searchmetrics email facebook github gplus instagram linkedin phone rss twitter whatsapp youtube arrow-right chevron-up chevron-down chevron-left chevron-right clock close menu search
30193019

Update: Google “Bad” SEO Update – Now named Penguin Update

Google, by way of Matt Cutts, announced some weeks ago that a special SEO over optimization update will come. Now it’s live with its first iteration. Yesterday evening Google also published the article “Another step to reward high-quality sites” with some more insights into the update. Interesting stuff that every SEO should read.  The official goal was “to take care” over-optimized websites, containing too many unnatural links, automated content (spinning), keyword stuffing etc. Google tries to kill webspam altogether. The impact on all keywords queries is about 3.1%, which compared to Panda (with around 12%) is much less. But Google said more short-head/visible keywords should be affected.

What’s definitely new and surprising for me was the day of the update. It’s unusual for Google to make a major update on a Tuesday or Wednesday.  Normally Google makes this kind of update on a Monday or Thursday. That’s why I presume that in the next few days we’ll see more updates and this update is just the beginning.

Update: I updated the data tonight so it is fresh and you’ll see more losers and much better correlations to the Google update. You can also click directly in the table to see more data in the Searchmetrics Essentials.

What happened?

I took nearly a huge set of keywords from short-head to medium and low search volume and looked at the current rankings from position 1 to 100 and compared the rankings to April 20th. In the data were also some glitches from the Panda 3.5 update which was going live from April 19th to 20th, Matt Cutts mentioned. But overall you see a trend of those domains which really lost visibility within the Google Penguin update.

In the data I saw some striking losers, which suggests to me there must have been another update today. There is a pattern in the winners and losers table and I’ll give you some insights below the table. Note I analyzed the visibility, not the overall traffic. The SEO visibility only includes calculation on organic rankings. Direct traffic, Social, PPC etc. is absolutely not included.

List of SEO visibility losers (google.com)

UPDATE: I updated the table (2012-04-26) with more keywords and fresher data.

Domain SEO Visibility Difference in %
songlyrics.com 128186 -128679 -50,10
great-quotes.com 4515 -72248 -94,12
comcast.com 143032 -66286 -31,67
dslreports.com 65264 -63770 -49,42
appbrain.com 100782 -53439 -34,65
hotels-rates.com 16868 -43554 -72,08
1-love-quotes.com 30366 -42772 -58,48
tvrage.com 124821 -40846 -24,66
shoplocal.com 39654 -39966 -50,20
vzw.com 31323 -38387 -55,07
free-games.net 40883 -37542 -47,87
hotelplanner.com 8933 -33676 -79,03
consumeraffairs.com 50384 -32243 -39,02
lottostrategies.co 1192 -31367 -96,34
edogo.com 2015 -29850 -93,68
2011blackfridayads.com 9514 -29339 -75,51
creditcards.com 47759 -29281 -38,01
download-free-games.com 33558 -28644 -46,05
htc.com 49104 -28268 -36,54
black-friday.net 49484 -28153 -36,26
freephonetracer.com 29783 -28085 -48,53
ghacks.net 38952 -27989 -41,81
customerservicenumbers.com 48318 -27985 -36,68
langenberg.com 22037 -27534 -55,54
karaoke-lyrics.net 1595 -27269 -94,47
dressupgirl.net 42109 -26965 -39,04
robtex.com 92320 -26162 -22,08
lotteryusa.com 76685 -23355 -23,35
bestsampleresume.com 17787 -22175 -55,49
resellerratings.com 48542 -21948 -31,14
dmv.com 15619 -21672 -58,12
songs-lyrics.net 4284 -21096 -83,12
unemployment-extension.org 214 -21025 -98,99
dictionary.com 27654 -20584 -42,67
thedomainfo.com 74737 -20497 -21,52
creditcardforum.com 13481 -20119 -59,88
thesmokinggun.com 41682 -19956 -32,38
ipaddresslocation.org 6050 -19357 -76,19
free-online-games-to-play.net 1489 -19288 -92,83
digg.com 68914 -19126 -21,72
mainkeys.com 44423 -19082 -30,05
lovepoemsandquotes.com 25694 -18719 -42,15
tjoob.com 5585 -18522 -76,83
promgirl.net 15993 -18204 -53,23
americanrhetoric.com 6652 -18079 -73,10
costumekingdom.com 8813 -17670 -66,72
doc-txt.com 13794 -17229 -55,54
education.com 54829 -17127 -23,80
freegames.net 827 -17023 -95,37
halloweencostumes.com 22190 -16951 -43,31
statscrop.com 14936 -16759 -52,88
latest-hairstyles.com 40126 -16733 -29,43
abovethelaw.com 19251 -16450 -46,08
celebjihad.com 23960 -16152 -40,27
aoltv.com 62290 -15872 -20,31
cardratings.com 11762 -15859 -57,42
watch-movies-tv.info 23640 -15717 -39,93
collegeview.com 15151 -15526 -50,61
interest.com 5606 -15389 -73,30
freecookinggames.net 3207 -15246 -82,62
scrabble-word-finder.com 2564 -14442 -84,92
americanheart.org 3732 -14035 -78,99
lotsofjokes.com 12157 -13954 -53,44
ticketcity.com 48204 -13954 -22,45

