Written by Merche Martínez
Índice
- When we think about renewing our Web, we have doubts such as: what will happen to the current traffic that my Web receives, will I maintain the positions in the search engines, what happens to the external links that point to my current pages, how do I make the new pages receive the traffic of the old ones, what happens to the external links that point to my current pages, how do I make the new pages receive the traffic of the old ones?
It is true that the migration of a Web site is a very delicate task for the positioning of a Web site and that depending on the situation in which our Web is, it must be faced in one way or another, although the main phases are the same:
- Study of the architecture of the new Web
- Content and optimization
- Redirections
- Follow-up
Study of the architecture of the new Web
First of all, a study of the current architecture of the site must be made in order to define a Web map with the pages organized in different levels following a balanced architecture that helps the correct indexing of the site. By balanced architecture I mean that there should be a balance between the importance of the content and the depth to which it is found in the architecture.
This architecture is the one that will be recommended for the new site. This step must be done prior to the content dump since the relevance study will be done on the new architecture.
Content and optimization
Content migration is perhaps the most entertaining task. This involves copying the content of each of the pages and generating content for the new pages. When copying the content of each of the pages, the internal links must be modified to point to the new pages.
As we have already mentioned, the migration of a Web site is a delicate process in terms of positioning, so if we are going to make improvements in the optimization it is preferable to do them before publishing the new site.
In this optimization phase we cannot go crazy: if a page is very well positioned and receives traffic, we must follow the rule that if something works well, don’t touch it.
Redirections
First of all, the candidate landing pages to be redirected must be identified, these pages can be:
- Pages that rank well for critical keywords in website traffic.
- Pages that receive more visits from natural results.
- Pages that receive a lot of direct traffic.
- Pages with the highest number and quality of inbound links.
Once identified, 301 redirects must be made from the old pages to the new ones.
There may be pages that do not have an equivalent on the new site, for example, discontinued products. In that case the redirection of the product page that no longer exists can be done to the category page of that product or a 404 server code can be returned on a friendly page where related products are displayed. These decisions are made on a case-by-case basis.
Follow-up
This stage is as important as the previous ones. No matter how careful we have been, no matter how thorough we have been in all the steps, it is possible that there are pages that have not been redirected correctly.
In Google Webmaster Tools we can keep track of new indexed pages, possible server errors and pages to be de-indexed.
It is normal that the pages that return a 404 server code will increase in the weeks following the migration since they are the pages that have been removed from the previous version and it is also normal that the indexation will increase since the new pages will be indexed first and then the old ones will be de-indexed.
Success or failure?
A website migration is considered a success if organic traffic remains virtually stable and objectives are still being met to the same extent weeks after migrating a website. Experience tells us that it is possible to successfully migrate a website, but there are also external factors that can have a negative influence and that Google has the last word in this case.
Finally I leave you an example of the traffic collected in Google Analytics before and after a migration that can be qualified as a success.
