How Google Authorship decreased our traffic by 90%

http://www.jitbit.com/news/183-how-google-authorship-decreased-our-traffic-by-90/

by scholia • 13 years ago

229 109 13 years ago

pdx• 13 years ago

I wonder if your Google+ Profile picture was your JitBit Name + Logo, you would have done better.

I think it is your face that is the problem. Not because it's ugly, but because it's obviously a face, which is out of context when searching for "macro recorder". We naturally gloss over things that don't sync with what we expect.

spindritf• 13 years ago

I believe Google will only show a "recognizable headshot,"[1] not any image you give them. So you cannot use a logo.

[1] https://plus.google.com/authorship

daniel-cussen• 13 years ago

Use a logo that passes machine pareidolia.

kanamekun2• 13 years ago

This is 100% right. A quick confirmation: "John Mu of Google has confirmed that you must have a headshot of yours in order to see your Google Plus avatar in Google’s rich snippet."

source: http://www.internetmarketingninjas.com/blog/social-media/log...

underdown• 13 years ago

google now accepts a micro data format for logos - although I haven't really noticed them in SERPS yet http://www.hobo-web.co.uk/how-to-put-your-company-logo-on-go...

alternize• 13 years ago

i agree. i have even misread "Marco" instead of "Macro" in the title. with the face next to the content, and me looking for a technical tool and not a person, i'd have skipped to #2 as well.

jonahx2• 13 years ago

Funny, I made the same misreading

axx• 13 years ago

I was like "Recorder is his last name? I bet he has a lot of fun with that!"

100% the same with me!

unreal372• 13 years ago

I do think the actual photo chosen is a turn off though.

I wonder if he tried a better more professional photo if the results would change.

theycallmemorty• 13 years ago

I would love to see someone experiment with headshots of different people with varying degrees of attractiveness.

doomslice• 13 years ago

My brain went through a few things when I saw that result screenshot:

1. I thought it was giving results for a guy named Marco.

2. After realizing it was Macro, I thought the #1 result was a blog or review.

3. I then focused on the 2nd result because it looked less cluttered.

mtgx• 13 years ago

Good observation. So basically, the fact that Google authorship did this is a good thing for people who search Google, because until now his title could "trick" people into clicking on his site (because it was more SEO optimized or for whatever reason), but the "real" result was always the Wikipedia one - because I suppose most people don't have a purchase intent when searching for that, but an intent to find out more information about it.

tonyedgecombe• 13 years ago

I think if you are searching for "macro recorder" then you know what it is and are looking for a solution rather than a description.

darylfritz• 13 years ago

Thoughtbot performed the same experiment; their results seemed favourable: http://robots.thoughtbot.com/post/51068089232/google-authors...

pfortuny• 13 years ago

The photo is a tell-tale to me. Each time I see one on the first search results, they get filtered out by my brain (they are either 'news' or 'youtube').

But this is just my reaction.

epistasis• 13 years ago

Yes, having a photo (or perhaps any image) that modifies the layout of a search result instantly makes that search result look like an ad and not a true search result for me. Particularly if it's the first in the list, then it gets categorized as an advertisement by my brain.

We have been trained by more than a decade of abusive internet ads to ignore them. Does Google have results that show that these abnormally layouts result in an improved number of clicks, I wonder? Is it just us?

jsdalton• 13 years ago

Even though I know what the photo is for, subconsciously it makes me think "Oh, this is going to lead me to some guys Google+ profile page."

I imagine I've skipped over a ton of search results that are just like the one in the article for just that reason.

pvilchez• 13 years ago

This is exactly what I think every time I see it. And sometimes it does take me to a g+ page, so I've trained myself to just avoid it.

Nursie• 13 years ago

Yup.

It screams blog/editorial content. I'm very, very rarely interested in blogs or editorials. Usually I seem to be after hard data or documentation.

unreal372• 13 years ago

Also screams "this site has been SEO'd to death". Better when every top result has a picture then it doesn't stand out so much.

dkroy• 13 years ago

I know this isn't suppose to be funny, but I couldn't help but laugh at how you presented this. I completely lost it where you had written "My Stupid Face" with an arrow pointed towards your google result.

joeblau• 13 years ago

That is pretty fascinating. It seems like Facebook, who I use because this was their early marketing strategy, is teaching us to ignore data that has peoples faces associated with it. I wonder if there is broader research being done on this?

arkitaip• 13 years ago

This is fairly well researched wrt banners. Basically once users associate visual elements with noise, the will start to automatically filter them to the point of not being conscious about their existence.

