In the past 2 months Google made big changes to its search results.
Webmasters saw disappearing Google authorship photos, reduced video snippets, changes to local packs and in-depth articles, and more.
Here at Moz, we've closely monitored our own URLs to measure the effect of these changes on our actual traffic. The results surprised us.
Authorship traffic—surprising results
In the early days of authorship, many webmasters worked hard to get their photo in Google search results. I confess, I doubt anyone worked harder at author snippets than me.
Search results soon became crowded with smiling faces staring back at us. Authors hired professional photographers. Publishers worked to correctly follow Google's guidelines to set up authorship for thousands of authors.
The race for more clicks was on.
Then on June 28th, Google cleared the page. No more author photos.
To gauge the effect on traffic, we examined eight weeks' worth of data from Google Analytics and Webmaster Tools, before and after the change. We then examined our top 15 authorship URLs (where author photos were known to show consistently) compared to non-authorship URLs.
The results broke down like this:
Change in Google organic traffic to Moz
- Total Site: -1.76%
- Top 15 Non-Authorship URLs: -5.96%
- Top 15 Authorship URLs: -2.86%
Surprisingly, authorship URLs performed as well as non-authorship URLs in terms of traffic. Even though Moz was highly optimized for authors, traffic didn't significantly change.
On an individual level, things looked much different. We actually observed big changes in traffic with authorship URLs increasing or decreasing in traffic by as much as 45%. There is no clear pattern: Some went up, some went down—exactly like any URL would over an extended time.
Authorship photos don't exist in a vacuum; each photo on the page competed for attention with all the other photos on the page. Each search result is as unique as a fingerprint. What worked for one result didn't work for another.
Consider what happens visually when multiple author photos exist in the same search result:
One hypothesis speculates that more photos has the effect of drawing eyes down the page. In the absence of rich snippets, search click-through rates might follow more closely studied models, which dictate that results closer to the top earn more clicks.
In the absence of author photos, it's likely click-through rate expectations have once again become more standardized.
Video snippets: a complex tale
Shortly after Google removed author photos, they took aim at video snippets as well. On July 17th, MozCast reported a sharp decline in video thumbnails.
Most sites, Moz included, lost 100% of their video results. Other sites appeared to be "white-listed" as reported by former Mozzer Casey Henry at Wistia.
A few of the sites Casey found where Google continues to show video thumbnails:
- youtube.com
- vimeo.com
- vevo.com
- ted.com
- today.com
- discovery.com
Aside from these "giants," most webmasters, even very large publishers at the top of the industry, saw their video snippets vanish in search results.
How did this loss affect traffic for our URLs with embedded videos? Fortunately, here at Moz we have a large collection of ready-made video URLs we could easily study: our Whiteboard Friday videos, which we produce every, well, Friday.
To our surprise, most URLs actually saw more traffic.
On average, our Whiteboard Friday videos saw a 10% jump in organic traffic after losing video snippets.
A few other with video saw dramatic increases:
- Domain Authority: +157%
- MozBar: +77%
- Learn SEO: +43%
The last example, the Learn SEO page, didn't have an actual video on it, but a bug with Google caused them to display an older video thumbnail. (Several folks we've talked to speculate that Google removed video snippets simply to clean up their bugs in the system)
We witnessed a significant increase in traffic after losing video snippets. How did this happen?
Did Google change the way they rank and show video pages?
It turns out that many of our URLs that contained videos also saw a significant change in the number of search impressions at the exact same time.
According to Google, impressions for the majority of our video URLs shot up dramatically around July 14th.
Impressions for Whiteboard Friday URLs also rose 20% during this time. For Moz, most of the video URLs saw many more impressions, but for others, it appears rankings dropped.
While Moz saw video impressions rise, other publishers saw the opposite effect.
Casey Henry, our friend at video hosting company Wistia, reports seeing rankings drop for many video URLs that had thin or little content.
"...it's only pages hosting video with thin content... the pages that only had video and a little bit of text went down."
- Casey Henry
For a broader perspective, we talked to Marshall Simmonds, founder of Define Media Group, who monitors traffic to millions of daily video pageviews for large publishers.
