Authority Distribution & Flow
- Authority Distribution
- Authority Distribution
- Analysis Techniques
- Common Problems.
- Canonical Chains
- Redirect Chains
- More Chains!
- Dead Ends
- Indexation Problems
- Canonicalization Problems
- Inconsistent Linking
- Linking to Non-canonical or Non-Indexable Resources
- Isolated Content / Island Content
- Lack of Deep Links
- External Links.
- Nofollow Meta Tag
- Home Page Accessibility Issues
- Other Accessibility Issues
- Canonical Chains
- Best Practice & Solutions
In this guide, we are looking at how authority delivered from your sites backlink profile flows through your site.
There are a number of issues that can heavily impact a site’s ability to allow authority to flow where you want it to go. So, we also cover common obstacles to authority flow and how to prevent authority leakage from your site in this article.
There are various metrics that exist to describe or attempt to quantify the ‘authority’ of a webpage or whole domain… Google used to provide a PageRank score which is now no longer used or updated.
Authority is one to two broad components used in Google’s algorithm (the other being relevance) to determine who should rank where and for what search. Consequently, authority is as an essential commodity for every website.
Website’s acquire authority through backlinks, and we cover this topic in more detail in another section of the knowledge base… What we want to look at is what that authority does once it enters your site.
Authority is like light in many ways, it doesn’t site still, so when it enters your site, for the most part, it keeps moving… Authority is passed on in a number of different ways, none of them are perfect, each are described below:
Conversely if you are using a standard 301 redirect correctly, you could pass on the vast majority of the authority to the target page, we estimate this to be between:
90% to 99% of authority passes through a proper 301 redirect.
Redirects will typically override other components, for example a page with 100 links and a canonical tag pointing to another page will all be ignored if a redirect is in place. This is because the redirect will take the crawler away from the page, not allowing it to see the links or canonical tag.
Google have said in the past that canonical tags work like redirects for the purposes of authority, meaning that they pass between 90% to 99% of a page’s authority onto the target page (if not self-referential).
That said, we have seen non-canonical pages outrank their canonical counterparts, so this is just a hint and is definitely not as strong a factor as a redirect for that reason. Google will not rank a page that redirects to another if they detect the redirect.
Google will generally ignore the links on a page if it has a non-self-referential canonical link present. This is because you are telling Google that this not the canonical version of the content.
We cover this concept in much more depth in our guide to canonical tags.
Internal links pass between 85% and 90% of a page’s authority… But this is divided by the number of links on a page. For example; each link on a page with 10 links will pass 8.5% to 9% of the pages authority to the target page.
This means that the more links you have on a page the less valuable each one is, and obviously the converse is also true.
You can stop the flow of authority through a by adding the ‘nofollow’ attribute to the link, or adding a nofollow meta tag to a page to prevent the flow of authority through all links on that page.
We cover these concepts in more depth in our guide to internal linking.
Authority Distribution is essential to every site; this is the process of ensuring that the pages that need the most authority have it and those that don’t need it so much have less. Due to the influence that authority has on a page’s ability to rank for its target keyword/s, understanding this flow can be critical to optimising your site.
Backlinks are a huge part of this and we look at backlink distribution analysis in another guide, if you’re interested.
The benefits of good authority distribution can be a hugely significant factor in ranking well. The impacts of poor authority distribution can be catastrophic to a site in the most extreme cases. We cover some of the most common problems later in this guide, which will provide some context for this.
This all depends on what you want, what pages are the highest value, etc… Typically we want to see a strong correlation between the most valuable pages and the number of followed internal links that they have. More valuable pages should have a larger share of the internal links.
Although the reality is much more complex than just doing this, it is a good start. Raptor helps you to understand this with our authority distribution tool.
There are a number of techniques that we can employ to assess whether your site is distributing authority as desired. Depending on why you are reading this or what you are trying to achieve the following techniques should help.
The first thing to understand or to choose are the metrics, depending on what tools you use you will have a range of metrics. We are not going to get into all of these here as we discuss many of them in more detail in other guides.
The most common authority metrics are Domain and Page Authority (Moz) or Trust & Citation Flow (Majestic). You can also look at the number of backlinks to any given page and weight this by authority.
Our distribution analysis tool calculates how much authority is passed on (based on whether it’s a canonical link, internal link, nofollow, or a redirect (based on type). We assess the entry points for backlinks and the authority that those links pas into your site, we also calculate how much authority is lost due to problems or external links.
This is presented primarily in a visual interactive map of your site colour coded to show the high and low authority pages. Without a tool like ours this is a very complex and processor intensive task.
We look at distribution analysis techniques in much more depth in another guide.