In a even broader scope, Jakob Nielsen has shared lots of data from his eye tracking usability studies that show just how little attention users pay to online content [1][2].

[1] http://www.nngroup.com/topic/advertising/

[2] http://www.nngroup.com/topic/eyetracking/all/

Kiro• 13 years ago

It's funny how clueless most SEO experts really are.

MrBlue• 13 years ago

Exactly! Stop taking SEOmoz as gospel.

VikingCoder• 13 years ago

Dear Google, please stop content skimmers.

Dear Google, thanks for trying to address content skimmers - but now we may have another problem...

gagan2020• 13 years ago

Firstly, google became a verb and now replacing Dear God... ;)

_k2• 13 years ago

I would love to hear more on this. Are others seeing this problem as well and what photo are they using ?

My first thought when seeing the photo was ok, this is a guy giving his opinion on the topic, I'm not interested in opinion, I want the facts. Can companies act like an author on G+ ?

danso• 13 years ago

Interesting...you think of all the A/B metrics that Google tests, this would be among one of the most obvious standouts. The percentage of users who click on lower-ranked results is reportedly a big factor in the decision of how results are laid out and paginated (e.g. if the 12th result in Google SERP had a relatively similar amount of "usefulness" as the 10th, it's likely that Google would have the search pages show 12 results)...you'd think a statistical drop in first-result clicks would trigger an analytics alarm.

jonahx2• 13 years ago

+1. You have to think many sites besides the OP's experience a similar effect. This, combined with Google's collective UX wisdom and the not-at-all unintuitive nature of the result, makes it pretty shocking that they have not put a stop to this. Maybe they are getting some value from the uptick in people maintaining G+ accounts, so it's being swept under the rug?

ben336• 13 years ago

You're assuming this is generally true. All we know is that it seems to be true for this one guys site/picture. Google has access to ALL of the data for this feature. They're in a much better position to make an informed decision on its value.

Matt_Cutts7• 13 years ago

You're making at least a couple incorrect assumptions. Google launched authorship before Google+, for example.

enjo• 13 years ago

There is clearly a strong internal focus on getting people to adopt G+, however. I wonder how that affects their decision making process here...

tomkarlo• 13 years ago

You can't take one very specific case and then extend it to the entire corpus of searches / web sites. Any change to the SRP might result in less-than-optimal impact on a particular search but if it improves performance for a large number of other searches, it's still a good change overall.

FollowSteph3• 13 years ago

For a lot of people the face gives credibility. It makes you think it's a more legit article because somehow google took the time to associate an image with the article which it would never with spam content.

And it's too bad they don't accept company logos, I think that would really help. It downplays company we pages versus articles.

themckman• 13 years ago

This happened with the Puppet type reference search result. Still feel weird clicking it with the guy's picture next to it. Makes it feel like it's his page when I want to believe it belongs Puppet.

kanamekun2• 13 years ago

The quality of the photo can make a huge impact on clickthroughs. Cyrus Shepard of (seo)Moz saw a 35% increase in click-through rates after a/b testing his profile photo: http://moz.com/blog/google-author-photos

He also mentioned that, "Bounce rate dropped while time-on-site and page views increased. It's as if having an authoritative photo in the search results raised users' trust in my site and expectations of authority."

tokenadult• 13 years ago

This is one of the reasons I asked my Facebook friends to A/B test two photographs I have used as profile photos before setting my Google+ photo. I should probably be a lot more rigorous about this, but at the very least a photograph should be compared to other available photographs for what it says about your online persona.

bmac27• 13 years ago

When you're assessing the viability of microdata or any SEO strategy for that matter, you have to consider the context and whether or not it works for the market that you're in.

I don't see this as an indictment of Google Authorship at all; rather this is an example of a situation where user intention is misaligned with what a webmaster is showing. If I'm looking for content (particularly the originator of said content), having the author's name and face is exactly what I'm looking for. When you're searching for car parts or baby strollers or software, that same name/face is going to throw you off.

mladenkovacevic• 13 years ago

If someone was looking for macro-recorder, they probably just wanted a download link.

His authorship link makes it look like a review of some macro recorder software. I would actually consider an authorship link that's being triggered by a product page a bug that needs to be corrected ASAP.

EDIT: Strangely enough his authorship image is still triggered by a search for "macro-recorder". Did he reactivate it?

Now imagine if Google allowed him to tag his page with a "Software Download" tag.. and enabled him to attach a appstore like icon next to the search result instead of his face. Would that drive more clicks?

massarog• 13 years ago

My question is, why did you add authorship sitewide? Authorship should be established only on jitbit.com/news/ where your blog is, NOT on your entire website. For example, if I search jitbit in google, your authorship photo is appearing for your domain, not good.

graeme• 13 years ago

Yes, this mars the result. I wonder if things are different when it's only on blog posts.

