#mm-preview-settings-bar { z-index: 100000; }
SEO Mythbusting: Does Google Have a Political Bias?

SEO Mythbusting: Does Google Have a Political Bias?

SEO is a field where testing assumptions can reveal truths about optimizing your website. Busting SEO myths helps us make better business decisions and get more from investments in SEO.

The last couple of years have been an interesting time for search and politics. Not separately, mind you — I am referring to the politics of search. On several occasions, Google has been accused of political bias — most recently and energetically by President Trump — and Google has even spoken before Congress on the subject.

I realize any political topic will get touchy, but I could not think of a more fun subject to try to test empirically.

The idea of explicit (or unintentional) biases in search results are concerning for many reasons; with so many people using search every day, the results they encounter can have huge political, social, cultural, or financial influence. The topic of bias in search has long garnered public interest, but in 2018 it reached an all-time high in web searches, according to data from Google Trends.

Donald Trump Tweet: Stop the Bias

It’s not much of a secret that Eric Schmidt, former Executive Chairman of Google and (at that time) Alphabet, both endorsed former President Obama and had a close relationship with his White House and administration. Additionally, it‘s known that Mr. Schmidt was an avid supporter of Hillary Clinton’s campaign.

Of course, there is nothing inherently illegal or unethical about this, nor should there be; people are allowed to have and support their political beliefs. I can, however, see why folks on the other side of the political aisle may be concerned.

Add to this the fact that it was discovered employees of Google had suggested they should adjust search results to favor specific websites in response to President Donald Trump’s immigration travel ban, and you have the makings of a really great conspiracy theory.

While all of this political intrigue may be fun to discuss and argue about, the discussion really boils down to one thing: is there a political bias coded into Google’s algorithm?

The more I thought about testing this, the more I realized that it could become an enormous undertaking. While I initially mapped out over 25 separate tests I could run, I determined that I needed to focus and test individual variables.

I decided to first test if the algorithm has a general bias against President Trump, and then, if it prefers his opponents over him.

Testing for Political Bias in Google’s Algorithm: The Process

In the first test, I wanted to see if Google would promote or punish a webpage based on mentions of a specific person. I first wanted to determine if the algorithm contains a political bias, or if it treats everyone equally.

To run this test, I created seven pages with exactly the same content — titles and meta descriptions need to be unique or else Google will not index the page. I then came up with a unique fake keyword that displays zero results on Google when searched.

Lastly, on all the pages, I needed a control name. I created a fake president name, “John Q. Smith,” and placed it in the title, meta description, and four times within the body copy. As one more note, I decided to mix up usage of the name. Instead of just typing “President John Q. Smith,” I used variations such as “President Smith,” “Mr. Smith,” and “John Smith.”

After the pages were launched, indexed, and ranking for the fake keyword, I noted which page was ranking in the #4 position out of the seven pages for the fake keyword. After about a week, the rankings usually settle down and a specific page typically ranks consistently in the 4th position. Once this occurred, I edited the page ranking 4th to change all the “John Q. Smith” references to “Donald J. Trump” — and then I waited.

If Google results are biased against the President, then the page should drop in the rankings.

Within several days, the page went to the #1 position for the fake keyword. Did this mean Google is biased towards President Trump? I found this equally unlikely, so I created two more variations of this test to confirm.

For each of these variations, I followed the same procedure. Each set had the same content, except for the title and meta description. Each set also had its own fake keyword, and I used “John Q. Smith” as the control name placeholder. For the first two sets, I decided to test against the well-known Republican personalities Rush Limbaugh and Sean Hannity.

Once all the pages were indexed, I waited for the 4th position ranking to settle. After it settled, I updated those pages by adding the new variable names in each set — this took about a week. Both the Rush Limbaugh and Sean Hannity variable pages moved to #1.

Did I just uncover a right-wing conspiracy buried within Google’s algorithm? While I found this even more unlikely than a vast left-wing conspiracy, I decided to create another two test variations.

I set them up just like the last set, but this time I used the names Barack Obama and Hillary Clinton as my variables. Just like before, after indexing, updating, and waiting, both the updated variable pages jumped from position #4 to position #1 for their respective fake keywords.

