Algo Change Targets Low Quality Exact Match Domains 9-28-2012

[martinibuster]

Matt tweeted [twitter.com]:

small upcoming Google algo change will reduce low-quality "exact-match" domains in search results.

[BeeDeeDubbleU]

What if you have site about a famouse historical figure or a poet or whatever? Let's say that it is about Edgar Allan Poe. Would edgarallanpoe.com be considered manipulative by Google?

This is a nonsense decision no matter which way you slice it.

[Jez123]

I think we need to plan for the future of google seeing any webmasters who strive for organic positions as manipulative parasitic spammers.

[Rocka]

It appears from this long thread of discussion that EMD's irrespective of quality are being affected. Most strange issues are that they are also being affected for Non-EMD related keywords.

[tigger]

@Rocka

not all - some of mine are fine, but working out a pattern of what triggers the EMD filter is another thing and so far as simple as it is everything is pointing towards the links

[lewis1]

No one seems to have said anything about jimbeetles post about Google's EMD patent.

I had a good read through it and three of our sites seem to fit a pattern mentioned in the patent.

The three sites all rank #1 (competitive niche) for their main keyphrase which is NOT a commercial phase. The domain is a PMD with this keyphrase in it.

However, our internal commercially orientated and SEO'd pages don't rank anywhere.

This seems to fit in with the stuff in the patent although how one recovers from this I have no idea at the moment.

[Rocka]

@tigger

Strange thing I am noting with one of my domains is that its still on the top for the EMD (2words) keyphrase however it seems to be penalized for another variant which was working quite well too.

[tigger]

just out of interest what type of link development have you done for this site? the sites I'm holding onto all have very poor link development and I'm sure this is a factor as to why these sites have held -

one is 10 years old and ranks for all its keywords most in the top 3 - other new site (last year) is holding strong

[Rocka]

its had a steady mix of directories and original articles, no spam links, no bookmarking, no blog comments etc.

Holding good for a the EMD term but out of the top 30 pages on other variants.

[petehall]

just out of interest what type of link development have you done for this site? the sites I'm holding onto all have very poor link development and I'm sure this is a factor as to why these sites have held

@tigger the only one I lost was recent and only has a single link from a very relevant source.

[Dymero]

I agree with this, but who is to say what a brand name is and is not? This is why I believe they are using something which words are generic and non-generic which then determines if you have an EMD or a brand. Who is to say there can't be a brand/company/etc... say... called 'FloridaHouses' and uses 'floridahouses.com'... would this site no longer show for the search term 'florida houses' eventhough this is in fact a brand in this example?

@mihomes:

The algorithm takes a look at other signals to judge quality before determining whether or not to knock it down. Like tedster said, not all EMDs were targeted, only low-quality ones.

[tigger]

@tigger the only one I lost was recent and only has a single link from a very relevant source.

mmmm scratches head in confusement...some would say G has done a good job in completely confusing the SEO market or at least this webmaster - I was under the impress a lot of penalty weight was down to exact anchor matching, maybe something else is thrown in ....

[BeeDeeDubbleU]

Like tedster said, not all EMDs were targeted, only low-quality ones.

This is probably true but from what we are reading in here it was not only low quality sites that were shot down. That is why this is another bad move by Google.

[WebPixie]

@BeeDeeDubbleU: This is probably true but from what we are reading in here it was not only low quality sites that were shot down.

Based on my own sites that experienced huge drops and some others that I follow it appears that the sites I would consider quality sites that lost big time with this update all have link profiles that could trigger the angry Penguin.

I have seen some EMDs with clean link profiles lose rankings, but not on a grand scale. The EMDs getting nuked in this update seem to be experiencing the exact same situation as the sites hit by Penquin previously.

So I think there were at least two major elements of this EMD update. A lessening or removal of any bonus having an EMD gave sites. Secondly, and more catastrophically for the effected webmasters, a lessening or removal of protection for EMDs from Penquin.

Obviously EMDs are particularly vulnerable to over-optimization of link text due to the site name being the targeted keywords.

[petehall]

@tigger me too... only thing it didn't have going for it was links and age. Ranked really well before this update.

I see such a mixture, almost makes me think its just a random car crash at the webmaster's expense to put people off this tactic in future as it was getting out of hand?

[mh_and]

I can only say that I had/have exact match domains that were hit by first, second and every other Panda update/refresh and that was fair. Yes I admit that and don't complain.

These domains never ranked for their main keywords ever after, but this EMD update has more questions then answers.

1. What is EMD to them? bluewidgets com is exact match but what about tomsbluewidgets com? Is that an exact match?

2. What is quality for them? All that guidelines they have or quality is more specific - content, links? Don't forget that Cutts tweeted EMD update that is unrelated to Panda (content) and Penguin (links).