Auditing your site on a regular basis can help to identify most of the problems that we detail in the following section. Again, Raptor automates this for you, but its more than possible to use crawl data, a few basic tools, Chrome plugins and a spreadsheet to identify these issues.
Looking at ranking and Google Search Console data can reveal things like; the wrong page ranking for a keyword, or a bunch of pages ranking for similar terms. This is a strong indicator that authority distribution is not working in your favour.
Review the section on solutions to see how to deal with this particular issue.
As mentioned, building a table and some charts to look at what the most and least linked to pages are can help to understand any systemic issues. Often these are caused by menus and templated navigational links or a lack in content (contextual) links.
We discuss link structures in more depth as part of another guide.
This is more off-page than on, but the lines are blurry in this area… Looking at what URLs have backlinks is essential to understanding where authority is entering your site. These pages should be noted and dealt with differently than other pages.
For example, retiring a page on your site that has 20% of the deep links (backlinks) would be a bad idea, if you just let it 404… Instead it would be better to update the page or canonicalize it to another page. Allowing a high value entry point page to return a status other than 200 can lead to people removing the links to that resource. If you saw that you had a bunch of links that pointed to error 404 pages or 301 redirects, you would (or should) consider updating or removing the links.
If you notice a drop in your rankings, it is worth checking your entry point pages first to see if they are still functioning properly.
Problems with authority distribution can result from a broad range of issues, typically those affecting larger numbers of page or the home page are the most severe. You work hard to earn backlink and authority, so it’s worth ensuring that you aren’t wasting that investment with poor flow or distribution.
Perhaps not exhaustive, this list is a pretty good list of the most common causes of authority distribution problems.
This represents a leak rather than a block or obstacle to authority flow & distribution. A canonical chain is where a series of pages are canonicalized, for example:
Page A - Canonically links to Page B
Page B - Canonically links to Page C
Page C - Canonically links to Page C
In the example above, if we assume that 95% of the authority is passed on through these canonical links we see a loss / leak of authority:
Page A - 100% authority
Page B - 95% authority (1 x 0.95 = 0.95)
Page C - 90% authority (0.95 x 0.95 = 0.9025)
So, you can see that around 5% of the original authority is lost in the chain; the more links in the chain the more authority is lost:
No of Links in Chain
A canonical link should only point to an indexable, canonical page.
This is the same as with canonical chains except that instead of canonical links we are talking about redirects. Assuming that the redirects are crawlable by Google and they pass on authority, we end up with the same problem of authority leaking out through unnecessary links in the chain.
A redirect should only point to an indexable, canonical page.
Just for fun there are variants of the above that will have similar problems such as:
- A redirect pointing to a page that is canonically linking to a different page
- A canonical link pointing to a page that redirects to another page
- Extended combinations of the above
This is characterised as a chain (see any of the above), a canonical link or a redirect that points to an inaccessible page. There are many ways that you can create a dead, any canonical link or redirect, in a chain or not, that points to a page that does the following is bad:
- Error page (any kind, 404, 501, etc)
- A non-indexable page (noindex tag)
- A page that can’t be crawled (disallowed from robots.txt)
In any of the above examples, the authority will be lost as it ends up on a page that cannot pass it on and does not require it.
You could create a legitimate dead end where the final page has a nofollow tag on it, and a self-referential canonical link, thus preventing any of the authority leaking out of the page. This could be done to create a one-way flow of authority to your top tier high value product pages for example.
There are as many types of loops as there are chains; canonical loops &redirect loops. The simplest form of a redirect loop is:
Page A - Redirects to Page B
Page B - Redirects to Page A
This would make both pages inaccessible (to users and robots), thus prevent them from passing on any authority as well as wasting the authority they have.
A canonical loop will at least allow both pages to be accessed by users and robots but will still create a black hole for authority… As well as potentially causing neither page to rank.
Loops can also include a chain component such as in the example below:
Page A - Redirects to Page B
Page B - Redirects to Page C
Page C - Redirects to Page A
Loops of all kinds should be avoided at all costs, if this only affects a couple of low value pages, the impact will be quite small. However, if this affects a high value page or affects access to a branch of the site, or if it affects the home page… This could cause major problems for your site both in terms of authority flow, distribution and indexation.
Conversely to the above, indexation issues can also affect the flow and distribution of authority throughout your site. Depending on the specific indexation issue, your pages may not be indexable or crawlable by Google. A page that is not indexed will not pass on authority as Google will not know about it.
Again, this may be relatively minor but I the case of a category page this could prevent a whole directory from receiving some link juice.
For more information on this topic read our guide to indexation.
As mentioned, canonical tags when implemented badly can adversely affect your site… Canonical duplication issues are characterised by content being accessible from multiple URLs. Failing to control how your content is canonicalized can result in leaking authority, we cover the principle mechanism for this in the next section on inconsistent linking.
