'. If the old address is deleted instead of being redirected, all external entries referencing it - search results, external links from other sites, user favorites - will end up on an empty page, and the accumulated weight from history will also be lost. The redirection table is a one-to-one correspondence between the old URL and the new URL, belonging to the site structure layer.

The website replacement solution is 'How to change that sentence on the page.' Navigation copy, product names in the text, and internal link text scattered in many documents, it is not realistic to change them one by one manually. AnQiCMS provides a website replacement tool for batch keyword or link replacement, covering a batch of documents in one task. It changes the content data, but does not change the address structure, so it will not replace the old links with the corresponding new links for you.

The difference between the two is that redirection is responsible for external references, and replacement is responsible for the reader's reading. The revision needs to be done at the same time, but the order is fixed.

Determine the URL rules first, then move the content.

The content of the redirect table depends on what the new address looks like, so the new rules must be established first. The static rule management is responsible for this: determining the path forms of category pages, document pages, and tag pages, confirming the file extension, hierarchy, and whether to include a trailing slash. Once the rules are switched, the internal link text written in the site-wide replacement has a target address to refer to.

This common mistake is to change rules while publishing articles. The new rules are not finalized and the main text is replaced, then the rules change again, and all the links previously replaced point to non-existent addresses, resulting in a larger amount of rework than not doing anything. This is more obvious in a multi-site environment: a set of rules is applied to multiple sites at the same time, and errors will occur together.

The redirect table must be set up before the switch.

It is recommended to establish a fixed order of four steps:

First step, export the old addresses in full, group them by category, mark which correspond to pages in the new structure, which should be merged, and which are ready to be retired.

Step two, when the new template is deployed but the traffic has not been cut yet, add the redirect table to the 301 redirect management, verify it once with a small number of addresses to confirm that the returned status code is a permanent redirect and not a temporary redirect.

Step three, cut the traffic, observe the crawling and collection logs, and confirm that the requests for the old addresses are all received.

Step four, do a full site replacement, change the old copy and the internal link text in the body at one time, and check several pages after replacement to see if there are any incorrect replacements.

The benefit of redirecting before replacing is that the old address can always reach the content at any time, and the switch is gradual rather than cliff-like.

Inspection points for batch verification of multi-site verification

Multi-site management allows a single backend to manage multiple sites, which brings a unique trap of the redesign: when five sites are cut and observed simultaneously, it is impossible to determine which site's rules are incorrect when a problem occurs.

The batch approach is to queue sites. First, choose the site with the smallest traffic to redesign, complete the jump table and full-site replacement, and then wait for a round of crawling cycle to pass. Confirm that the inclusion and new address are correct, and then push the next site. After each site is redesigned, leave three checkpoints: the return status code of the old address, whether the new address renders normally, and whether the address in the site map has been changed to the new rules.

Replace tasks need to confirm an additional matter: does the replacement range exceed other sites? The full-site replacement is executed based on conditions, and if the conditions are written broadly, it is easy to change the content of unrelated sites. It is safer to run a small-scale trial on a single site first compared to executing it across the entire site.

Sequential reference table

Link Action object Position in the revision Manifestation of problems
Static rules URL form First draft Internal link points to a non-existent path
301 redirect URL attribution Configure before cutting traffic The old address falls to an empty page
Replace site-wide Main text and link text Execute after traffic cut Copywriting is inconsistent with the new structure
Batch verification Single site Check after each site is executed Cannot locate the specific site of the problem

Common questions

Should we back up the old address before the revision?Yes. The full site replacement is an irreversible batch write, so make a backup before the revision. If it goes wrong, it can be restored directly, which is much faster than rolling back article by article.

Can the jump be made for the old address that has been invalid for a while?Yes, but you need to confirm first that these addresses still exist in external links and search caches. The jump is only meaningful for addresses that are still being visited, and adding them for addresses that no longer are does not create value.

