Anchor Text Selection & Optimization
Selecting the right anchor text for your backlinks whether its for a press release, a social profile, pbn link or guest post, can be the difference between climbing up the serp, or receiving the old "google smackdown". If you aren't careful you can over optimize, but if you are too careful you can miss out on an opportunity to help google understand exactly what your content is about. So lets dive in to how we can use what information google gives us to select the "best possible" anchor text.
I got so frustrated with the process of manually doing this I decided to hunker down and right a web application that could to what was taking me hours ... in seconds. I just went live with it and you can check it out here > AnchorProfiler.com
What Is An Anchor Tag Anyway?
First off lets make sure we understand exactly what an html anchor tag is as well as what the anchor text is. An anchor tag is the html tag that allows you to link from one html document to another somewhere on the interwebs.
For example the word "anchor tag" above has been "hyperlinked" which is derived from the link created between HTML (hyper text markup lang) documents.
As you can see the text is underlined which is typically the default style applied to the element.
*Pro Tip: If you want to make your links less or more clickable this styling can be overridden, by adding some css code to your sites.
The anchor tag has a few different properties, a couple of which we should be familiar with.
The href property which stands for hypertext reference, is the path to the document that is being linked to
The Target property allows you to tell the browser how you would like the new document opened (same window, new window, etc)
Now we have the Anchor Text which is everything that is inside the opening and closing a tags. <a>This Is Anchor Text</a>.
Why Does Anchor Text Matter?
Anchor text helps identify to the reader as well as the search engines what the content on the other side of the hyperlink is going to contain. Since we know that links from one page to another are similar to a vote being cast for that page, the anchor text helps identify exactly why you are casting the vote.
Once SEO's figured out that this was a ranking factor though they began to abuse the use of it. Since ranking #1 for a search term can sometime mean literally millions of dollars going hard on something you know works, until it doesn't anymore is understandable even if its not a long term strategy.
With google algo updates that have rolled out over the years like panda and penguin, google has taken action against the abuse or "over optimization" of anchor text. Now if you don't pay attention to how you are linking back to your site you can get in hot water. This is why we need a method to systematically review the anchor text profiles of competitors on page 1, and then do our best to optimize our anchor text profile to land within that range.
How To Select The Ideal Anchor Text
Selecting the ideal anchor text is not a sexy process that's for sure. It takes time, and effort to identify competitors, analyze backlink profiles combine data, and then write formulas to do some calculations for your.
The process in itself though is not complicated though.
To select the ideal anchor text you will need to
- Google your keyword and identify the competition, and log the url's they are ranking with.
- Use an SEO tool like ahrefs or majestic to get all the backlinks to their site, and then download them into a csv
- Create a spreadsheet where you can combine all of the csv's into one
- Go through each backlink and set the "type" of backlink it is
- Calculate the ratio of total backlinks by type for each url
- Aggregate that data into a blended page 1 anchor text profile
The anchor text types we will be identifying are...
In the end we will end up with something like this ...
What About Hidden Links?
I've gotten this question / comment a lot when talking about this.
"Well, sites are going to have links that are blocking seo bots so you can't know what their true anchor text profile looks like, so you can't really generate an accurate anchor text profile"
My question to that is: Does that mean because you can't have 100% accurate information that you shouldn't try?
If thats the case then we shouldn't be doing SEO in the first place since you will NEVER be able to see 100% of the picture. Google themselves doesn't know about EVERY link EVERYWHERE on the web. Its our job to learn as much as we can from our competition & take action.
If you see a site that is ranking in the top 10 for a competitive keyword and they have no links or its obvious that they are blocking most of their links, then don't use that site in your calculation.
Its that simple. Don't throw the baby out with the bathwater here folks.
Step By Step Instructions
For this example I'm going to use "best ping pong table" as our keyword. We will use AHREFS for our backlinks. I'm not currently trying to rank for this so I will choose a site on page 2 that at least has some links so that we can identify where they could improve.
We can either use the top organic results that ahrefs returns in their keyword tool or we can just go to google in an private browsing window and perform the search ourselves.
From here we will take the top 3-5 results that appear to have a strong backlink profile. We talk about this more below, but if you have any experience in PBN's you know that PBN sites typically block seo bots, this means that we won't have ALL the links that are helping a page rank. So when picking the sites you want to aggregate you want to try and "sniff out" any ranking where it seems there may be hidden links.
An example of this would be a site that is ranking in the top 3 and only has a few links while the other sites that are ranking have hundreds of links to the page. This is where you will just have to use your best judgement, when choosing the sites you base your calculations off of.
Ok so now lets go ahead and choose the sites that we are going to base our calculations off.
Here we have performed our search, and we can see that there are a bunch of ads, and google shopping results... so lets scroll down to the organic search results.
Here we can see the different domains that are currently ranking.
Looking at these results we can see that Amazon has the number two position and then the other three results appear to either be authority sites in the table tennis niche or affiliate / review sites for ping pong tables.