For more information on this topic read the section of our knowledge base on canonicalization.
The above issues are also often coupled with internal linking inconsistencies that allow them to become a problem. If a site is accessible with and without the www, but there isn’t a single internal link pointing to the www version of the URL; the problem will be pretty minor at worst.
However, if you are lining to non-canonical pages, you will be leaking authority as this qualifies as a kind of chain. If all of your links point to the http and www version of your site, they might pass through two redirects before getting to any page, this means you lose around 5% of your authority before you’ve even started!
Once you add in other redirects, you start to lose even more authority… Ensure link consistency throughout your site to maximise authority flow.
If linking to non-canonical, non-indexable or uncrawlable pages, you should always use a nofollow link to prevent losing that authority to a page that does not require it. This can be a major leak on large sites.
You should only link to resources that redirect under specific circumstances where it is unavoidable or has little impact… In general, you should always link directly to the resource you are pointing users to.
On large sites, it’s easy to end up with isolated pages that are either not linked to or only linked to from nofollow links, making them what we call ‘island pages’.
These pages will be unlikely to rank due to the lack of authority flowing into them, these pages do not leak or lose authority they are the product of a lack of authority.
This is more relevant to off-page SEO and backlinks but is worth a mention as it can have various effects on your site.
Most sites will typically have a higher ratio of links to the home page than to all of the other pages put together. If your site has almost no deep links (links to internal pages) you are relying on all of the authority to flow from the home page to the rest of the site.
The further a page is from the home page the less authority will flow into it, as we mentioned in the sections on canonical chains, the last page in the chain (in this case a click chain) gets a very small amount of authority.
If you have a lot of external links all set to follow on high value pages you could be unnecessarily leaking authority to external resources.
If you’ll recall we said that a link will pass through authority to other pages as a percent roughly equal to what percent it represents of all links… So, if you have a page with only a few links, those links can be sending authority away from your site in larger qualities.
There is nothing wrong with a (follow) external link, just be aware of where they are and how they affect the flow of authority off-page.
We have mentioned this tag before in this guide, if it applied to all or most pages it will prevent authority from flowing between them, essentially creating a site of island pages. You can legitimately use this tag on pages with lots of external links for example, but they should be used sparingly and only when necessary.
Due to the fact that on most sites the home page is the entry point for the vast majority of the authority the site receives; accessibility issues with the home page can prevent all or most of the authority from being passed on.
This is really just an extension of the above, any page that is inaccessible will not pass on authority or rank for its target keywords. In sites with directories and more complex structures, the accessibility of a few pages could prevent authority flowing into huge segments of the site.
We have already covered, the basics, common problems and some techniques for how to identify those problems… Here we discuss a few of the best practices that will help to prevent issues from creeping in and some solutions for how to fix them if they do.
Some tools (like Raptor!) will send you alerts to your phone or email if certain critical issues occur, this can help you to identify and resolve thee before they are picked up by Google. This is a great semi-preventative measure.
This is an audit of sorts, using ranking data and a URL to Keyword Map of your site you can see what pages are ranking for what keywords. Where find that there are many pages ranking for similar terms that shouldn’t or the wrong page is ranking… This is an indication that these pages are either more relevant or have more authority (or both) than the desired landing page. Hence this type of analysis is useful for identifying distribution and flow problems.
Consider consolidating this content using redirects and canonical tags to push a single preferred page above all of the others… This can often have a very positive impact on your rankings for head terms.
We cover this in more detail in our guide to canonical duplication and canonical tags, so read either of those guides for more information on this topic. For the purposes of this guide, just ensure that canonical tags and paginated series are setup correctly. As mentioned earlier; avoid canonical chains and loops.
Ensuring that you have a properly formatted and structured XML sitemap containing consistent URLs and only canonical, indexable, and crawlable pages. This helps to prevent some indexation problems.
Also ensure that you have a HTML sitemap listing all canonical, indexable, and crawlable pages as this helps to provide at least one HTML follow keyword rich anchor text link to every page. This ensures that every page that needs authority has at least some.
We have discussed these previously in this guide, but it is one of the simplest and easiest types of analysis to perform. Simply sum all follow links by type pointing to each page and sort from highest to lowest. You will easily be able to see some problem pages or an undesirable order emerging.
This is a good starting point to begin a wider and more granular analysis of authority distribution and flow around your site.
One of the most effective ways to mitigate problems, especially on larger sites; is to perform regular checks and audits. Often there is no better way to determine if or what problems your site has than a thorough technical audit.
Regular checks allow you to compare data to see if problems are increasing, if there is a pattern to them; and helps to determine whether things are getting better or worse.