A lot of these losers are database-driven websites – they mainly aggregate information and use large database systems to create as many pages as possible. Sites such as songlyrics.com, great-quotes.com, cubestat.com or lotsofjokes.com fall into this pattern. It makes sense that these sites will lose visibility.

Press portals and feed aggregators such as pressabout.us, newsalloy.com and bloglines.com were also affected, which makes sense from a Google point of view since these are the website types that are very often created by very aggressive (possibly overly aggressive) SEOs and often contain similar content.

A couple of heavily template-based websites were also affected – ticketnetwork.com/ticketcity.com, hotelscombined.com and customerservicenumbers.com fit Google’s anti-SEO bill perfectly when it comes automatically (possibly also spun) content.

Furthermore, a lot of sites that copy or rehash other peoples’ content (or are used by their users to do that) were demoted – examples include mayor sites such as digg.com, folkd.com and pastebin.com.

First Preview of SEO visibility winners

Domain Win of SEO Visibility in %
spotify.com > 30%
yellowbook.com > 30%
observer.com > 30%
nme.com > 30%
menshealth.com > 30%
poynter.org > 30%
mysanantonio.com > 30%
slideshare.net > 30%
azstarnet.com > 20%
lasvegassun.com > 20%
stackoverflow.com > 20%
epinions.com > 20%
ebates.com > 20%
locatetv.com > 20%
ovguide.com > 20%
gq.com > 20%
newyorker.com > 20%
giantbomb.com > 20%
networkworld.com > 20%
morningstar.com > 20%
livescience.com > 20%
snopes.com > 20%
marvel.com > 20%
lyrics007.com > 20%
esnips.com > 20%
techspot.com > 20%
usnews.com > 20%
drupal.org > 20%
theverge.com > 20%
miamiherald.com > 20%
webs.com > 20%
thehollywoodgossip.com > 20%

 

On the winners side it’s not as easy as on the losers side to find a pattern. But a lot of brand sites won visibility, which makes sense to Google strategy. Sites like menshealth.com, gq.com or lots of other magazines fill perfectly the gap that the losers leave.

The day started exciting and it will end exciting and tomorrow will start again exciting because I have so much data to analyze. 🙂 I keep you updated.

Marcus Tober

Marcus Tober

My name is Marcus Tober and I’m the founder of Searchmetrics. Because we really love to analyze all kinds of online data, we can give you more insights than any other company in SEO, SEM and Social Media. It’s not a job, it’s passion.

115 thoughts on “Update: Google “Bad” SEO Update – Now named Penguin Update


  • David Spanierman 2012/04/26 at 4:45 pm

    Hi Marcus, great article as always. Can you explain how you conjure the organic rankings percentage increase or decrease? I am not 100% sure what the percentage represents.

  • Marcus, have you or any one else seen a negative impact on websites with 301 redirects following this update? One of my web properties that had a 301 redirect to it lost about 30% of its traffic on Wednesday. The homepage is still ranking very high, but the internal pages (which were housing content from the old domain) have disappeared? Prior to Tuesday this website was ranking very well for all of long tail keywords.

    Also, I have to agree somewhat with Jimmy, most of my other keyword rich domains dropped, but a few moved up. Surprisingly the keyword rich domains that went up were the ones with little or no SEO. Go Figure!