Blog post search results are also more likely to field multiple results with authorship pictures.

Still, worrisome if it's the only top ranked result with a photo. I imagine this will fade over time as people get used to it, took me a few months to figure out what was going on, as a user.

nhebb2• 13 years ago

If you hadn't mentioned this, I wouldn't even know that Authorship could be associated with just a directory. I bet a lot of people, including the OP, fall into that category. If the OP just associated it with his onsite blog (/news/), then essentially the only purpose of Google Authorship would be promote his company blog, not his overall site. So he would be G+ blogging to promote his blog to promote his products. Sounds a bit convoluted to me.

crazygringo• 13 years ago

Not surprised at all. Any time I ever see a social photo next to a search result, I automatically assume it's a social media post or page and won't even bother to look at it, if I'm looking for anything else.

Why on earth would anyone want a face associated with a program, website, or anything that's not a social media account?

But, very good to know the statistical result.

noodle• 13 years ago

I'm in the same boat. If I'm looking for actual information, products, etc., I'll skip the posts with the headshots. It strangely reduces my trust level of the content.

Having said that, we're all in the tech field and this guy's product seems to be geared towards the tech field. Perhaps the result for the population-at-large is different?

muyuu• 13 years ago

I actively avoid mug shots. Face pics are often used to manipulate, esp. if it's "happy faces".

I've developed a rejection to faces in websites. But honestly I didn't expect this to be common. Much the opposite.

PavlovsCat• 13 years ago

As someone who never felt like putting his own mug on his own website, I appreciate that at least for one person out there it's not a minus :)

agrona2• 13 years ago

>Why on earth would anyone want a face associated with a program, website, or anything that's not a social media account?

A coworker runs one of those aggregate-amazon-affiliate-link sites (think http://www.thisiswhyimbroke.com/) for camping and outdoorsy things. He saw a noticeable increase in traffic by associating his wife's picture in his search results. He's of the opinion that a pretty face helps sell things.

mkhaytman• 13 years ago

I've watched that website launch and develop with some interest. Without disclosing any particulars, does the site do well for him? I imagined he was raking in the dough, but if he's still your co-worker I guess that's not the case?

k3n3• 13 years ago

Sex sells....who knew??

nhebb2• 13 years ago

It makes the search result look like it's to a personal site or a blog - not a business. It would be good of Google if they allowed you to associated authorship with your site but opt out of the head shot in search results.

adrr2• 13 years ago

This. IMHO a lot of people are tired of blog spam/social spam and just want the original content.

rhizome• 13 years ago

I would love it if Google figured out a way to surface original content over pages that reference it.

greggman• 13 years ago

I would want a face associated with a program, website etc.

If I search for "digital camera" and see results from friends that are heavily into digital cameras I will likely consider that result.

If I search for "video editing software" or "ramen recipes" or "pizza in sf" I would love to see which things my friends are recommending more than I would like to see what is highest ranked by some algorithm.

k3n3• 13 years ago

I unconsciously ignore anything that isn't in the standard format that the normal results are presented in, a side-effect of so many "sponsored" links always being shoved to the top.

There's the "crap at the top" that is almost never what I'm looking for, and then there's the real results, which is what I scan for.

weinzierl4• 13 years ago

Just as an counter example: I search a lot of Java questions, and there is a guy who has a lot of useful answers. When I see his face in the SERP I often click it without reading the text.

bobsoap• 13 years ago

Yes but you know the guy already. That's branding.

Also, searching for dev advice is not the same as searching for a product like in OP's case.

walshemj2• 13 years ago

for Branding purporses

mhoad• 13 years ago

As someone who for the past several years has done very high end legitimate SEO consulting to fortune 500 style companies (i.e. not the spammy mess than the HN crowd would usually associate with SEO), I am more than happy to go on the record and say that for what it's worth, this is the exception and not the rule.

As a few others here have pointed out, slapping Google Authorship markup all over your site is probably not a good idea unless you run a pure news / blog style site.

One other point to note though as well is that I am very familiar with that message in GWT telling you that traffic dropped by a crazy percentage overnight. Given everything going on in the SEO space at the moment, I am not entirely convinced just yet that you don't have a case of correlation rather than causation at the moment.

gfodor• 13 years ago

The author's takeaways are pure speculation as to causation, but he presents them as if they are facts. I could claim that people just don't like the color of his photo, and I would have as much evidence to support this has he has for his other claims re: authorship placement. And he doesn't mention that since he is not doing a split test (since it's not possible) and he's not measuring confidence intervals that his "A/B test" may be meaningless.