The interesting thing about testing is that I often learn unexpected things about the search engine. In this case, Google is favoring my updated variable pages. To me, this indicates that Google favors “freshness,” “uniqueness,” or both. Any real person would agree that these test pages could hardly be described as fresh or unique, but to a computer, they are fresher and more unique than the control pages. I have seen this in past tests and actually expected to see this result, but decided to run them anyways to establish a baseline.

So, test set #1 did not expose bias against Donald Trump, nor any of my test subjects for that matter. Therefore, from a larger context, Google is not automatically demoting content about our President.

The next argument could be that Google may be biased against, or biased in favor of, specific people. For example, it would rank former president Barack Obama over current president Donald Trump. I had my next test.

I set up the second test very similarly to the previous test set: seven identical pages, except for titles and meta descriptions, and a fake keyword on each. This time, instead of a fake placeholder control name, I used Barack Obama as my control.

After the pages were indexed and the rankings settled, I took the page ranking in the 4th position and replaced Mr. Obama’s name with Mr. Trump’s name. After about a week, the page with Donald Trump’s name ranked #1. I then performed the exact same test, but in reverse: Donald Trump’s name was the control, and Barack Obama’s name was placed on the page ranking in the 4th position. Again, the updated variable page, with Mr. Obama’s name, jumped to the 1st position for the fake keyword.

I wanted to be extra clear, so I decided to run two more tests. I called these #2C and #2D, to help keep all these tests straight. This time, I ran the exact same tests as already described, but instead of using former president Barack Obama’s name first as the control and then as the variable, I used Hillary Clinton’s name instead.

Once again, the page with the variable name ultimately ranked 1st for the fake keyword. Therefore, the Donald Trump variable beat the control pages — the Hillary pages — and the Hillary Clinton variable pages beat the Donald Trump control pages. I was definitely seeing a pattern here.

After running nine separate tests (five using test #1, and four using test #2) the page that was most fresh and most unique always ranked in the first position, regardless of whose name I used.

The Findings

What, if anything, have we learned?

First, the results seem to confirm the validity of Google’s never-ending counsel that we build the best content possible. The base algorithm appears to be aimed at ranking the freshest and unique content available.

I spent six weeks trying to determine if Google is biased, and I learned that the adage “content is king” seems to be true. That is what I like about testing: sometimes I learn something profound, and sometimes you just reinforce the basics.

Secondly, I would suggest that the results demonstrate Google’s algorithm is likely not overtly biased against any person. While I will agree many employees of Google clearly have a political preference, so far I am happy to say that I see no indication of a political bias programmed into the algorithm. It could be biased in a more subtle way — which is why I am running further tests.

I am currently running tests on sentiment. The control pages contain either negative or positive sentiment, and the variable page is the other. I will test if there is indication that a page with positive sentiment towards President Trump will hurt or help the page in the rankings. Regardless of the results, I will run it against either Obama, Clinton, or both as well.

I suspect the results will be consistent with the tests I’ve already run — but you never know for certain!

Charles T on Twitter
Charles T
Charles T
Charles has been actively involved in online marketing since 2000. For the past 15 years, he's focused on SEO in a number of B2B and B2C verticals – legal services, eCommerce, information marketing and affiliate marketing. He is currently the SEO Manager for Verizon's Fios division. Charles is always looking for new ways to help new and established companies solve their SEO challenges.
SEO Mythbusting: Can Google Recognize All Phone Number Formats?

SEO Mythbusting: Can Google Recognize All Phone Number Formats?

How do you know what you know about search engine optimization?

Think about it for a second. Even those of us that have years of experience can only optimize so many websites — meaning in the course of our work, we sometimes encounter issues we may have never been exposed to and must look beyond the scope of our own first-hand knowledge to determine the best course of action. Often, we base knowledge on what we have heard or learned from trusted sources. But is this the best method for our SEO knowledge? In a court of law, this would be called hearsay and would not be an acceptable form of evidence upon which to make important conclusions.

I have found that this is often where SEO myths come from. A statement is made, and then repeated so often that it becomes a “fact.” What makes this dangerous is that often we base business decisions for our own websites, or even worse, client websites, on these often-repeated myths.

We can put an end to this cycle, but it requires us to be skeptical of everything we hear. We must put all of it, including our beliefs, through a rigorous testing process — one that should state the hypothesis we are testing, and eliminate as many variables as possible.

To illustrate my point, let’s go back to July 10th. An individual tweeted at John Mueller and Danny Sullivan; he wanted to know the best format to use when listing his company’s phone number online. Obviously, this is important: if customers are not able to find your number, they are likely to go to your competitors instead.

