top | item 5792268

How Google Authorship decreased our traffic by 90%

229 points| scholia | 13 years ago |jitbit.com | reply

109 comments

order
[+] Matt_Cutts|13 years ago|reply
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.
[+] jitbit|13 years ago|reply
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...

[+] adrr|13 years ago|reply
His position on the SERPs didn't change, he's still #1 for the keywords that were driving the traffic.
[+] ignostic|13 years ago|reply
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?

[+] mattmanser|13 years ago|reply
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.

[+] crazygringo|13 years ago|reply
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|reply
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?

[+] agrona|13 years ago|reply
>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.

[+] k3n|13 years ago|reply
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.

[+] nhebb|13 years ago|reply
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.
[+] pdx|13 years ago|reply
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.

[+] alternize|13 years ago|reply
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.
[+] unreal37|13 years ago|reply
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|reply
I would love to see someone experiment with headshots of different people with varying degrees of attractiveness.
[+] pfortuny|13 years ago|reply
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|reply
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|reply
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.

[+] Nursie|13 years ago|reply
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.

[+] gfodor|13 years ago|reply
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.

[+] mhoad|13 years ago|reply
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.

[+] massarog|13 years ago|reply
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.
[+] nhebb|13 years ago|reply
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.
[+] graeme|13 years ago|reply
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.

[+] kanamekun|13 years ago|reply
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."

[+] doomslice|13 years ago|reply
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|reply
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.
[+] danso|13 years ago|reply
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.
[+] enjo|13 years ago|reply
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|reply
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.
[+] jonahx|13 years ago|reply
+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?
[+] VikingCoder|13 years ago|reply
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|reply
Firstly, google became a verb and now replacing Dear God... ;)
[+] specialp|13 years ago|reply
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

[+] mladenkovacevic|13 years ago|reply
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?

[+] bmac27|13 years ago|reply
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.

[+] FollowSteph3|13 years ago|reply
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.

[+] _k|13 years ago|reply
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+ ?

[+] themckman|13 years ago|reply
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.
[+] Kiro|13 years ago|reply
It's funny how clueless most SEO experts really are.
[+] MrBlue|13 years ago|reply
Exactly! Stop taking SEOmoz as gospel.