His intuition makes sense, and his conclusions are probably correct in this specific case, but this is shoddy analysis and shouldn't be presented in a generalized manner as he's done here.

6d0debc071• 13 years ago

On seeing what it did, my first thought was that the result with his picture next to it was one of those horrible promoted pages/adverts. I'm used to just skimming over the first results that look different to the rest with Google.

specialp• 13 years ago

Google is basically twisting everyone's arms to join G+. Here is the Google progression of things:

1. Hey we support all open protocols all you need to do is use our FREE service that is superior to all others.

2. OK now you need to unify your profile with Google+

3. Either kill the product entirely or drop the open standard support because the open standard does not support some awesome feature they made.

4.You are stuck in a proprietary world

Matt_Cutts7• 13 years ago

You may want to defend this (incorrect) blog post with that world view, but Google launched authorship markup before Google+: http://www.seoinc.com/seo-blog/google-launches-authorship-ma...

weinzierl4• 13 years ago

Assuming your ordinary niche blogger (no celebrity, no New York Times author, no one with a dedicated Wikipedia page): Is there a way to have a photo next to his or her search result if he or she doesn't have a Google account?

TomGullen• 13 years ago

> most people are clicking #2

Maybe a little nit picky but that's a bit of an assumption especially in the context of the article.

I guess it shows the importance of having more diverse results, you want other results appearing that strongly link to you positively (our official twitter/fb/software reviews for example)

npsimons• 13 years ago

That sucks; sorry to hear it happened to you. But I can tell you why I will usually scan search results (or C-s) for Wikipedia over other results: Wikipedia won't jerk me around. What I mean by that is that (currently) Wikipedia doesn't require JS, flash or any other BS; WP pages load quickly, and are often informative, not some wink-wink-nudge-nudge advertising that may or may not tell you about the product without requiring an email address. Oh, and WP doesn't (currently) have advertising.

That being said, I've been noticing that results returned in Google, even for technical posts (eg, Emacs howtos) have more and more been including headshots from the writer. It's the same posts I would have trusted before, it just seems weird to actually have a face to go along with the post.

205guy• 13 years ago

I think you meant: it just seems weird to actually have a face to go along with the search results.

And when I think about it, that's the whole issue: I'm beginning to think that images (of authors or webmasters or product logos) don't belong on the serarch results page--ever. I can't see a case where it helps the average person who is searching because the photo is always once removed from the actual content.

If I search for a "thingamabob" how does the photo of somebody who wrote about thingamabob help me choose a result? If I'm searching for a company, even the company logo doesn't help because I may not know it ahead of time. Or if I'm looking for a blog, how would I know the face of the blogger ahead of time? I just can't see that many searches where the author's face is relevant to choosing the content you want to see.

Then there is the issue of Google putting the photo in the hottest part of the user's heat-map eye-scan of the search results page. If you put an irrelevant item where the user is looking for immediate relevancy, that result will get skipped (just as the OP postulated and so many commenters are confirming here on HN).

Finally, you have to wonder what Google was really thinking. If images do have an impact on click-through rate, then the images will get SEO'd and become useless. There is a comment on this thread that confirms that putting a woman's picture for the authorhip increased hits. Great, soon every website will appear to be written by a hot babe showing skin (or hot guy depending on the target audience), or if logos are allowed, all competitors will have a logo that looks like the #1 in the field.

Google is killing the golden goose.

Matt_Cutts7• 13 years ago

Authorship had nothing to do with this site's drop. The site has been affected by our Penguin webspam algorithm and that accounts for the drop. See http://www.mattcutts.com/blog/penguin-2-0-rolled-out-today/ for details of the recent Penguin launch, or http://googlewebmastercentral.blogspot.com/2012/04/another-s... for more information about Penguin in general.

mattmanser3• 13 years ago

Still #1 when I search and that's not even what he's talking about.

He's claiming that having a G+ picture inserted next to search results is resulting in a much lower CTR than they previously had.

Which I think is plausible as it detracts from the search experience, I often skip those results as the text isn't lined up so I can't scan it properly. Wish you could turn them off, it's really off putting. I also wonder what the CTRs of pretty people are like compared to munters. I still don't get why Google added it.

anoncow• 13 years ago

Authorship ties Google+ to search. Can I have my photo next to my search result if I don't have a Google account?

