I usually don’t like to write about Google updates just after they are released. I don’t like it because I try to be objective, to see the dust falling on the ground before analyzing their effects. Something I find ever truer when Google wants to have fun an roll out two major updates as EMD and Panda 20 one day after the other.
If you want to know what I think about the EMD Update, you can read here below my comment to the post my friend Martin MacDonald wrote few days ago.
But here and now I don’t want to talk about the update itself, but about the reactions to it I have seen in the SEOs community so far.
One thing is sure: the SEO community never stops amazing me.
I mean… years and years complaining about the exaggerated power EMD had in the SERPs, and now that Google finally rolls out an update, which purpose is to solve that injustice, the same SEOs complain for the opposite reason? #WTF!
I don’t deny that Google should have tried to refine better this new algorithmic update, having somehow replicated the same “black or white” mistake done with Penguin, but its purpose is surely laudable and the real complaint should be why the hell Google waited so long before rolling it, if it is true, as Bill Slawski with many reasons suspects, it was something that was patented almost 10 years ago.
It’s this unjustified delay what is causing that also respectable and not spammy EMD sites are now dropping like flies.
Few days ago Dr. Pete wrote an excellent “instant post”, describing and commenting what the Mozcast’s metrics are saying in relation to EMDs and the update. If you have not read it yet, I really suggest you to do it.
And read also the comments, a sort of anthology of everything can be said against Google and Matt Cutts. Somehow it was like reading Webmaster forum into SEOmoz.
The most common reaction was something like this:
Hey Google, my EMD was totally fine. It had gazillions pages of content with gazillion words. It was all White Hat and legit and useful and You – tricky b*ast*rd – You screw it!
Surely it is an understandable reaction. I’d react the same way if I were seeing my site falling from the first page into the Index Limbo.
But, let’s try to analyze what that kind of reaction actually tells us.
My EMD was totally fine
Sure it was. And it was an anomaly Google waited for too long to correct.
- To have such an un-proportional SEO benefit just because our site had an EMD was absurd.
- To see EMDs ranking in the top 5 without having a decent link profile (if not any at all) was absurd.
- To see the long tail (but also the middle tail) dominated by EMDs (even dashed) was absurd most of the times.
- And absurd was Google not doing nothing about it.
As absurd was that laziness that the EMDs success caused in so many SEOs. We all knew that EMDs was going to be the next objective, there were signs of their slow decreasing power… and those SEOs did nothing about it? Ok, Matt Cutts is not sexy like Rihanna, but nobody saw this video?
It had gazillions pages of content with gazillion words.
This is a very common justification some SEOs use to affirm their site is a quality one.
Sorry to tell you, but gazillions pages with gazillions words are not synonym of quality… if it was so Allison Copperfield, Lilian Genton, Linda Call and all the novelists specialized in romantic “read and throw” pocket books should be considered better writers than Ernest Hemingway, Kurt Vonnegut or Ray Bradbury.
First of all is not quantity that determines quality. Secondly, it is not length a direct causation of quality either.
Simply they are consequences of two “best practices”:
1) A blog should have regularity in the publication of its post. Not necessarily every day, but neither once in while. Especially the rhythm of their publication must be constant.
If they are SEO compliant, then, those posts will help you ranking for long tail keywords.
But do you think that it is the long tail domination the main purpose of a blog? No, it is branding, voice promotion, social media and link building. A blog is a quality one if it is a recognized source of information and the expression of a thoughtful voice, which could be considered as belonging to a thoughtful leader in its niche;
2) Lengthy posts, maybe enriched by videos and/or images and/or charts, have a better chance to obtain a positive reaction from the readers (check this post by John Doherty on SEOmoz to dig more into this subject). That reaction usually translates itself in the form of social shares and, eventually, links.
But it is not always so… just think to writers like Seth Godin or Chris Dixon, whose posts usually are very short, without images or videos: they are considered useful, shared and linked also when they are one paragraph long. And now, try to reflect about what is the reason of that: thoughtful leadership.