    Any input regarding the 301s would be very much appreciated

  • I suggest you all look at this: http://searchengineland.com/winners-losers-from-googles-webspam-update-119493

    There was a Panda update on the 19th April which has significantly skewed this study.

  • Is it due to a 3.5 version of the filter Panda launched on or about 19 April (and confirmed by Matt Cutts site Search Engine Land)? Or the new algorithm spam “black hat” Google announced yesterday (though it was his understanding that according to the announcement of Matt Cutts, the new algorithm was not yet launched …)? The fact is that Google’s results have been greatly modified in recent days.

  • @ Remote Keyless Jimmy

    Well I’ve listed plenty examples over here and on Search Engine Land of keyword rich sub and root domains getting a boost. Many of which violate nearly every major guideline Google has tried to instill in us.

    Whether the boost was directly related to the name or not, is still up for debate. But the examples of it happening are numerous.

    -And just a friendly tip, using keywords as your name in the comments could easily be interpreted by Google a SPAM.

  • Make money online 2012/04/27 at 12:00 am

    Are google for real – Are they really so desperate for more adwords customers? If so they should just google “Make money online” and look at the first result to see what they’ve done. C’mon guys. This penalty against quality sites who may not have even done their own seo is desperate and already backfiring.

  • Make money online 2012/04/27 at 12:24 am

    Oh, I forgot to add a ps.
    P.S. If you really want to Make Money Online (www.notgoogle.com) I’d suggest you consider buying shares in Bing and Yahoo right now.

  • @ Marcus et all,

    I don’t know about the other sites in your list, but I do know similarsites.com very well, especially the online marketing, and the site has not lost organic traffic from google over the past few days.

    Either your numbers are wrong, or your not measuring the right parameters, but search traffic has not been hit recently.

    Knowing what I do about one specific site in your list makes me weary of accepting the rest of your data as well as your conclusions.

    A few additional points:

    Similarsites is not a spammy aggregator/spinner, the site takes information, processes it quite intelligently, and then creates fresh content which adds value for the millions of people that use it each month.

    The real organic earthquake hit a few months ago, and it seems to be more about localized results for organic searches, i.e. – person in NY searching for “widgets” (NOT widgets in NY), and SERPS serving a blend heavily salted with widget sites in NY (or aimed at NY). This is not just for terms like “pizza”, but also true for more general informative terms.

    All this is effecting the bigger national directory-type sites. I’d be interested to hear about how the sites in your list rank across regions, i.e. – search results per keyword per major US metro area in google.

    You can’t really check SERPS for a broad USA google anymore, it’s not the death of SEO, but there is no longer a “real” first page listing for many terms, it all depends on where the searcher’s IP is logged from.

  • First of all congratulations to Mr. Cutts and all the members of the webspam team, they surely have rolled out an algo update that will not only kill webspam but also force closure of so many mom and pops stores that depend on SE traffic to bring home the bacon.

    Google clearly wants everyone to use PPC and pay them to get traffic. This is a shady business practice. You may be trying to kill SEO but did you guys realize you are giving rise to a much bigger market of NEGATIVE SEO?

    If not, then check out this little experiment http://trafficplanet.com/topic/2369-case-study-negative-seo-results/#entry33318

    If that does ring a bell of alarm in your ears then nothing would and gaming dumb search engines (read Google in this case) and blasting your competitors out of the SERPS became a much easier task.

    You sure hate webspam but you are the ones who are actually promoting it.
    —===End of Rant===—–

  • @Alison
    Its pretty true as people can play dirty on competitors now, without leaving a trace.
    The update, I hope is incomplete as I saw genuine sites disappearing nowhere!, which kinda sucks.

  • @David Spanierman
    The percentage means the increase or decrease in SEO visibility based on our keyword set. I updated the list and now you also see the difference in SEO visibility and the percentage.

