Internal links are the hyperlinks within a website. The total web of internal links on a website is also called the link structure of the website. This structure is crucial to enable good navigation for users, but it also has a lot of influence on the indexation and average ranking of a website in search engines.
In this HowTo, various guidelines explain how you can optimize the internal link structure of a website for search engines. Specifically, Google is regularly discussed, because there is relatively much information available about this supreme leader of search engines.
Ensure better indexing with internal links
Make your pages easily accessible using internal links, then it becomes easier for search engines to index them. During a crawl, search engine bots follow the links they encounter and in this way discover new pages that can be indexed.
If there are no internal links to a page, this page runs the risk of not being indexed, or relatively late. This page will also rank less well on average. These types of pages are also referred to as orphan pages or orphaned pages.
If you want to help search engines to find and index pages faster, create an XML sitemap that contains the links that point to all pages on the website that you want to have indexed. Especially with large websites, a sitemap helps to significantly improve and speed up the indexing process.
You can also add meta-data in a sitemap. You can define a crawl priority and specify how often the content on a page is changed. However, Google does little more with this information, because the content is often changed less quickly than indicated in the sitemap. For more certainty that Google uses the data, add timestamps to the URLs in the sitemap, which indicate when the pages were last modified.
Marketing Facts How-
Because not every marketer is an SEO ninja or social media guru, we like to publish HOW-TO’S on Nichesitebasics. They will therefore not be surprising for everyone. That does not mean that there is no tip that everyone can learn from.
Give the most link value to your most important pages
Search engines look at a large number of factors when determining the ranking of a page. One of the most important ranking factors is the link value that a page is assigned. Pages on a website build up link value (called by Google PageRank ) when other websites link to them. The value that is passed on depends on the authority and reliability of the linking website. A part of this link value is then distributed over the website via internal links.
A good internal link structure is therefore of great importance to correctly distribute the link value on a website. When considering a suitable structure, bear in mind that the homepage can usually distribute the most link value (because other websites usually link to the homepage most often).
A pyramid structure is a suitable link structure for most websites. This structure fits well with user-friendly navigation and ensures that the link value from the homepage is distributed somewhat evenly over the underlying pages. First comes the homepage, then the category pages (and if necessary subcategory pages), and below that more specific pages for, for example, products or articles.
Any deviation may, of course, be desirable. If you have very popular or important pages that are lower in the hierarchy of the link structure, it may be wise to link to these pages directly from the homepage. If there are pages high in the hierarchy that are not that important, then it may be wise to link to these pages to a limited extent.
For internal links that refer to pages on your own website, avoid using the Nofollow attribute. Until 2008, link value could be controlled in a smart way (this was also called ‘PageRank sculpting‘). To prevent such manipulation, Google changed the operation of the Nofollow attribute in 2008. For internal links, the attribute hardly has any function. Moreover, it can hinder good indexation of the website.
Link most often to the most important pages
The number of internal links that refer to a page is one of the signals to search engines about how important a page is compared to other pages on a website. So make sure that a relatively large number of internal links refer to the most important pages and relatively fewer links to less important pages.
How do you know which pages on your website most links refer to? For this, you can consult Google’s Search Console internal links report. You can also crawl the website with a program like Screaming Frog and then organize the pages based on the number of internal links that refer to a page.
The number of internal links that refer to a page is easiest to influence with sitewide links, which are, for example, included in the header, footer or sidebar. For example, think of links in the main navigation of a website. If the main navigation is on every page of the website, then the links in the main navigation are also on every page of the website. For example, one link in the main navigation can provide a large number of internal links to a page.
Think about the location and context of links
In the first articles about PageRank, written by the founders of Google, the ‘random surfer model’ was explained. This model assumed that users clicked on links on a page in a random manner. According to this logic, the link value in the original Google algorithm was evenly distributed across all links on a page. In the article the PageRank Citation Ranking: Bringing Order to the Web from 1998 we read the following:
“The definition of PageRank above has another intuitive basis in random walks on graphs. (…) Intuitively, this can be thought of as modeling the behavior of a “random surfer.” The “random surfer” simply keeps clicking on successive links at random. ”
That Google wanted to distance itself from this model after a few years is evidenced by a patent filed by Google in 2004: Ranking documents based on user behavior and / or feature data. The ‘reasonable surfer model’ is set out in this patent. This model assumes that not every link on a page is followed as often by users. For this reason, different values must be assigned to different types of links:
“When a surfer accesses a document with a set of links, the surfer will follow some of the links with higher probability than others.”