Matt_Cutts7• 13 years ago

I'm traveling, so it will be a while before I can check with the team, but the initial message that kicked off his investigation referred to clicks to his site, which I believe can include overall traffic to his site. I'll double-check when I'm back from traveling though.

I was also responding to the overall headline of his site ("How Google Authorship decreased our traffic by 90%") because there are other reasons for the decline in traffic to his site. The headline is definitely overstating things; the article itself concludes "Turns out, it's not that simple. Some pages moved down, but some have improved drastically." And on the graph he shows, the worst CTR decrease is -57% while the best CTR increase is 329%.

So I think the headline of this article ("How Google Authorship decreased our traffic by 90%") remains inaccurate.

adrr2• 13 years ago

His position on the SERPs didn't change, he's still #1 for the keywords that were driving the traffic.

ignostic• 13 years ago

Ooh this is embarrassing; the article was about click-through rates and not rankings. But maybe now is a good time to tell people how terrible the CTR data in WMT is.

Also, are you sure the site was hit by Penguin? This doesn't look like the typical Penguinized site. You would know, but have you checked?

Matt_Cutts7• 13 years ago

Yes, I did check that the site was affected by Penguin before I posted my comment.

uts_2• 13 years ago

These guys seem to have no problem with manipulating the search results will hundreds of thousands of paid links:

https://ahrefs.com/site-explorer/backlinks/subdomains/jitbit...

So it's really no surprise that Google's latest Penguin rollout has hurt them.

Sad to see it blamed on authorship.

jitbit6• 13 years ago

Hi, the original post author here.

Hundreds of thousands of paid links? Excuse me? We do NOT pay for links. The only thing I can imagine has initiated Penguin - we sell web-software (forum software, helpdesk software etc) that has a footer link "powered by XXX". We dis see this affected our rankings a year ago and we're still trying to recover (making those links "nofollow" trying to contact the linking webmasters etc).

jitbit6• 13 years ago

Hi Matt, the original post author here.

Our site WAS affected by Penguin indeed, even by the first version of Penguin a year ago. Because we sell web-forum software and ticket-software - that both have a "powered by" link at the bottom, our SEO agency advised to add that...

And we're still trying to recover... I'm contacting our clients one-by-one and we're changing those links to "nofollow".

We have never paid for links, the only paid ADs our agency buys are CPC-campaings on download sites like "download.com" etc... Can this be a reason? Should I tell them to stop doing that?

UPDATE: anyways THANK YOU MATT for commenting this and letting us know we're penalized by Penguin... Not many people are lucky to have Matt Cutts look at their issues!! PS. I guess we should fire these guys, remove/disavow all links and start over...

b1daly• 13 years ago

This article made me realize that I am unconsciously filtering out things that look like G+ posts. In part because my expectation of it being high quality content is low, and because the experience of switching to the G+ app on my phone is slow.

onemorepassword• 13 years ago

So basically people are learning that the first Google result, especially if visually "enhanced", is usually irrelevant advertising?

That would concur with my personal experience, and is pretty much the same as what happened to standard size display ads.

iconicmind• 13 years ago

I wonder if niche also plays a part in this - perhaps for certain search queries it is best to have a face associated with the result. Thakns for this post though, it's some food for thought.

EGreg• 13 years ago

Why don't you change your icon to the icon of your program or something users associate with what they need? A face says "social" - which is great for your personal Google+ account but not great for your company. At least try it -- if everyone is like me it should increase the clicks.

humanspecies• 13 years ago

The WWW desperately needs a free and open search engine that everyone can hold accountable. The WWW is much too valuable to hand over to a company with blatant conflicts of interest in their search results.

kcbanner• 13 years ago

aerolite• 13 years ago

Really, people will most likely skip the #1 search result? Please, teach me web marketing.

stefanve• 13 years ago

My own experience is that I would skip it but not because it is an possible ad. Especially if I'm looking for a product I will look at the ads, sometimes the ads are better than the actual result.

But there is a problem with the SERP, and that is that the result under the ads (if any) is sometimes a related sub search.

So if i'm looking for some thing scientificy it will show me some sub search of Google scientific results. Or if I'm searching form some thing that has been in the news recently it will show me some Google news results. Now unless I'm looking for papers or news I will skip over it. Since it shows your face next to the link it looks like an news sub search result or indeed a blog result.

erikb• 13 years ago

I am more likely to click a link with a "stupid" face in front of it, because it looks more qualitative to me. More I simply don't know about this topic or SEO. But maybe because most people know as much as me, I guess that most people will react the same way.

How Google Authorship decreased our traffic by 90%

Related Stories