I personally seen a site with a following format domain tomsbluewidgets.com, that stands out among its competitors with its content and links, hit by this update and another site that is bluewidets.com has 1 page content, 3 incoming links and was ranking nowhere for blue widgets now is in top 10.

[WebPixie]

@mh_and I personally seen a site with a following format domain tomsbluewidgets.com, that stands out among its competitors with its content and links, hit by this update and another site that is bluewidets.com has 1 page content, 3 incoming links and was ranking nowhere for blue widgets now is in top 10.

I have also seen examples of both of these situations. A site that was basically BlueWidgetsNow.com was hit and a site, BabyBlueWidgets.com which is a glorified parked page went from nowhere to top of page two.

[SevenCubed]

I can understand google's intention overall. I don't think any of us would strongly disagree that some of the EMD that got hit will not be missed. But as londrum points out above where he says "but if i just gave you the two domain names (signposts) on their own" he's right in the sense that google cannot differentiate between the two. Therefore they should not have rolled out something like this until they had a much better handle on it. I can see 1 site in the batch that I'm closely watching that should have been sent to -950 land but it's still on page 1. Clearly many of the desired targets have been missed and some that should not have been hit were. They absolutely could not have done enough quality control testing prior to unleashing this destruction.

[muejl1]

I think Webpixie is right on with what they said.......this makes the EMD's more susceptible to Penguin and having their links look spammy. I got whacked as well and the first thing I'm doing is going through my links.....making sure to remove any bad ones. The second will be on-page elements....maybe I need to remove the keyword in my domain from my title tag.....etc. I addition, I think EMD's need to be even more concerned with anchor text diversity.....

I have noticed one thing though.....for terms that I've not done any link building on...their position in the SERPs have remained steady......it's the ones I did link building on that have caused issues. I did receive an unnatural link notice back in late March, but didn't really suffer. Maybe this update just made me more likely to feel the consequences of rookie SEO mistakes.

[mihomes]

The algorithm takes a look at other signals to judge quality before determining whether or not to knock it down. Like tedster said, not all EMDs were targeted, only low-quality ones.

I think everyone in this topic is aware what Matt said by now and that other factors determine a ranking... why most people are here is because they are questioning it... why are people seeing anything BUT low quality EMD's being hit...not all of course, but some.

I feel like I keep reiterating myself in this topic, but I believe there is something more at play... perhaps a system which judges domain words to determine brand vs non-brand or non-generic vs generic. If majority of ones backlinks are anchor text with a 'brand' (Macys and Macys.com) is that seen as un-natural? What if those backlinks are generic (widgets1widgets2 and widgets1widgets2.com)? Mind you for both cases it is natural for links to be your name which is also the domain name.

The topic should focus on opinions and thoughts about what is actually going on. If it were as simple as low-quality EMD's were hit then there would not be much discussion - I think most of us can safely determine a low vs good fairly well these days and have a good grasp on it.

At this point in time we have reports of good-quality EMD's hit as well as low-quality EMD's still ranking or even going higher. Let's discuss...

[Leosghost]

And we had the same reports after Panda 1 of supposedly good sites being hit, and what it couldn't possible be about..and then a long time later the "Google reviewers guidelines" got leaked ..and some people realised how much time they had wasted in denial ..and many of the original deniers then denied having said what they had said previously, about what it could not possibly be when Panda 1 rolled out....

I really don't see how going down that path , yet again, is going to help anyone who has been hit..because any discussion, of what else it might be, and "how it can't be" is, like the last time , is also going to be full of folks who also know damn well why they were , hit , but are going to publicly say otherwise..

And they are going to be jumping into every thread , no matter the subject, and posting about "it can't be" and it's not fair on my site, revenues etc etc".."it's the end, somebody help" just like the last time ..it has already started..

While some are sympathising with the genuine cases of collateral damage ..they are also sympathising in some cases, and threads, with the very scrapers, and /or owners of really low quality EMDs, who have brought this upon everyone..

Many here forget who has said what down the months and years..and as a consequence give their sympathy, and their credence to those who abuse..

[jimbeetle]

I think most of us can safely determine a low vs good fairly well these days and have a good grasp on it.

No, we really can't, and I think this thread shows just that. We have no idea how Google is measuring quality. Take some of the factors in the thread I referenced some number of posts above that Google or any other SE *might* use to measure quality:

A Mathematical Model for Assessing Page Quality

Behavioral score of page – How people perceive a page is a big indicator of the quality of the page. This can be measured by analyzing the user behavior. Some of the factors that are traditionally used are –

1. Conversion score
2. Bounce rate
3. Number of page views
4. Number of repeat visitors to this page
5. How many people add products to cart after visiting this page?
6. Average amount of time that is spend on this page.

Now, that' just how a user interacts with the page. What about throwing in how the site now interacts with the user, it's social profile? And, even more intriguing to me, is the possibility of the "are you who you really say you are?" entity analysis. Google has put in a huge effort on this over the past few years.

