Google’s virtual meetings
Like many other live events in recent months (and also in the comings…), Google Webmaster conferences have been canceled due to the Covid-19 pandemic. To ensure that the public still has access to the expected information on the SEO and the latest news in the industry, the American giant has involved its speakers in the Webmaster Conference Lightning Talks series, reduced video versions of the presentations they would have given in cancelled conferences.
Reminding the importance of links
Before embarking on the technical aspects, the Googler makes a premise to reiterate the value of links and their sensitivity on the fate of a site (and the entire Web). Not surprisingly, he begins, “I believe that all those who have used the web have seen them”, because they are essential to bring users from one page to another, within the same site or on another site.
Links were the “killer feature of the web when it was introduced, and even today they are a rather important thing”: in addition to allowing users to browse through different contents, they serve bots and search engines like Google Search. The key function is to allow crawlers to find other pages on our website and to switch from page to page, finding links on a page and then following them.
The other important purpose is that, by following the links, the crawler acquires an understanding of the site structure and information architecture, and this is useful for ranking purposes, to understand which pages might be relevant for a given topic.
The right way to create links
Simplicity is the watchword for not making mistakes, according to the Googler, who invites us not to complicate our lives too much for the management of this topic (as well).
The easiest way to insert a link into a page is to use the good old HTML <a> tag with the destination URL in the href attribute, as in the example
Do not leave out the href attribute
Ultimately, “it’s not a good idea,” Splitt summarizes, because it makes the link fragile.
Do not use pseudo-URLs
Another bad idea is to add an ahref attribute without a useful URL or with a “pseudo-URL”, because we will get the same result as a link without a href attribute.
Do not use buttons
The third thing “not to do” concerns buttons: although it may seem like a workable option to add a link to a page, even the use of a button is not a good idea according to Google.
The rule of thumb is: if a link activates something on the current page, then we can use a button. Instead, if a link takes a user to another content that was not on the page before, then it is better to use a standard link.
How to implement a correct URL
The basic advice that comes from the video is “to use semantic HTML markup and point the link to a correct URL”, and Martin Splitt spends a few minutes to explain what the correct URL means by analyzing the basic structure of the links through two example strings:
URLs themselves contain the following attributes:
- A protocol
- A host
- A path toward a specific content
- Fragment identifiers (optional)
It all starts with the protocol, which basically represents the language you speak when accessing the URL, which can be HTTP or HTTPS. Then there is the host – example.com in the example – a name that some computers or multiple computers respond to. Last but not least, there is a path to a specific content of that computer, to which is added an optional part, a fragment identifier.
Properly use the fragment identifiers
The fragment identifier is marked with the symbol “#” within a URL and indicates the subsections of a given page, not necessarily a different page. As Splitt sums up, it does not lead to a different content, but rather points to specific locations within the same content – if we do not use it, we get the same content.
Crawlers ignore the fragment identifier, just pretending it does not exist. This element is particularly important to evaluate if we create a single page application with links full of fragment identifiers: crawlers will not follow the links, so they will not be able to understand the Web app.
- Use an appropriate semantic link marking.
- Do not use fragments to load different contents into single-page apps.
The practices to follow so to avoid compromising the result are (at least) four:
- Do not leave out the ahref attribute.
- Do not use pseudo URLs.
- Do not rely on click managers.
- Do not use the buttons.
Further info on Googlebot and links
- What types of links can Googlebot discover?
Googlebot analyzes the HTML of a page, looking for links to discover the Urls of the related pages to be scanned: to discover these pages, we must make the links true HTML links.
- What types of Urls are good for Googlebot?
Googlebot extracts Urls from the href attribute of links and queues them for scanning. This means that the URL must be solvable: simply, the URL should work when inserted in the address bar of a browser and refer to a real web address to which Googlebot can send requests.
Yes, as long as these links meet the criteria of the webmaster guidelines.
- Does Googlebot include URL fragments?
The URL fragments, also known as “hash URLs“, are technically valid, but may not work as intended with Googlebot.
The fragments should serve to direct a content within the same page and are absolutely fine when we use them for this purpose.
- Does Googlebot still use the AJAX crawling scheme?
The AJAX crawling scheme has been deprecated for some time now, so we cannot rely on it on our pages.
Google’s recommendation is to use History API and migrate Web apps to URLs that do not rely on fragments to load different contents.