    @LC
    I also saw this drop especially for keyword domains. But these domains often don’t appear in loser lists because they rank for too less keywords and thats why the drop in SEO visibility is not high enough to show these domains. But you can check all your domains in our Essentials: http://suite.searchmetrics.com/en/essentials

    @Chris Gedge
    It doesn’t matter if there was a Panda update in between or not. We update our data every week since years. So the trend is most important and if you lose X percent of your SEO visibility you have to take a look at your site. This is what matters. The only thing is that the interpretation of the analysis is more complex because of some noise of another update. But that’s why I updated the data this morning. 😉

    @Boaz
    similarsites.com is also ranking well in many other Google indexes, e.g. co.uk and .pl. The drop in visibility is mainly in the US index. That’s why the overall traffic could be stable. Look here: http://suite.searchmetrics.com/en/essentials?url=similarsites.com&cc=US. If you register for the Essentials you can also see the keywords where they lost rankings.

  • @marcus, I don’t need to register with your program, I have access to the similarsites.com analytics, and I repeat that organic traffic, including organic US traffic, did not drop this week.

    Besides that, the site has dedicated subdomains per language/region. So, co.uk traffic, for example, mostly hits the UK version of the site (uk.similarsites.com).

  • @Boaz
    To be honest. Similarsites.com is a site which typically has to be hit by this update. The function to show similar sites is something that has value for the users, but it’s no real content with real information. The loss in SEO visibility means that you don’t lose this percentage in traffic, it means you lose this visibility. And if you’re below the fold for lots of keywords where you has no value for the user, then you rank, you get visibility, but if you lose lots of these rankings you lose a lot visibility.

    This site used to rank for lots for domain and brand names. No wonder that this site isn’t ranking for keywords like: wuzam, craigslist evansville, datehookup, backpage portland, tubekitty anymore. I assume that this site will lose more SEO visibility in the future. Nothing against the product, it seems great, but no useful content if user are looking for brands etc.

  • @marcus, you’re making a circular argument, saying we were hit by the update because we had to be! The rest is kind of smoky.

    If we really lost visibility, then de facto, we lose organic traffic, the two points are directly correlated to each other (i.e., more results and higher rankings equal more clicks). We don’t track most individual keywords, but since we have not lost traffic, if anything, we had a small increase, then I don’t see the point you are trying to make.

    The site ranks for a huge variety of terms, with a considerable churn rate among them, which is what is probably throwing off your visibility program.

    So, having never used your data before, I can still say that it seems to have some flaws when dealing with large and complex sites. Besides that, I don’t know how valuable your visibility data is unless it is tied to a specific region (for example, google.com for a NY IP) – to quote Dylan, “the times, they are a changin'”.

    Can your product take location into account? If not, it may not have much use for current SEO practice across a huge range of keywords and site types.

  • @Boaz
    You are a target of these updates because you have no valuable content for a search engine. I also guess that your “considerable” churn rate of changing keywords is much higher than for a normal website. Your product looks fine, but you’re a spam for a search engine. You have nothing unique in terms of content. You scrape the title and description and build a relation between other sites where you also show the title and the description. If I were you I won’t build a business around organic traffic… because this is just automated content and spam for SEs. Sorry, but that’s the truth. This what you’re doing I did 8 years ago. Look we have millions of domains and lots of data for these domains in our database. The reason I don’t create well optimized landing pages is that I know it will only rank temporarily, because it’s spam.

    And your argument with the dependancy on the location makes no sense if I take a look at your keywords. There is mostly nothing related to possible local searches. You build sites liks these http://www.similarsites.com/site/youtube.com and try to rank on brands/domain names and some long tail content around the content that belongs to this domains.

  • @Marcus, no, we were not affected by this update, why is that so hard for me to explain? Traffic has not dropped, not US, or any other oragnic. The new spam update is also supposed to address spam in other languages than English, and we did not suffer there either.

    The site itself is not spam. Just because something is DB driven does not make it spam. We create unique correlations between data which users find relevant and useful. Our user base includes millions of toolbar users and repeat visitors that find our free product useful on a daily basis.

    What we really offer users is a combination of user data for a huge range of sites and relational data between millions of websites. We add new data and features continuously. Our competitors are folks like Alexa and Xmarks. We are not some whois-driven spam, our widgets and APIs help power many other sites, including Compete.com.

    My location argument is more general and not specific to similarsites. The main point is that for many keywords, there is no longer a “real” ranking on advanced engines like google, its now a ranking set that is qualified per term/location.

  • Thanks a lot for this news very good for my business.