Marshall found that despite the fact that most of the sites they monitor lost video snippets, they observed no visible change in either traffic or pageviews across hundreds of millions of visits.
Define Media Group also recently released its 2014 Mid-Year Digital Traffic Report which sheds fascinating light on current web traffic trends.
What does it all mean?
While we have anecdotal evidence of ranking and impression changes for video URLs on individual sites, on the grand scale across all Google search results these differences aren't visible.
If you have video content, the evidence suggests it's now worth more than ever to follow video SEO best practices: (taken from video SEO expert Phil Nottingham)
- Use a crawlable player (all the major video hosting platforms use these today)
- Surround the video with supporting information (caption files and transcripts work great)
- Include schema.org video markup
SEO finds a way
For the past several years web marketers competed for image and video snippets, and it's with a sense of sadness that they've been taken away.
The smart strategy follows the data, which suggest that more traditional click-through rate optimization techniques and strategies could now be more effective. This means strong titles, meta descriptions, rich snippets (those that remain), brand building and traditional ranking signals.
What happened to your site when Google removed author photos and video snippets? Let us know in the comments below.
Hey Cyrus,
Every SEO needs to read & re-read your last paragraph "SEO finds a way".
We're now back to "words" and how it was when Google first appeared as the new kid on the block. Back to creating enticing meta descriptions, bolded titles, brand-building and ranking signals. But, goodness me, did some think they had ever left?!
Well before video snippets and authorship mug-shots dazzled us, companies ranked for key search terms and sold (or gained leads) online.
One of the key (mandatory) skills of an SEO is to adapt to enforced changes and that's what forces us (willingly or un-willingly) to re-assess, re-examine and move forward (for the sake of our companies / clients).. regardless.
Way too many SEOs still look for the silver-bullet that'll catapult their websites into the stratosphere. Let me break the news to everyone (once and for all)..there isn't one. Actually, there's four:
Strategy
Marketing nous
Boldness
Technical knowledge
Get good at these four best-friends and, in the future, whenever Google throws a wobbly, you'll respond with giddy laughter ;-)
Amen!
This is the information I was waiting to see, Cyrus! Thanks so much for providing it. Moz makes an excellent guinea pig for this case study because
I'm hoping this article finally puts to rest the (in my opinion) unwarranted hype about the supposed huge and almost automatic CTR boost of all rich snippets.
I agree with others that your last section is golden. It's time for SEO's to grow up and stop playing with toys ;-)
Thanks Mark.
It really shows how nuanced SERP features can be. When I first gained authorship snippets, the verticals I was competing in at the time were relatively untouched. Mine was often the only photo on the page. It seems to work that way with most new SERP features. Everyone wants the shiny new thing. Over time, too many shiny new things and nobody knows what to look at anymore.
My ego took a hit when I lost my authorship photos and didn't lose traffic (or gained traffic) but my brain is better for it.
Thank you Cyrus for this very interesting post.
I am actually not all that surprised that you saw an increase in traffic after the SERP change. With all the author images being removed, people will have to do some sort of screening of the displayed result to see what they believe answer their questions better. I guess having the author photo as part of the result actually gave a lot of traffic to the "wrong" pages.
Agreeing with you and your wrap up at the end: SEO will find a way. SEO will (should)never be just about the rankings. It's about the relevant traffic, the right audience, and thne delivering on everything you promised them out in the SERPs.
Interesting research Cyrus, when I had a chat with you at SMX you were right that more case studies are needed in the SEO world. This is the type of blog post which backs up the content with research and analysis to show exactly what has changed. This change does really affect websites owners who have implemented author mark up for Google and gone out of their way to get the whole team involved and take new images and what not so they look good for CTR. We had a few clients we actually went and did this for. Now it supports those who did not move with Googles changes and try to do things to make their site Markup/ Author friendly.
Thanks James. Also, we have another case study/experiment we're looking forward to publishing in two weeks.
While all of my rich video snippets were lost my rankings still stayed the same. Then the new Google update happened and everything was dropped for a few days. Then the rankings and impressions jumped up to where they were before the update. Not really sure what to make of it
Is it wrong that I thumbed up your comment because your name is Dr. of Funk?
