The article identified here indicates that a substantial bias by Google Search exists and is obvious and measurable. The bias is in essence censorship of conservative websites (making information that they present difficult to find or causing links to such information on conservative websites to be effectively absent from search-return lists) versus ranking items that fit the liberal agenda to be high and at the top of search-return lists.
The article analyzes that issue, but the analysis is a bit shallow. Some examples are indicated in the article, but those are in essence anecdotal evidence, although some efforts were made to make the study sample somewhat representative, to compensate for the sample not having been randomly selected.
The article is not necessarily easy to comprehend even for someone who makes it through the first sentence and understands what it means:
“The percentage of domain traffic, referred by Google Search, net of brand searches (PGSTN), tends to be in or around the range 25%-30% for a broad class of web domains.”
That is a mouthful, somewhat correct but not very enlightening. Rephrased, this definition of PGSTN puts it more into the context in which it is being used:
»PGSTN, a term used extensively by Google Analytics. It appears to stand for: The Percentage of all traffic directed to a website or domain (URL for a website or for a specific web page) that originates from Google Search Traffic, Not including views or visits that may have resulted from directing traffic to specific instances of brand names.«
I spent more than 20 hours of searching for an authentic definition of PGSTN and failed to find one. In the absence of an authentic, official, definition by Google, I feel that it is fairly safe to assume that PGSTN is an abbreviation of these terms: Percentage… Google Search Traffic, Not…
Absent a better definition from Google, that what I use by which to remember the meaning of PGSTN.
The indicated article identifies that PGSTN for websites is “in or around the range 25%-30% for a broad class of web domains.” That is the range of the portion of Google-Search-related traffic out of the total of all non-ad/or-brand-name-related web traffic directed to a domain or URL.
That percentage is important for any web ranking service (Alexa – now owned by Amazon – is one service that ranks websites), while search engines rank individual web pages and their URLs based on how frequently they are being accessed through search engines. It appears that web ranking services rely on search engine statistics, and that Google Search results play an important role in giving them the data from search-engine-originated traffic with which to do the ranking.
There is no correlation between actual traffic counts of daily views (metered by WordPress) and the portion of the traffic that represents daily views directed to blog.fathersforlife.org through Google searches (PGSTN, metered by Google Analytics). If anything, whenever the PGSTN experiences a rising trend, something is at work to depress the increase in traffic, so as to bring it closer to zero. (Note, the chart will be updated now and then, to reflect what Google is up to and how “accurate” its traffic data are over time.)
My website (fathersforlife.org) and blog (blog.fathersforlife.org) can be called conservative, although some opinions hold that the website and blog present a libertarian perspective. They are obviously not liked very much by Google, as the PGSTN for the domain fathersforlife.org was five percent or less for the last few weeks of all search-engine-directed traffic, with the website and the blog effectively being blacklisted by Google. That will of course not be much of a problem, the less of the traffic comes through Google Search, and the more of it comes through links or other search engines, such as DuckDuckGo, the search engine I preferably use.
Mind you, the Google Search bias will be a problem if a website-ranking service relies solely or primarily on traffic data based on Google-Search traffic information obtained from either Google search-engine crawlers or from search-engine crawlers that collect traffic data collected by Google for a given website, blog or any other domain.
Other than that, I see no good reason anyone should rely on either Google Search for finding objective sources of information, as Google Search parameters and algorithms appear designed to prevent anyone from finding such objective sources. Moreover, there is no good reason anyone should bother analyzing the characteristics of traffic to his website using an analysis tool such as Google Analytics, when that tool obviously provides results that reflect very distorted and unrealistic information on as little as 5 percent or even none of the traffic directed to a given website of interest.
At any rate, the essay by Leo Goldstein states: “Google is very good at its job. Sites and domains that are less popular with the visitors tend to be less likely to receive traffic from Google, and vice versa.” I agree.
It seems that a feedback effect is at work, when Google does its best to ensure that it keeps visitors away from websites it deems people must be kept away from. The good thing about that is that the fewer visitors come to a website through Google searches, the less harm Google can do to influence the traffic and ranking of a given website, except for ranking by services that rely on Google traffic data for websites.
It remains to be seen what the impact will be on Google’s business prospects in the long run, on account of Google doing serious harm to the extent of trust people put into Google.