The patent says that Google can judge links on, among other things, the following properties:
- The size and color of the anchor text font;
- Position of a link on the page (Height on the page / in the HTML code, the link is in the content, sidebar or footer, etc.);
- Position of a link in a list of links;
- Link attributes (e.g., bold, italics, underlined, etc.);
- Number of words in the anchor text;
- Whether or not the anchor text is of a commercial nature;
- The link type (for example, an image or text link);
- The context of words before and after the link;
- The topics with which an anchor text can be associated;
- Whether the link refers to your own website / a website with the same host, or to another website.
Link to content that is relevant to users
The Google patent discussed under the previous heading not only deals with link properties such as the location and form of links, but also user data that can be linked to links. The patent describes that data could be collected with a web browser:
“For example, the web browser or browser assistant may record data concerning the documents accessed by the user and the links within the documents (if any) the user selected.”
For example, Google can look at how often a link has been clicked, and generates positive and negative events that are associated with the link. Clicking on a link on a page can be a positive event, while not selecting the other links can be a negative event for those links. Even if none of the links on a page is selected, this can still be considered a negative event.
In addition, Google formulates the idea of looking at the session time of a page as a ranking factor. Here too, the linked pages play an important role in making a correct assessment of the value of the page.
The news.google.com website is cited as an example. The homepage can have a relatively short session time, because the page functions as a hub that links to all kinds of news articles. If Google calculates the session time of this hub page, it can add up the session time of the hub page and the linked articles (the route data).
The above examples are a clear signal that Google is increasingly looking at user signals when evaluating links. So if you link to relevant content for users, you increase the chance that positive user signals are sent to search engines that can help improve the average ranking of a page.
Keep in mind the number of links per page
Years ago there was a guideline from Google that said a page should not contain more than about a hundred links due to limited crawl capacity. Nowadays, websites are on average much larger and pages regularly contain more links. Google nowadays has much more crawl capacity and this guideline no longer applies.
Be aware that link value is distributed over all links on a page. If you have a lot of internal links on a page, the value that you provide per link is therefore much smaller. Therefore, do not add unnecessary links to a page.
Optimize anchor texts and alt texts
The anchor text of a link (the link text) is used by search engines to find out more about the subject of the landing page. The text around the link can also be used by search engines to better understand the subject of the landing page. Therefore, provide clear anchor texts that describe the subject of the landing page well. Do you want to rank on a relevant keyword with the landing page? Include this keyword in the anchor text of a link.
However, do not always use the exact keyword in all your internal links, but try alternating with anchor texts that partially match the keyword. This way you prevent over-optimization of the anchor texts.
If you link to a page with an image, you can optimize the alt text of the image. With links in images, this text performs a similar function to the anchor text.
Consider setting up a silo structure for all your pages and posts, especially affiliate posts.
Watch out for too many internal redirects and broken links
Not all links on a website refer directly to the correct page. Adjustments and / or poor maintenance can result in unnecessary internal redirects and broken links on a website. You can recognize these links by the HTTP status code:
- 301 = redirect with a permanent redirect to another page.
- 302 = redirect with a temporary redirect to another page.
- 404 = the link is broken, refers to a page not found.
- On Wikipedia is an overview of all HTTP status codes.
Many internal redirects have a negative impact. The loading time of pages is extended and a link that is redirected passes slightly less link value (90-99 percent) than a normal link. If it concerns temporary redirects, then no link value is passed on at all. Broken links are also not desirable, because this link value indicates pages that are not found. Many pages that have not been found can also have a negative impact on the user experience.
You can track internal redirects and broken links with various tools. You can find broken internal links on your website by looking in the crawl errors report in Search Console. You can also easily find internal redirects and broken links with the aforementioned program Screaming Frog. You can then eliminate unnecessary internal redirects by adjusting the links and linking directly to the correct pages. You can have the broken links refer to another page or remove it. You can of course also try to make the page that has not been found traceable again.