Google Authorship – the 3 reasons why it failed

Google intended Authorship photos in search results to convey trust. The average Joe didn’t buy it. Here’s why.

There are so many theories floating around the Internet about why Google Authorship was canned, but let’s begin this article by quoting right from the official announcement:

“Unfortunately, we’ve also observed that this information isn’t as useful to our users as we’d hoped, and can even distract from those results. With this in mind, we’ve made the difficult decision to stop showing authorship in search results.”

In other words:

A) people were not clicking more on search entries with little author pictures attached, and;
B) in some cases people were clicking away from search entries with little author pictures attached

And this was predictable from the start. Hindsight is 20/20 vision, so let’s put on our hindsight goggles and review the three reasons.

  1. Trust and authority differ for different types of searches.
  2. People trust institutions more than strangers.
  3. People select between news and opinion.

The 3 reasons Google Authorship failed

Trust and authority differ for different types of searches

To really discover how the faces in the search results affected the average Joe (not us webmasters and online marketers), we have to reverse engineer it back to the actual searches and their intent. There are various reasons people search

  • They search for something to buy
  • They search for entertainment
  • They search for information

Let’s look at each of these three searchers one at a time.

THE BUYER is looking for a product.  In most cases, the only “authority” on that product is a known brand name.  A face next to a search result means nothing to a buyer.  If he pays any attention to it at all, it is to skip over somebody’s opinion of the product or somebody’s report on how they used the product to play a prank or make Thanksgiving dinner.  Google Authorship kindly flagged your blog post as unhelpful, so that people could skip over it. The buyer is looking to buy.

Unless, of course, the buyer is looking to first research the product, which is the case sometimes when:

  • The product is fairly unknown
  • The product is fairly expensive
  • The buyer is picky or indecisive

The fact is that most people won’t even research a real estate agent before trusting her with their most valuable possession.  Most shopping searches are not looking for reviews.  But some are.  Are they looking for one guy’s opinion, or are they looking for several people’s opinion in one place?  Yes, the big LAZY in all of us searches for a forum thread or a review site like TripAdvisor or ePinions where we can quickly see what several people have to say.  All those search entries with a single face next to them look like a whole lot of extra work for nothing. Google Authorship kindly flagged your blog for people to skip over and save time.

But wait!  What if you saw a trusted face that you recognized?  Someone you knew to be an expert on that product?

Exactly.  How many trusted experts on birdhouses or cookware or hose extensions or bedding or winter boots do you know?  I suppose if you saw Oprah’s face or Martha Stewart’s face or Consumer Reports face…  But two of those are true celebrities, and the other is an institution.  People don’t know your face, so your opinion means nothing to them.  The average blogger’s face in Google search results means nothing to 99.9 percent of searchers.

STOP THE PRESSES!

Who qualifies as a trusted source that most people would click on?  At very least it needs to be someone they know.  Here is a good first triage step: if their name is not in Wikipedia, most people don’t know them.

But even if their name is in Wikiepdia, that does not mean the average person knows them. How many country music stars are listed in Wikipedia? Now how many of them would you recognize if you saw their face in passing among the search results?  (If you are a big country music fan, feel free to replace the words “country music” above with “gastric bypass” or “LEED certification” or “contract negotiations”.)

You see?  There are very few people who are so famous that they are universally known outside of their field, and even fewer whom people might consider to be an authority on a given subject.

RESUME THE PRESSES!

Nobody cares what some blogger or journalist has to say, except those few people who actually know that blogger or journalist. Google Authorship kindly flagged those blog posts for people so they could skip over them.

THE RELAXER  is looking for a video, for humor,  for something to entertain her and help fill some down-time.  She does not want to think.  She does not want to read about entertainment.  She wants to be entertained.  If the faces next to a post are not Lady Gaga or Jimmy Fallon or Scarlett Johansson, it’s just some irritating blah-blah-blah clogging up the search results.  Nothing irritates someone in the mood for a party more than somebody who wants to just talk about partying. Google probably had to dump Authorship just to avoid being called a party-pooper.

THE RESEARCHER is looking for information.  There is some overlap with the other two categories here.

She might be researching to buy something, in which case (as I have already mentioned), she wants good, solid information from the company itself, from a trusted source like Consumer Reports, or from a review site where there are multiple user reviews at once. She couldn’t less what some unknown blogger has to say, and Google Authorship kindly flagged your blog so she would not waste any time clicking on it.

The researcher might be looking for information about entertainment.  Perhaps he loves watching Jimmy Fallon, but right now he wants to know the latest gossip on him.  If that gossip is coming from another well-known entertainer or from Perez Hilton, the face might stop quite a few searchers, and they might click through.

But if they don’t know you (Remember the Wikipedia test?), your face in Google’s search results just flags for them that this is something they can feel free to ignore, since they don’t know you and therefore don’t give a hoot what you have to say.