Nice study! Thank you very much for this interesting data!! I also believe that traditional click-through rate optimization techniques and strategies could now be more effective.
Great case study! I was waiting for such case study which I can shows to the client but due to less I wasn't unable to produce. Thanks Cyrus Shepard for your time to live it.
Very good post and the impression jumps Moz is seeing is crazy. I work for a retailer and we have hardly seen any changes within our landscape as our blog was relatively new and as a result authorship had only just started showing and not many of our competition took advantage of video snippets.
On our most organically visited page on Brandpoint.com and one that was receiving an authorship photo, /how-to-use-googles-structure-data-markup-helper-to-add-schema-to-your-wordpress-blog/, we saw this break down after the June 28 cut off: 23% increase in impressions, 20% loss in clicks, even average position, and a decrease in CTR of 35%.
Thanks for the precise data!
There's surely a lesson here about SERP features, assumptions, and the importance of using data to back up your marketing efforts. The video data is intriguing - are the session #s above for WBF inclusive of all visitors to a page that just happens to have video on it, or are they actual video-viewing sessions?
It's a very good question. About 70% of Whiteboard Friday pageviews include a video view.
More to your point - what's the intent of the visit? I thought a lot about this myself when researching this article. If people are searching for music videos or movie trailers, you would expect video click-through rates to be quiet high.
On the other hand, when I'm searching for SEO information, often I'll skip over a video result because I don't want to take the time to watch it.
Intent is everything.
I like the way you chose to wrap up this post, Cyrus. It seems the basics of SEO - strong titles, descriptions, content above the fold that users expect to see when they click through, etc. - have become more important again w/ SERPs being more simplified.
It appears SEO is cyclical. :)
In my area of business, real estate, we are being hammered by the big 'Z' (unmentionable) and those snippets did bring more business...the ctr has definitely slowed a bit but my web traffic is up...
Haha if the big Z you're referring too ends in an 'a', that would be one of our clients:p edit: nvm, you're in the US so probably ends with a 'w'!
I am truly amazed at the results of your small study, especially in this part: "According to Google, for the majorité impressions of our video URLs Dramatically shot up around July 14th."The trend is such that it deserves to be studied further.
Personally glad to see the end of authorship, beaome too common and the number of obviously ripped istock photos was bordering on crazy in some niches. As i'm in local SEO i haven't noticed any impact from the reduced video snippets. YouTube still ranking well, maybe taking a touch longer but still rocking it.
Phenomenal article! These are some really interesting developments. I can't wait to see how this plays out. It seems to me like Google is valuing video more post update- even though they removed video snippets. I will continue to use wistia, and video SEO optimization on pages and will see how that performs. Getting transcripts of the video for content is also a great, and cost-effective way to add a large amount of content to a page as well :)
I know I'm late to the party here - got a bit behind in my reading. To answer your very last question - we have not seen any decrease in either rank nor traffic as a result of the removal of authorship photos. Our ranks have increased slightly, but that's not reflective of the authorship issue because we can't split out the various effects which have led to the rise - i.e. not a valid test.
Glad you have time to play with the data and some good, solid conclusions - we are all for making data-driven decisions!
Thanks Cyrus.
Thanks Martin.
Call me a conspiracy theorist, but one thing I think is interesting is this: If every website only had a 0.5% drop in traffic, this wouldn't be statistically significant enough to notice. But if Google collected that 0.5% in the form of higher add clicks, Google would definitely benefit from millions of extra clicks.
Like skimming fractions of pennies off every paycheck :)
I don't think you'll find yourself alone thinking that conspiracy Cyrus - myself included in that :)
Is there any obvious hack/filter for Analytics - or some process to try and segregate out a drop in traffic purely due to visuals? Given the 200+ ranking factors it's kinda difficult to ascribe the rank increase/decrease purely to the Authorship photo. Just curious - 10-to-1 that's a week's work for you, but it would be an interesting exercise to try and do with our affected clients.
Cheers!