So for our example lets take
Our first example http://www.bestpingpongtables.review/ has the following backlink profile. As you can see rom the screenshot this is where the time intensive portion is going to come in. We are going to need to manually go through all of these links and identify which type of link they are.
our next example is http://ponghero.com/best-ping-pong-table-guide.html. Here is the backlink profile metrics according to ahrefs.
next we have http://thepingpongguys.com/
Lastly we have https://www.best-table-tennis-tables.com/
A little gotcha here, that I didn't do on my first pass... Make sure you select the URL option (see below) when having ahrefs do the analysis.
We want links that are specific to this URL, and not all links across the entire domain.
Loading Data Into Excel
Now that we have identified our competition, and have gotten to the back link portion we want to go through and download the backlinks for each of these sites.
You can do this by clicking on the backlinks link in the left hand sidebar in ahrefs, then clicking the export button.
now lets open that file, and copy the contents into a new sheet in our google sheet workbook.
From here I like to move the anchor text column to the first column in the sheet so that its easier to navigate. Once we have moved it we will create 1 additional column in the A position which will be our Anchor Text Type column.
Once we have this done we can leverage some of excels functionality to help do the math for us as well as give us a visual representation of the anchor text distribution.
Now we will highlight the first two columns which contain the Anchor Text Type and the Anchor text itself.
In the top menu we will select data, and then pivot table. This will create a new sheet with a pivot table in it. If you aren't familiar with pivot tables they are what I like to call Awesomesauce...you can get to know them here.
Here is a screenshot of our new pivot table.
On the right hand side you will see your pivot table controls. We will want to add the anchor type to the rows section, and then we will want to add Link Anchor to the values section.
Once this is done we will need to update the function the pivot table performs on the values to be COUNTA.
now as you can see we have a breakdown of each of the anchor text types that this site has coming towards its site.
and now that we have the data broken down we can add some visuals. Click on the insert tab, and then on then toward the bottom you can add a chart to pivot table sheet.
If you'r chart doesn't automatically detect the data in your sheet all you need to do is define it in the chart's data section on the right.
Our data is in cells a1:b6 .. so that is what we will place in the what data input field.
Now we should see a chart with some data in it.
personally I prefer a pie chart for looking at the anchor text distribution as it gives you a better idea of the whole picture. In the chart data section though you have a variety of different charts to choose from though.
From here I like to add a title to my chart so that I can export it as an image, and put it into my notes. This is also something you can send to your clients that looks schnazzy 🙂
Once you have added the title you can click the three dots in the upper right hand corner of the chart and click the save as image button, and it will let you download a png of the chart.
Rinse & Repeat
We will follow this same process for each of the urls that we are modeling our anchor text ratio after. Here is an image I created so we can visualize how this ties in with the search results.
Aggregate Your Results
From here we can begin to aggregate your results. We will take each of the results sets, and place them next to each other in order.
Now we can begin to craft an anchor text distribution that will blend in with the distribution of what google is saying that they like.
Now we have our anchor text distribution ratio.
Lets get an average number of links that it takes to get on page 1, by adding up all of the links each of the url's has and dividing by 4.
Now that we have percentages lets get the distribution of the anchor text that we would need. This would be the anchor text distribution that I would go out and try and acquire based on this research.
Anchor Text Suggestions
We can then get actual anchor text suggestions, and not just the anchor text type suggestion by aggregating all of the anchor text across the domains, grouping them by type, and then showing a count.
To do this merely copy the Link Anchor, and Anchor Type Columns from all of your competitor sheets and combine them into one sheet. From there we will create another pivot table where we can manipulate the data.
Now we have a list of "all" of the anchor text that is being used in backlinks for page 1 rankings. So this gives us an idea of what anchors are "most" important, and then also gives us other suggestions for anchors text we can try.
Example Link Acquisition Table
Here is an example of how I would lay out my links to acquire for a fictitious site of ... BestPingPongTables.guide
If you would like to check out a copy of the google sheet you can check it out here.
This Is Not The End!
I want to make sure that this is clear that this gives us a great place to start.
We have done the leg work to determine what google is signaling that they like, in terms of the amount of links, and the anchor text distribution.
This is NOT a guarantee that if you add these links you are going to rank #1. It is however a way to be systematic in your approach, and allow google to show us a path to how to rank.
I recommend that if you haven't already, you read Matt Diggity's BackLink Blueprint for recommendations on link velocity, and other factors to consider when linking.
Specail thanks to Matt Diggity & Darryl Rosser for their Q&A session that covered this topic. The approach they discuss to systematically analyze backlink profiles, and anchor text ratios based on currently ranking sites, is simple & elegant.
Probably 8 years ago or so I took the in person Bruce Clay (the grandfather of SEO) course, and one thing he said that has always stuck with me.
SEO is not about building out the perfectly optimized site, it is about being the least imperfect.
SEO isn't performed in a vacuum where we can control all of the variables, and we are given very few "rules of the game". Even the rules we are given are ambiguous, and sometimes just outright lies. So the best way we can do our job is to look at the people who are already ranking. Gain as much information as them as possible, and then try and do it better. Then we have to TEST, TEST, TEST.
Hopefully this article helps you understand that there is a systematic approach to selecting your anchor text, that keeps you from feeling like you are playing anchor text roulette.
To all of you who made it to the end ...