The Next Generation SEO thread [webmasterworld.com] touches on a few more of the things we have to think about going forward.

[WebPixie]

@Jim

Do you really think people are speaking to the quality of their site without taking into account, bounce rate, conversion, page views, time on site/page, repeat visitors and social?

[Ralph_Slate]

Google obviously can't compare those metrics for dissimilar sites, so they are likely comparing what they believe to be similar sites or even similar pages. But who is to say that they are correct in their comparison?

Let's say that I have a site describing the history of widgets, and you have a site that displays the current market value of widgets. Similar, but not quite the same because with the latter you would consult it to look up a value whereas you would consult the former if you wanted to learn more about the widget.

Google might classify the two sites as equivalent and make one disappear from the search results because they say that it has better metrics than the other when in reality it is being used in a different way.

I can't believe that Google would be dumb enough to use site-wide metrics for comparison because some sites have various sub-sections - message forums, blogs, data, reviews, ecommerce, etc., and that would be a really bad comparison. It really only could be a page-to-page comparison, but again, how do they know the comparison is between equal competitors?

[jimbeetle]

Do you really think people are speaking to the quality of their site without taking into account, bounce rate, conversion, page views, time on site/page, repeat visitors and social?

Yeah, just look at the comments in this thread. Most people made no mention of those factors, all that was talked about was content and backlinks.

[Ralph_Slate]

jimbeetle, how can you compare those factors across disparate sites? Here are my numbers;

Conversion Score: N/A. I run an informational site.
Bounce Rate: 49.34%
Number of page views: 5.83
Number of repeat visitors: 61%
Number of products added to cart: N/A
Avg time spent on page: 3:28

My site has apparently been classified as "poor quality" - despite that unique visitors over the past month is 423,801, and 30% of the organic visitors arrive at my site by requesting my site's name in their search along with other terms" (such as "blue widget widgetsrus" where widgetsrus is my site). A fair number of my pages still rank well, but certain sections of my site will not rank at all, no matter what I do (except if I get backlinks).

Are my numbers good or bad? I can't tell. If you post your numbers, can you say if your site is better or worse than mine? I doubt it - bounce rate could signify how well the site satisfies the immediate need, or how the site is able to draw the visitor in deeper.

[jimbeetle]

jimbeetle, how can you compare those factors across disparate sites?

I dunno', I'm not the math guy, just reporting it. Did you read the thread [webmasterworld.com] I referenced some posts back or click through to the original post [stonetemple.com]?

[MrSavage]

After a bit more peeking around various keyword search results that I'm familiar with, I am wondering if trademark/brand usage has something to do with this update. Was your website tied into a particular brand/trademark and has been part of this update in a bad way?

I'm not going to be surprised to hear Google come out with something about trademark usage and copyright usage. Something along the lines of the rightful owners getting the traffic as it's rightfully theirs. If you're trying to ride on the coat tails of a brand, did you experience a tumble with this update?

[RishiRich]

Hey Guys! What do you feel about the latest EMD Update ?
I personally feel that Google did not devalue EMD's but imposed a severe Penalty! Also i feel that many EMD's with quality content and link profile got hit as well! Do you feel the same way about genuine EMD's being hit.

Google should have devalued the importance of EMD names but imposing a penalty was really uncalled for

[courier]

I have a very expensive EMD I purchased a few years ago in the intention of creating an offline brand with an online presence. As it has come about, my creation of an offline brand has been slower than I would have liked and with the present financial climate, and how Google has performed with my other sites it has been a set back.

It had 10 pages of crappy content which was nothing more than to hold it, normally sits around page 5 to 7, and was unaffected by the EMD update.

I had decided two weeks ago to start with an offline presence, (opening a shop) so I decided to update the crappy content that held this site on friday. I put all of my 10 years experience into writting the best content which is better than every other source on the internet at present. I added two new pages of content and re-wrote my home page, guess what... withiin 24 hours the site dropped to #99.

What does Google want? They certainly don't want good content. Where do I go with this next, the EMD describes perfectly what service I offer, I only offer this service unlike evryone else who offers this service as an extra.

[RishiRich]

Just a small weather update costed companies millions of dollars!

Domain name buying and selling is a major business and so is securing an expired domain name when it goes back to the domain pool and thanks to Google all is screwed now!

What about companies sitting on a reserve of 4000 + EMD names trying to sell them one by one? It is not about search quality. It is about letting the webmaster know that with Google there is no fixed set of rules that you can follow to get high rankings. The only thing stable is their advertisement platform and you should seriously start thinking about advertising now!

I do not understand what is wrong in getting an EMD. EMD's are better than branding because when user sees an EMD he or she already knows what the website is about before actually seeing the meta description or title and this is the reason many big brands are going after micro websites rather than having a single big domain. A single big domain is "Ugly" and "Confusing"

Why in the world would Google Penalize someone of having an EMD despite the fact he has quality content and links.