In Google’s typical fashion, John Mueller gave this vague reply:

Maybe it’s me, but I felt that this didn’t really answer the question.

The exchange was picked up by Barry Schwartz from Search Engine Roundtable, with an article headlined “Google Implies Phone Number Formats Don’t Matter.” The article states: “I suspect Google can figure out many types of phone number formats based on your preferred style.”

A simple question received a vague reply, and then the next person extrapolated meaning that may or not be there. Now, this article may be used as “evidence,” when in fact, nothing was actually proven. This makes a great scenario for some SEO mythbusting!

The Test

So, the hypothesis is simple: Google can recognize many types of phone number formats. For this test, I setup 10 HTML pages and used the same content in each (to eliminate variables), except on each page, I included a phone number in a different format.

I tested the following formats:

  1.       ##########
  2.       ###-###-####
  3.       ###.###.####
  4.       (###)#######
  5.       (###) #######
  6.       (###)###-####
  7.       (###) ###-####
  8.       ### ### ####
  9.       (###) ### ####
  10.       (###)### ####

The first thing I’d like to point out is this can be an arduous process. Testing takes time, it is a pain to document (notes that are clear the day you wrote them often seem like gibberish a month later), and once you are done, you always notice things you forgot. In this case, I forgot to try the format ###_###_####. I’m running it now, so I’ll have the updated results in a week or so…but it still bugs me that I forgot.

I ran this test three times, using a different phone number in each set.

For the first two tests, I included the copy, “Feel free to contact or call my phone anytime at”. I did this because I assumed that Google needed some kind of context to know that those numbers were a phone number.

On the third test, I just placed the phone number in its own paragraph tag (<p>).

I waited for the pages to be crawled, indexed, and cached by Google — this took about two weeks. Afterwards, I ran a search for each phone number format in both the general results, and using the “site:” command to include only my testing domain.

As SEOs or business owners, we cannot control how our customers or clients type our phone number into the search engines. Therefore, we need to identify the format that Google returns most often. The results I got were fascinating.

The first thing I learned is that Google did not need the extra text for context. The results for all of the tests were almost identical, whether it included copy or just a naked phone number. Second, while Google is pretty good at figuring out most phone number formats, there are definitely formats it prefers and formats it does not recognize as a phone number.

The phone number format you should avoid at all costs is #3: [###.###.###]. For none of the other 9 format searches did this page ever get returned (even when using the “site:” command). The only time Google returned the page containing this format is when I used this format. And when I did the search with this format, none of the other pages were returned (again, even when using the “site:” command). I think that clearly shows that Google does not recognize this as a phone number.

There were two other formats that Google did not rank in most cases: #4 [(###)#######] and #5 [(###) #######]. I suspect the lack of a dash or space gave Google trouble. Without those, it was not able to parse the numbers to understand that it is a phone number. I would suggest not using those format either.

What format should you use? Simple: formats #2 [###-###-####] or #7 [(###) ###-#### ] always ranked the best for every phone number format search. Unlike the previous two examples that Google had trouble with, these were parsed with a dash between the prefix and line number — a format that Google understands to signify a phone number.

I would then give an honorable mention to formats #6 [(###)###-####], #8 [### ### ####], #9 [(###) ### ####] and #10 [(###)### ####]. In most of my test searches, the pages containing these formats would appear…but not always. I also noticed that if the pages were returned by Google, they were always rankings behind the winners (#2 & #7).

So where does this all leave us? If your phone number were placed on a website, I would always request they use one of the two “winning formats.” I suggest you pick one of those two formats and stick with it, so that all your citations are consistent across the internet. If a site uses formats #6, #8, #9 and #10, I would not worry at all, Google will likely recognize it. Lastly, if a site uses formats #3, #4 or #5 I would ask them to adjust it. If it is a directory and that is their standard format, I may consider bypassing it if it’s a cumbersome submission process, or they require payment for inclusion. It likely won’t be worth your time or money.

After performing all these tests and discussing the results with my team, I was asked by one of my SEO Specialists if I had considered using the HTML telephone link code. To be honest, I had forgotten all about this code.

For those of you, like me, who had forgotten about this code, or were unaware of it, it is code that allows you to click or tap on the phone number and have your mobile device or desktop app make a call — also sometimes referred to as “click to call.”