Many people doing research are not seeking information about entertainment or about products.  Many people just want information, and they want the most accurate and quickest information they can get.  Typical searches for information, and this list is far from complete,  might be:

  • for a recipe
  • about symptoms they are having
  • about nutrition
  • for fitness tips
  • how to build, repair or maintain something
  • for translation or definitions
  • for the latest in a current event (such as a war or a natural disaster or proposed legislation)
  • to fix a computer or software issue

If I am looking for a specific recipe or a recipe that combines certain ingredients or what spices go well with something, I want a recipe site, where there are multiple options all in one place.  I do NOT want to go through a dozen blogs about different people’s personal experience with the ingredients.  Google Authorship kindly flagged those pages, saving me the time I would have wasted clicking on them.

I will skip the one-by-one review of searches about medical information and how to build or repair things and updates about current events, etc.  I assure you that it will get repetitive.  People want solid information that they can trust, and to understand how Google Authorship repels researchers, let’s get straight to the second reason that Authorship failed…

Does a stranger's face convey trust?

People trust institutions more than strangers

You can say that you distrust institutions.  Most people do.

They say don’t trust government.  Yet, they are more likely to believe government information than information from an unknown source.

They say they don’t trust the media, that you can’t trust something just because you read it in the newspaper or see it on TV.  But if they do read it in the newspaper or see it on TV, most people will just automatically assume it’s true.  In fact, there is a whole “As Seen On TV” retail sector based on this simple premise.

They say they don’t trust big business, but ROI on advertising proves them wrong.

On the one hand, people distrust big institutions because they suspect there might be a hidden agenda.  And there often is. On the other hand, they assume that anything big institutions say is based on testing and experiments and scientific proof. And it often is.  At the same time, they assume what some random person says is not based on science or fact, but just some fool mouthing off.

Here would be an interesting experiment (Google, are you listening?):

Imagine a split test in the search results, for a few articles from USA Today or The New York Times. Half of searchers are served up results that include the journalists’ faces.  The other half are served up results with the New York Times or USA Today  logo next to them.  Everything else is random; the actual search queries, time of day, geography, etc.

I wonder how many more people would click on the logo article than would click on the face article.  Remember – it’s the same article, only the visual image would change.

Back to Google Authorship and how people reacted to it, let’s look at an example from the list of information searches in the section above. For medical information, whom would I trust?  I’ll bet you some people would recognize Doctor Weil. Or Doctor  Oz.  Or Doctor Phil.  And many of those people would therefore trust them. I’ll bet you that more people would recognize each of their names than their faces (so the picture probably doesn’t really help increase clicks to their own named websites).   And I’ll bet that many people would not clue in even on their names, much less their faces, so the picture might even detract from them.

As for anybody else, like some health blogger or health reporter for a daily newspaper, would you trust the unknown face over:

  • The Mayo Clinic?
  • A government department with the word “health” in it?
  • A university site with the word “health” in it?
  • A site with the word “doctor” in it?
  • A site with the word “clinic” in it?

Most people will look for some sign of authority, and an unknown face just doesn’t count as a medical authority.

People select between news and opinion

The same goes for other searches, such as updates on current events.  It might be very handy to pull up the results of that New York Times research project I suggested in the previous section. Although I am quite sure I know which of the two identical entries would get more clicks, the important question is who would click more on the entry accompanied by the New York Times logo, and who would click more on the entry accompanied by the journalist’s face?  And, lucky for you, I look into my crystal ball and I know the answer.

Drum roll please….

  • People searching for the latest news – the hard facts – of what happened, will click more on the entry accompanied by the media outlet’s logo.
  • People curious about what the latest developments mean, what the implications might be, what political slants there might be – opinion and analysis –  will click more on the entry accompanied by the journalist’s face.

How do I know this?  Because we have decades of training on how to read newspapers.  The Internet might be a new medium, but we take online our assumptions passed down in the offline world. We have always looked to newspapers to deliver us the news, and we will read the headlines and some of the articles to get the information we want.

There are never any faces attached to those articles.

But there are faces attached to regular columns on politics, international affairs and other topics.  We expect a less”journalistic” style when we read these.  We expect to be challenged to think about the news, not to just read it and accept it.

Flash forward to 2014 (before Google canned authorship, of course) and people searching for news would be predisposed to click on an entry that appeared to be from a trusted news source, such as CNN or BBC or The Globe and Mail. People searching to dig deeper – those prepared to invest some effort thinking about what it all means – will be predisposed to click on an entry with a face.

Big caveat: there are many other factors that will lead people to click through to a given result, including the title and the domain/URL of the article.  But in aggregate, Google authorship would have helped people choose between news and opinion.  Whether it would have done so accurately, I cannot tell.

And whether more people would have chosen to click on news, without the faces, I cannot tell (although I suspect that more people would search for news from a trusted media outlet than opinions of people they don’t know, even if they are interested in opinions).

If my suspicions are correct, Google would have incorrectly seen this as a failure of Authorship.  They likely assumed that faces are not helpful if fewer people click on articles with faces, rather than seeing this as a means of triage helping both news-seekers and opinion-seekers better find what they want.