  • @Boaz
    This conversion leads to nothing. Look SEO visibility is not traffic. It’s more. It’s a calculation which includes ranking of a keyword, search volume, type like navigational, informational and transactional search. It does NOT take a usual CTR, also keywords below the fold and keywords on page 2 will be considered on a diminishing scale. But you just talk the whole time about traffic. You don’t want to think about what visibility is.

    You say you offer a service similar to compete.com, but this is totally different than talking about the SEO of your site! Compete.com does not spam the internet like you do. You created these hubs http://www.similarsites.com/browse on purpose to index all these pages. These pages only offer scraped content, nothing else.

    Look at your SERP spread. It’s absolutely not natural and you’ll see a drop in page 1 and 2 rankings this week: http://img404.imageshack.us/img404/2287/similarsitescom.png
    You also see that for all the keywords that we track for the US index you only have 2% keywords on page 1. This is far far away from a natural site.

    Your algorithm behind your product seems to be cool, but what you do is spamming. This has nothing to do with your user base and the product at all, I just want to say that your spamming is against what you wants. Be honest to yourself. Spamming the index with content that Google already knows is no rocket science and not useful for Google users. I hope we can discuss this out in person sometime. Cheers Marcus

  • @Marcus, wow, what started off as a factual error on your part (our getting hit by an update) has turned into a bit of a tirade.

    Quick responses:
    -Re visibility: this started as an algo post, saying a new update is killing spam sites, and that our site was hit hard (was at the top of your list). I replied that our site was not affected by it. If an algo kills hits a site, rankings drop and so does traffic, an algo hit directly implies a traffic drop.

    -Re compete: we don’t offer a service similar to compete, their site uses our API for some of the data they show their users.

    The /browse page is not a hub, but part of the internal site index (sitemap), yes, we want our pages indexed, but that is the what a sitemap does – it helps users and spiders find relevant pages on a site based on whatever hierarchy.

    -Re serp spread: nothing could be more natural for a type of site like ours. Their is no one-size-fits-all “natural” profile, it will vary across markets, site types and regions.

    Re spam: the site does not scrape & serve. We have both a ‘cool algo’ (thanks) and a large data set of user behavior, so the results we show are very useful to a large pool of users.

    We are not “spamming the index with content google knows”, since we take various data sources, then manipulate it to add value to our users (browsers/surfers, widgets and APIs to a wide range of other sites).

    What happens is the opposite – we get hit by scrapers and spammers daily, spammers are scraping our content and trying to feed it to search engines left and right. Since our unique content gets stolen/scraped so often, it’s a bit annoying to be labeled a spammer by someone who I stumbled onto while reading a defamatory post about a site I’m involved with.

    Why don’t we just argue about politics and religion? LOL. Anyways, I’m trying to think what I can do with your blog post’s data, and how I can use it to improve similarsites. If you have a free full-working version I’ll give it a whirl.

  • I think the sites that dropped are doing ad related cloaking.
    Not showing the ad when Google bot visits, but showing when everyone else visits.
    One of them shows a Google ad below another ad, so when the ad is not shown, a google ad appears.

    They all seem to be doing the same thing.

  • Most affected are content aggregators with large databases…

  • Chicago web development 2012/04/28 at 11:11 am

    Thanks for inform me because i have lose site ranking … thanks again.

  • Web Design Coimbatore 2012/04/28 at 2:00 pm

    Thanks for sharing a latest update of Google..I really glad to see…

  • I have lose 10 positions on google. thanks for the information

  • Hi,

    we were hit very hard by the penguin update, we have lost 72% visibility in germany. After two days of frustration I’ve started a blog for pengiun victims.

    Best wishes,
    Georg.

  • Доставка цветов 2012/04/29 at 12:22 am

    I think it`s true, wating…

  • Hard time for bloggers, many of my fellow bloggers were hit by Panda, and now Penguin making things worse. Some of them quite blogging and some have started updating their blogs, but its gonna be a real hard work.


Write a Comment

Note: If you enter something other than a name here (such as a keyword), or if your entry seems to have been made for commercial or advertising purposes, we reserve the right to delete or edit your comment. So please only post genuine comments here!

Also, please note that, with the submission of your comment, you allow your data to be stored by blog.searchmetrics.com/us/. To enable comments to be reviewed and to prevent abuse, this website stores the name, email address, comment text, and the IP address and timestamp of your comment. The comments can be deleted at any time. Detailed information can be found in our privacy statement.