The code looks like this:

<a href=tel:+1-###-###-####>phone number or text goes here</a>.

I decided to take one of my test sets and add these variables. First, I wrapped the #3 format (the worst performing format) in this code: <a href=tel:+1-###-###-####>###.###.####</a>. Then, I added an additional page using just text: <a href=tel:+1-###-###-####>click to call</a>.

It has been over a week and in both cases, neither of these pages are returned for any of the different phone number searches. So far, it appears that Google does not use the content of that code for ranking purposes. But, I’ll keep these pages up and monitor if anything changes.

The Takeaways: What We Learned

Let’s take a step back now and think about where we started, and where we have come.

We began this journey with a question, and vague answer by a Google employee. That answer was interpreted one way by a reliable resource, the interpretation was shared with other SEOs, and it became a “fact”. However, my tests have demonstrated that this “fact” is anything but.

Do not accept anecdotal evidence; strive for SEO truth by using rigorous testing processes.

Let me know what you thought about my test — did I miss something? If anyone wants to try to replicate and verify my test, please do. These testing processes help prevent us from accepting SEO myths as facts. It grants us a better understanding of Google’s algorithm, and makes both our sites and our clients’ sites stronger.

And at the end of the day, this understanding allows us to make better business decisions, and identify opportunities to improve our sites in ways that competitors haven’t. The decision to dive deeper, instead of taking information at the surface level, can make all the difference in the SERPs.

Charles T on Twitter
Charles T
Charles T
Charles has been actively involved in online marketing since 2000. For the past 15 years, he's focused on SEO in a number of B2B and B2C verticals – legal services, eCommerce, information marketing and affiliate marketing. He is currently the SEO Manager for Verizon's Fios division. Charles is always looking for new ways to help new and established companies solve their SEO challenges.
SEO Mythbusting: Does Google Recognize Click-to-Call as a Phone Number?

SEO Mythbusting: Does Google Recognize Click-to-Call as a Phone Number?

SEO is a field where testing assumptions can reveal truths about optimizing your website. Busting SEO myths helps us make better business decisions and get more from investments in SEO.

In Issue #4 of PAGES, I busted a myth about phone number formats and their impact on search rankings. We determined that the format used for a phone number does impact the way Google recognizes the phone number.

As is often the case with SEO testing, I invariably forget something.

In the last issue, I tested the best telephone formats to use on the web so that Google can understand them.  If you did not read that article, I suggest you get your hands on the Q3 issue of PAGES SEO Magazine and review it.  

I mentioned that I had not tested the phone number format ###_###_####.  I added this format to the three iterations of my test grid. After a couple weeks, the results were in.  

While that format did not perform as well as the “winning formats”, it did deserve an honorable mention.  So again, I would not use ###_###_#### on my website but if forced to use that format on another site I would not worry too much – Google will likely figure it out.

The second test variation I had not considered was wrapping the phone numbers in the HTML telephone link code:

<a href=tel:+1-###-###-####>

I tried two different variations of this test. First, I wrapped this code around the “losing format” of the phone number (the one using periods – ###.###.####). The second variation was to wrap the code around just text.  The phone number never appears on the page, only within the telephone code. If Google reads the number in the code and uses it for on-page optimization purposes, the pages should rank in the search results.

In both cases, Google did not seem to use the phone number as an on-page optimization factor.  In the first case, the “losing format” never ranked for any of the other phone number format searches, only searches explicitly for ###.###.####.  In the second case, (text wrapped in code) the page never appeared for searches for the phone number – even when using the “site:” command. I also tried searching for the full phone number “1-###-###-####” and even “+1-###-###-####” both without and with the “site:” command – again, no results.  

I feel this demonstrates that Google does not use the telephone code for on-page optimization purposes.

I hope this test helps you make more informed decisions about the content on your site! This is always the goal of SEO testing. If you have questions or comments about the test, let me know on Twitter: @CharlesHTaylor

Charles T on Twitter
Charles T
Charles T
Charles has been actively involved in online marketing since 2000. For the past 15 years, he's focused on SEO in a number of B2B and B2C verticals – legal services, eCommerce, information marketing and affiliate marketing. He is currently the SEO Manager for Verizon's Fios division. Charles is always looking for new ways to help new and established companies solve their SEO challenges.