In this week’s hangout, John Mueller discusses using canonical tags for content migration, redirecting affiliate links, and methods for testing structured markup quality issues.
hreflang tags without reciprocal tags will be ignored
Any hreflang tag which is not reciprocated with an equivalent hreflang tag on the target page, will be ignored, because Google assume it’s incorrect.
Redirected affiliate links are OK
Structured markup page quality issues can be tested with site:
Structured markup has to be technially correct, and they are accepted from a policy, and also meet quality criteria to be displayed.
If you do a site: command for a URL, and can see the markup appearing, but it doesn’t appear for regular searches, it means the page does not meet the quality requirements.
Google doesn’t look at your URL structure
Google doesn’t look at your URL structure semantically to understand your site architecture. They just crawl the URLs you link to.
Traffic volume doesn’t affect crawl frequency
Traffic volume doesn’t affect crawl frequency, as Google don’t know this.
If Google can recognise important pages which frequently contain links to new pages with unique content, they will be crawled more quickly.
But crawl rate, or rate of change of a page, doesn’t have any direct relationship to ranking.
Canonical tags can be used for cross domain migration
If you want to migrate part of your site to a different domain, but keep serving that content on the site, you can canonicalise across the domains, but Google might not accept it and may keep the original.
Large site navigation can make it harder for Google to read pages
If your navigation is very large, it can add a lot of text to the page which might make it harder for Google to identify the parts of the page which are relevant.
The conversation suggests that Google is trying to identify boilerplate elements which it can ignore, but the harder this is, the more likely that genuine content might not get classified as relevant.