So, no… thousand of unique lengthy content is not necessarily the same as saying quality content.
What that so common declaration also demonstrates is a still common myopic idea of content many SEOs have: that content is written content.
I tend to justify that belief with the fact that for many years SEOs were working with keywords and words as tools. But now, in 2012 almost 2013?
a) Content is every expression of a message regardless of the medium used. Therefore, words are content, but also images, videos, audio and code itself can be considered content.
b) Quality content is that one which unanimously is considered useful by its receivers.
c) Useful is something that answers and fulfills someone needs, which – in turn and using a terminology we all know – may be informational, transactional and – probably the most important one nowadays – emotional.
And how Google can translate this definition of quality into an algorithm? It does it using signals and translating the daily job of its quality raters’ tests into algorithmic formulas (was not this what they did for creating Panda?).
Until last Friday EMD was considered a signal, now it is not if it’s not strongly followed by all the others: quality link profile, readability, social signals, relevance of its writers in case of blogs and, probably, user metrics, like bounce rate and time on page.
This explains why EMDs like Booking.com or Diapers.com are still ranking wonderfully, while others no, for instance Travel.com, which doesn’t appear anywhere for “travel”, also due to a demential use of subdomains.
Certainly, an algorithmic translation of the concept of quality is not perfect, and maybe it will never be no matter how many efforts are done. For that reason there will be always someone, who will not agree with the algorithm results.
It was all White Hat
Compliments, because you did not surrendered to the Dark Side of the Force.
But being White Hat is not synonym of quality or a passport for the 1st page.
Even if you are whiter than Rand Fishkin, if your site has problems like content duplication, extremely diluted link equity, thin content issues and bad information architecture, if the EMD factor disappear, also your 1st page is going to peril (again, investigate the case of Travel.com and you will understand mine assumption).
Ah… I was forgetting another typical complaint:
I am very active in Social Media!
Again a phrase that hides another great misunderstanding many SEOs still have: that being active in Social Media means gaining rankings.
Unfortunately it is not so. Social signals are correlated to higher ranking not their cause, and the correlation is true only if the activity we have in Social Media causes a production of backlinks, being these one still those determining in a great measure how a site ranks or not.
So, if you had an EMD and were active in Social Media, but that activity was poor in terms of links’ generation, then you have probably gained a new and valuable inbound source of traffic, but not obtained better rankings because of it.
But there’s another fact that I believe many SEOs still have not understood, even though it is something that is working in the Google algo since many years: the concept of Entities.
As it usually happens, Bill Slawski have written the most insightful posts about this topic, so I am not going to repeat them, but I give you directly the links:
- Providing related links to documents;
- Boosting brands and other entities. How a Search Engine might assume a query implies a site search;
- Not Brands but Entities. The influence of named entities on Google and Yahoo search results;
- Search Engines and Entities.
The signals that algorithmically Google use to identify entities, then, are quite common to be found in what we generically call “Brands”. But don’t fall in the mistake to identify the two concepts, because the concept of Entity is bigger than the Brand one.
But, being brands a classic example of Entity, Rand Fishkin, there in 2010, designed a very understandable checklist describing what may help the algo defining if a site is a Brand site or not.
If you see the “don’ts”, you can easily identify how them are typically common in EMDs, especially in affiliates’ sites.
So, in my opinion, if an EMD does not own these “entity signals”, then Google could probably consider it less relevant and useful, also because people usually interacts with brands and looks for brands in their searches.
Somehow the fact that some sites, technically to be considered EMDs, were not affected by the EMD update because owning those “Entity signals” (booking.com and diapers.com, cited before) seems confirming my idea.
To conclude, before starting complaining let’s all try to understand if we really deserved to be penalized. We would understand – as it was in the case of many penalizations by Penguin – that from a Google point of view probably we were to be penalized.
Just after that analysis, then, we will be able to work on our site, correct what’s wrong and recover.
And please… change that domain name and start doing RCS!