How should batch 301 redirects be arranged during the domain name and version update, and how long will it take for old links to recover from dropping收录
How to arrange batch 301 redirect when changing domain or redesigning, how long will the old link drop收录 be recovered? The arrangement is divided into three levels according to the complexity of mapping: the whole site can be moved with only one rule and template when the address structure does not change; the column level changes need to be matched according to the new structure; if the corresponding new page cannot be found, merge it into the upper-level column or search page. The speed of recovery does not depend on the number of rules, but on whether the crawling end can see the new address in time - when the site map is still the old address and the internal links in the text still point to the old domain name, even if the redirect is configured in full, it will also drag on a long tail.
Lock the new URL structure before redesigning
The order is crucial: first define the new structure, then discuss the redirection. Doing it the other way around can lead to a difficult situation - the new address is still being adjusted, while the redirection for the old address has already been set up. Each adjustment of the structure requires a new configuration, and historical mapping relationships may overlap.
This step of locking the structure also determines the form of the pseudo-static rules. The pseudo-static rule management supports custom URL writing, which is exactly what it can be used for during a revision to define the new form: how to express the category page, whether the detail page should have a category prefix, and how to arrange the pagination address. Once the rules are finalized, the mapping table is a mechanical correspondence of 'old address - new address', and batch configuration can only be automated.
301 Redirect Batch Processing
| Migration Type | Mapping Method | Rule writing | Missing Consequences |
|---|---|---|---|
| Whole Site Translation, Structure Unchanged | Template Batch | Old Domain Redirects to New Domain Path | Only Home Page Redirect, Deep Pages Lost |
| Column Level Adjustment | Grouping by Column | Prefix of old column path replaced with new prefix | Some articles fall into 404 |
| Articles merged or deleted | Bottom-up | Merged into parent column or in-site search page | External link weight is hanging, long-term collection does not drop |
| URL with parameters | First jump and then redirect | Parameter addresses are unified and converged to standard addresses | Two addresses for the same content coexist |
In practice, it is recommended to run large template mapping first, then handle a few special cases. The 301 Redirect Management is responsible for putting these URL redirection relationships into the background, the key points are three: first, use permanent redirection instead of temporary redirection, as the temporary state not being撤销 will make the search end treat the new and old addresses as two sets of content; second, the old domain name should be retained for a sufficiently long time, the launch of the rules does not mean that external links and cached visitors have disappeared; third, the redirection chain should not be linked, as the old address will only fall to the new address after two or three times, and the weight and access will decline.
Replace the old links and navigation in the text with the site-wide replacement processing.
It is not enough to handle the address layer, there are still a batch of 'old addresses written in the content' within the site. The internal links in the article, the entries in the navigation menu, and the addresses pointed to by the anchor text, these three types of addresses need to be changed accordingly. The full-site replacement tool is responsible for batch replacement of keywords or links, suitable for cleaning up the old domain names in the text at one time; the navigation settings and the number of single-page entry pages are limited, and it is faster to check them item by item.
The typical manifestation of omitting this step is that the old domain names have all been redirected, but the site is still pointing to the old addresses, and the access needs to jump once before landing on the new page, and the path seen by the crawling end is also therefore longer.
Site map and collection recovery rhythm
The recovery speed is basically determined by two things: the timing when the new address is seen, and the timing when the old address is stopped from being accessed.
The first thing relies on the list and push. The Sitemap needs to be regenerated after the structure is finalized, and all the new addresses must be in the site map. Writing the old address into the list is equivalent to letting the crawling end repeatedly collide with the redirect. Link push is responsible for actively sending the new and changed addresses, Baidu push and Bing each connect to one side of the search system, push the homepage and main column pages first on the day of the revision, and then push the deep articles in batches, which is easier to be processed in time than pushing all at once.
The second thing is to retain the old entry, this step cannot be rushed: the old domain and old address should be kept for a sufficient period of time until the visit volume of historical entries naturally decreases to almost zero. How long does it take for the index to recover, most sites are counted by week rather than by day, and the specific speed depends on the original update frequency of the site and the distribution of external links - for old sites that are not updated for a long time, the crawling cycle is already long, and the redirection will take effect more slowly. To judge whether there is a recovery, just look at two curves: whether the amount of list submission and the amount of new address inclusion increase synchronously, and whether the amount of requests from the old address continues to decline.
Common questions
Can I only set up 301 for the homepage?It is not recommended. The links in the external links are often old articles rather than the homepage, and only jumping to the homepage is equivalent to losing the most valuable part of traffic and weight.
Do you need to resubmit the sitemap after the revision?Yes. The old list will cause the crawling end to repeatedly visit the redirected addresses, so it is recommended to regenerate and submit it first, and then cooperate with active push to batch update new addresses.
Can the old domain be stopped in advance?Not too early. After the shutdown, all historical entries will directly become 404, and the accumulated signals of redirection will disappear, usually retaining to the old address when the number of requests tends to zero.
Is it麻烦 to change the temporary redirection to a permanent redirection?Need to be corrected item by item, and the temporary status that has taken effect will coexist with the old and new addresses for a long time. It is more convenient to configure permanent redirection uniformly before the revision.