I can hardly see any insights from this post, authorship URLS and non-authorship urls saw both decrease and increase, this cannot mean the removal of authorship pictures has disastrous impact on search. Remeber that searchers are looking for content on Google, not the author pictures! If your site title, description and in-depth content still attracts the searcher, they will click through. We are not Google, we cannot control Google! When there are changes made by Google, we should embrace these changes and study how we could improve with these changes, instead of trying to convince every other SEOer that Google has made it worse and we have lost this or that. The SEO world needs a spirit of optimistic. Dont forget Google bear no responsibilties for SEOers, they serve searchers all the time.
So what was the intended purpose of the whole Google Authorship Photos adventure? Was it just a carrot to encourage G+ adoption?
I still found a author snippet for my google plus share of my Graphic Designing personal portfolio websites Blog post. See the Google Search Result (Number ninth position). Thank you for the informative post.
Interesting, I don't see any author snippets or images when I do the search in incognito mode in Seattle.
Yes in incognito mode i also didn't found any rich snippet. Could you please explain why this is happening? When i have searched in my chrome browser i found the results in number 8 today. I have found another Google rich snippet results today. Here is the screenshot what i have found.
I think as an SEO you have to consider what you think is going to stay and what could be a Google trial. I didn't bother going down the route of author photos which I am now thankful of.
More often than not I now sit back and think about parts of best practice SEO and consider can this be abused, if so, Google is likely to drop it at some point in the future.
The Removal of Author images does;t impact on our e-commerce site. The customer search for Brand , not the image.
Thank you Cyrus.
This is a major change to the SERP features in the recent. The site owners (authors) really enjoyed the nice feature, their face being shown in the SERPs. But over time, almost all search listings accompanied with the nice feature and nobody knows what to look at anymore.
Yep... It was like a nightmare without video and image snippets. Still there are strong credibility for traditional optimization techniques unless those will not be spammed by poor SEO ppl. Google going stronger!
I got so sad when they were removed but hey ho its how Google plays if you don't like it don't play I guess. I'm waiting for the next Authorship style Gold rush!
Found the info really interesting but still wonder where people will go if they can't get authorship as a quick fix.
Thanks for the post!
Great insight. It seems to have only made it more difficult to convince non-SEOs of the reasons to appropriately markup their pages with the recent removal of authorship photos and video snippets. Otherwise, SEO strategies should remain constant, carry on.
I too lamented over the fact that I couldn't get my image to show up properly. Even though the structured data testing tool said it was working I never saw by image in the SERPs next to my content. After a few months I just gave up and assumed it would show up sooner or later. And now it seems like I didn't need to worry so much in the long run anyway!
I also faced the same issues Nick hahaha :D
The question is that Google wants distribute the clicks over the results and over the Ads :-). Incomes are the key.
I think that an important point is that traffic might have changed but in the whole scheme of things, the change is probably only temporary. If the ranking of the listing remains the same, there is an argument that user behaviour will quickly return to type and the traffic that you will get is more of a function of ranking and information relating to your ranking listing. I've always sat on the sidelines with regards to author images, especially if the photo isn't that great, it might not actually help.
I also think that Google should have at least trialed business authorship with an image replaced by a logo. For example, Wikipedia or Which in the UK might well deserve to have their logo next to a listing, just a thought!
It might interest you that at the same time we saw the removal of Authorship snippets in globals search, here in the US we began seeing brand logo snippets of Google+ posts by brands in personalized (logged-in) search for relevant searches. Even though it's only for Google+ page content now, it looks like Google is beginning to experiment with "brand authorship" at least for brands you follow.
Very interesting stuff! The before and after data are clicks, right? Impact on CTR would be interesting. During summer months we often experience lower interest in the digital marketing "niche". So maybe no real impact at all?
It will be very interesting to see if anyone has any data that either reinforces or disagrees with your sites data. Although, the big change that is annoying me at the moment is the local listings. It is just yell.com everywhere!
Pigeon complaint - https://www.integratedigitalmarketing.com/pigeon-local-results-are-useless/
Hi Cyrus,
Thanks for the mention! Well researched article on the effects of pulling snippets and interesting conclusions. I wouldve thought the removal of snippets would cause traffic to plummet. Great study!