The Future of Authorship

The real future of Authorship, should there in fact be one, lies in Google better understanding how people view authority for different types of sources.  You and I do NOT have authority beyond out limited niches and networks.  But some people do.  And many institutions do.

I did say a short time ago on a UK marketing blog (My Online Marketer) that:

“Unless Google creates a new “Opinion” search (like the News, Videos and other searches), I suspect that authorship is dead. “

I might not have been completely accurate at the time. If Google can harness this understanding of what “authority” means for various searches and flag individual author expertise and institutional expertise accordingly, it might still be able to help people find the most trusted authorities for a given search.

Or here’s a novel idea: Google could do what it is already doing: trying to float the most trustworthy authoritative pages to the top of its results, where people tend to click through the most anyway.  The face, or the logo, would not give the entry authority – it’s ranking would (and does).

 

How Google reads your backlinks

People spend a lot of time scratching their heads, trying to understand how Google reads their backlinks.  They want to know what links they should seek to their websites that are still “safe”.

With all the turmoil over unnatural links and Penguin penalties over the past year or two, ever more people are sorting through their backlink profiles trying to understand which links to keep and which to try to cull.  What confuses many people the most is why some links would be valued over others.  “Why doesn’t Google like the links I worked so hard to build?”

The problem is that people are used to assuming that:

  • Every link is good.
  • High PageRank is what counts the most
  • Automation is good, because more is better.

These are wrong assumptions.  Remember that Google looks at each link to your website as a vote of confidence or a recommendation.  And not all recommendations are of equal value.  For instance, suppose you need headache medication…

 

Add the Infographic above to your site!

 

If one person recommends a headache medication, you might be inclined to try it. But if several people recommend a different headache medication…yes, exactly.  More is better.

But wait!  What if a doctor recommends a different headache medication.  Yup, authority trumps quantity.  And if several doctors recommend a completely different headache medication…exactly!  More is better, after all, especially when it comes with authority.

Now, what if the drug pusher around the corner offers his recommendation?  No thanks.  But what if a dozen drug pushers all recommend the same headache medication?  Of course you’ll take their advice, because more is better, right?

No way!

And Google is at least as smart as you are.  If hundreds of spammy sites link to your website, that is not a better recommendation than if one spammy website links to yours.  The more “drug pusher” websites recommend your website, the more likely Google is to label your website…

So, just as you would not want a throng of drug pushers recommending your product, make sure there is no throng of spammy websites recommending your website. Google will see more value in your website if inbound links come from trusted or – even better – highly trusted sources.

SEO Shotgun or SEO Rifle?

Or both?

For a huge website (ecommerce, directory, etc.) with many variations of the same product or service, whether by location or by brand, the effort to work individually on each one would be monumental.  For that reason, we often focus on:

a)   The home page, which is naturally where a fair number of links will have to go.

b)   A selection of the most important interior pages (such as those cities which might yield the best ROI) with a purposeful effort to help them rank better for relevant searches.

Some of the activities we do will help just those pages; some will help the entire site.  To understand this better, it helps to understand what types of ranking signals the search engines look for.  They include hundreds of specific signals, but most of them can be grouped as follows:

On-page relevance to a specific search query.

The changes we will make to the template(s) will bring benefits across the site to every page they apply.  In other words, even if we identify 10 city-specific pages on which to focus, every city-specific page will benefit.  If we add text or other elements on a page-by page basis, only the pages we work on will benefit.

Off-page relevance to a specific query.

Links that we obtain to 10 city-specific pages will often (but not always) confer relevancy.  The extent to which this occurs will depend on the content of the page that is linking, the anchor text of the link itself, and a number of other factors.  This relevancy is specific only to the page being linked to.  For instance, a link to the Chicago page of the website confers no relevancy to the London page.

Off-page importance/popularity.

Inbound links to a page also convey “importance” or “popularity”.  They represent a “vote” for the page in the eyes of the search engines.  That importance or that vote is specific to the page that is being linked to.  But, Google’s PageRank algorithm also spread the link-love to other pages that are directly linked. 

For instance, let us assume the Chicago page links directly to other Illinois city-specific pages, such as Rock Island, but not to any Florida city-specific pages.  If we obtain 20 links to the Chicago page, that will greatly boost the popularity of the Chicago page.  It will also boost the popularity of the Rock Island page, but not the Miami page (at least, not noticeably). 

This is why internal linking patterns for a big site like this are so important.

Domain credibility/authority/popularity

This is the exciting part.  Every quality link we build into the domain, strengthens the credibility/authority/popularity of the entire domain.  Every day the domain ages, strengthens the entire domain.  Every time a high-authority site links into the domain, every time there is a social media mention, every time the domain is renewed for a longer period of time…the entire domain – every page – benefits.

So the efforts we make for a few specific pages can benefit them all to some degree.  For a highly competitive sub-niche, that might not be enough.  For a smaller, less-competitive niche, the page might rank well without any direct attention to it.