Can we only replace and not jump?It only applies to the main text. Not creating a jump when the address changes is equivalent to leaving all external references in place, with the old page going from 200 to 404. The time saved by replacing it will be doubled in the loss of inclusion.

What should be done when the rules of multiple sites are inconsistent?Set up jump tables separately and verify them. The unified management of multiple sites is the backend entry, not the address rules. Forcing inconsistent rules to merge will make it harder to trace.

Related articles

How to avoid simultaneous online launch of scheduled releases for multi-site groups

When multiple site groups share a set of scheduled release schedules, the impact of simultaneous online launch falls on the crawling and indexing queues: at the same time, a large number of page changes occur, the marginal benefits of crawling decrease, and the inclusion pace actually slows down. Peak-shaving is divided into three steps: group by site to open up time windows, give up the whole points, and reserve time slots for revision and backup; the order of priority is arranged according to the impact of inclusion, not according to the size of the site, and the status of pending release is used to carry the schedule instead of as a draft.

2026-10-26

Import in bulk using both zip packages and tables. When images and fields do not match, check which side first

When importing in bulk, check the table side first and then the image side: if fields do not match, it is mostly because the column names and custom fields are not aligned, and if images are missing, it is mostly because the file paths in the zip package do not match the names referenced in the table. AnQiCMS supports ZIP package and Excel imports, the import interface can be queried by title to check if it already exists, and it should be used to judge before repeating the execution

2026-10-26

How to arrange the order of sensitive word filtering and manual review before the collected content goes live

The order before the content goes live is: first draft, then sensitive word filtering and keyword replacement, followed by manual review, and finally the formal document is changed. The rule layer is responsible for determinable issues, the manual layer is responsible for facts and scope, and the recycling station catches accidental deletions. Reversing the order will allow low-quality content to enter the visible range of search with a complete index.

2026-10-26

What are the recovery methods available in the background if an article is deleted or modified

First, determine whether it was a deletion error or a modification error. In AnQiCMS, the deletion action is to move the formal document to the recycle bin, not a physical deletion, and it can be restored from the recycle bin; for modification errors, check if there are draft and historical versions of the pending release status. If there are no rollback versions available, there is only the last line of defense, which is the data backup, and the backup range should include static files, and it should be verified regularly whether it can really be restored.

2026-10-25

Before importing tens of thousands of documents in bulk, you need to confirm which few things

Before importing tens of thousands of documents in bulk, you need to confirm the field mapping, duplicate check, batch rhythm, and rollback path. This article takes the AnQiCMS ZIP and Excel batch import, and document content import interface as an example, explaining the inspection methods before import, and how to recover after import errors using the recycle bin and backup.

2026-10-25

How do the fields and tags correspond between the Chinese and English stations?

When the Chinese and English stations share a set of content models, the field correspondence should follow the principle of 'define structure first, then language': use a custom content model to define bilingual common fields, maintain Chinese-English matching words with a keyword library, the whole page HTML translation is responsible for the final draft, and multi-site management is responsible for switching and isolating. This article provides common errors in field mapping and the implementation methods of label correspondence.

2026-10-25

Before going live, use the online experience site to inspect the functionality, which can help identify real problems in advance.

Before going live, use the online trial site to verify the function, which can help to discover three types of real problems in advance: the background action path is inconsistent with expectations, the function has an entry but lacks key configuration, and it is impossible to judge the scope of impact of batch operations. The trial site verifies the function and interaction, while deployment, data migration, and backup recovery require a separate test instance to be run independently, with two people responsible for a section of verification each.

2026-10-26

The test version announcement is only suitable for test environments, and how to set up a pre-release site.

Announcement of the test version written by oneself, 'Only suitable for test and development environments', means that a pre-release site with the same version, configuration, and data form as the official site is needed to carry out verification. The method is four steps: copy data副本 from the official library, replicate the environment according to the configuration of the official site, isolate the verification content with drafts and pending release status, and synchronize the conclusion to the official site after acceptance.

2026-10-26