Don't under-estimate how complex and important faces are - changes to pictures can change the conversion of a webpage by 100% easy. Perhaps that explains the variance
Mark, that is very interesting, I see it as potentially a positive and a negative but at least it gives consumers choice. I think that search results are over dominated by big brands that all do the same thing, especially in ecommerce. If branding were allowed, it might allow the small guys to stand out a bit more and for Google to understand that not everyone wants to buy the same old tat from the major players and that sometimes, original products and thought are worth a bit more money!
Quite a few of my client's sites have retained video snippets. We definitely go hard with on-page and the Schema video markup. A couple of the sites are ~1 year old. So domain age may not have a lot to do with it.
There is no reason to think one single change in the algorithm caused these changes. G could have made 36 changes on June 28, and we wouldn't know any better. Someone at Google is having a big laugh because they're flipping algorithm switches on and off, and the entire SEO industry goes crazy trying to figure it all out.
We'll never know the "truth" behind the algorithm. We can't even guess in a logical, experimentally sound manner because when we think we understand it, it then changes. We're not dealing with astrophysics or biology here. The "data" you collected 6 months ago is likely useless if the underlying fundamentals (algorithm) have changed.
Can you imagine a biologist collecting data, only to then find that the rules of DNA replication have changed overnight? Or the astrophysicist when he finds that the rules governing planetary orbits have changed overnight?
I think the SEO community has this quasi-religious belief that if they run enough searches and collect enough data, then they'll "figure it out" and sell this knowledge (or some software) to site owners. At the end of the day, it's all about proper technical implementation (i.e., schema), persuasive language (which has nothing to do with SEO and everything to do with copy), testing what you can test (like A/B testing ad copy) and diversification of marketing channels.
That's it.
Google changes the Algorithm over 200 times a year. The "big" changes we often notice, and Dr. Pete includes them in the Google Algorithm Update History
You have a point that there's a lot of bad science in SEO. Cause and effect are hard to map. Correlation doesn't equal causation.
On the other hand, I do disagree that we never know the truth behind the algorithm. Maybe not the whole truth, but we can get glimpses. We've learned so much trough testing about keywords, site architecture, page speed, and dozens if not 100s of other ranking factors that I'm grateful for the SEOs who keep researching and performing experiments.
But I see your point - people profess the "one truth" when it's often much more nuanced than that.
I, for one, am going to keep on trying to peel back the layers of the algorithm.
I always believe SEO is a blend of both Knowledge and Speculation. While nobody, for sure, can figure it all out from any amount of comparative data analysis, but the ones with better understanding of Google Algo are likely get better results from their data analysis.
It's only through understanding various correlations, SEO's can get as close to the causation as possible. I call this speculation because this is pretty much what happens in the Stock Market - the ones who studies the trend and market well enough are the ones that reap better results than others.
Great copy and technical compliance will always matter anyway, but it's very important to conduct your own research and study that of others. It's almost critical to staying ahead of the game. And, that's exactly what makes Moz a definitive resource we always come back to.
This is indeed sad. I don't believe that Google thinks that I will not affect the SERP. A lot of hard work is down the drain, or is it? The authorship feature is still a valid goal to achieve and the profile picture will sometimes appear for users who have you in their G+ circle.
So it is not all that bad. Google AdWords marketeers will certainly gain more clicks to their ads, as the images will draw less people down the page towards the bottom. Just my opinion.
Great post! Fascinating numbers. And yes, it's OK that you gave a thumbs up to Dr. of Funk.
What does it all mean?
It means that Google was losing ADword clicks so whoops no more pics. These amateurs that are running Google need to figure out how to inspire authority in the organic results (because god knows people are always looking for self proclaimed bloviating experts) while not taking away clicks from their monopolistic Cash Machine The License To Print Money Adwords machine
Was this just a test for Google? why yahoo and bing didnt take part in this? My client's website has not dropped CTR but it was looking more truthful.
Yahoo and Bing did not implement Google's rich snippets. This is about a special feature of Google search.
Interesting fact..! In one of my friend's site, he witnessed a decrease in the overall CTR after the Google authorship image was removed.
Rank Watch, and there's the problem that I think Cyrus addresses so well in this article. Everyone's a blind man feeling the part of the elephant right in front of him. One says an elephant is a snake-like creature, the other that an elephant is like a massive tree trunk. Only by comparing across a broad spectrum can we get near the truth.