Linknetwork visualization

SISTRIX LabsDuring the course of our ongoing development of the Toolbox, we also look at ways to show new features as best as possible – the best data becomes useless if you are unable to easily analyze it. The backlink-module especially has a large increase in the amount of data that is available and visualizing the connections among themselves has become rather problematic in form of charts. Because of this I want to introduce an alternative in the SISTRIX Labs: being inspired by the Java-Applet Touchgraph, which shows site-connections based on the Google-informations for ”similar sites“, we are testing how it works when it is based on links. When you input a domain, we show 10 domains that link to the queried domain. By clicking on the green plus that shows up for each domain when you mouse over it, you can expand the displayed links and show more of the linknetwork. Additionally, you can also see the amount of outgoing links for a domain when you move your mouse over the icon. The lower part of the tool lets you switch to full-screen mode and scale the size to your liking. This implementation is based on JavaScript – this has the advantage that most of the current browsers will be able to run it without extra plugins, sadly the Internet Explorer is the only browser that does not like to play nice. I will make this tool available to everyone until the end of the week and would be thankful for some feedback in return.

Johannes Beus - 21.11.2010 12:04


How Google uses legal loopholes

Today finally saw the launch of the German version of Google Street View, after a month-long tug-of-war. A lot has been written about the advantages and disadvantages of this service and there has been even more discussions on the legality of this service. What I find interesting in Streen View is that it is a nice example of how Google uses legal loopholes. When our lawmakers put the “Freedom of Panorama” into the copyright law, I am sure not one of them ever thought that one day a monopolist from the USA would mount digital cameras onto cars and plan to take pictures of the whole world – they were concerned that you would not run into any problems when you took a picture of anything that could also be seen freely from the public streets. A similar approach can seen with Google's Booksearch: Google just started digitizing a huge mass of books and then later, will bother with making sure that there are laws governing the process. YouTube is another example of a service that would probably not have been such a huge success, if people would have only been allowed to upload their personal videos – today, the Film- and Music-industry has to somehow make arrangements with this power-player.

All these examples have in common that now, we have applications that, a few years ago, no one would have thought possible to even exist. The computing power of Google's and Facebook's server-farms is what makes many of these services even possible. When our minister for consumer protection Aigner deletes her Facebook-profile, then this might get her one or two notices in newspapers, which is certainly good publicity but fails to recognize the fact of the matter: Internet giants are driving lawmakers in front of them and they will only react years, if not decades, in the future. As a nerd that was computer-socialized in a time when Microsoft was the embodiment of evil, I view this anxiously. Back then, Microsoft might have had a monopoly – but they did not possess the amount of personal data that Google and Facebook hold today.

Johannes Beus - 18.11.2010 19:41


Google’s Interface Iterations

In the last few weeks, Google has once again made numerous changes to how they display their searchresults. After launching Google Instant, they added screenshots for each result as well as a noticeable change to how they display the local search integration. While we can only speculate on the why, I have the feeling that they are not making much headway in improving the quality of the content anymore and are therefore moving on to the presentation thereof.

Screenshots
Google ScreenshotsLooking past Google's impertinence of just using my favicon to integrate the screenshots into the SERPs, I think another aspect of this issue is much more interesting: it seems that most of these screenshots are already completely rendered on Google's servers. In reverse this means that Google runs sites through a HTML-parser ahead of time to have them in stock. Everyone who has done something like this before knows that purely crawling and extracting text from a website and parsing complete HTML-websites are worlds apart when it comes to the computing power that is needed. I find it hard to imagine that Google only takes this step to make their SERPs prettier, I rather think that the screenshots are by-products of their website-evaluations. This means that Google now knows quite well where text and links are positioned on the page, regardless of the their positioning in the HTML-sourcecode and is able to use this information as a ranking-factor. SEO-advice that deals with positioning important links high up in the source-code and then show them at the end of the page with the help of CSS will not die out, but they should keep on losing their effect.

Google Maps Places
Google MapsThe changes for many local searchresults are also quite interesting. Until now, everything was rather simple: there were 10 organic results from the Google-Index and then there were the results from the local Universal-Search. The later were shown in their own Box (the map as well as the results), when Google thought it appropriate. Now this has gotten somewhat more complicated – Google partially rescinded the strict separation between the organic results and the vertical search in displaying the results. This becomes more apparent through this example: here I have a screenshot from the end of September for the search for “delicacies cologne”, which still shows the old results. At the top we have the Unversal-Search integration and then there are the 10 organic results. Something that is also apparent is that a domain like “hoss-delikatessen.de” is shown in both lists. This screenshot shows the results for the same keyword for today, therefore showing us the new way results are displayed: the map moved to the right and will move over the AdWords when you scroll down, additionally, the Universal-Search results have gotten a picture, making them more prominently visible in the SERPs. The real difference though is how the two lists of results are combined. If we look at position #4, we are still shown the domain “hoss-delikatessen.de” - the result is now also upgraded with additional information from the vertical search. This change has rather decreased the amount of results on the first page but the remaining results are more prominently integrated – this should increase the competition for keywords which trigger a local search integration.

Johannes Beus - 14.11.2010 13:29


IndexWatch 10/2010

The first couple of days of the month might have already gone past, but seeing how Martin has so charmingly asked for the new IndexWatch numbers in his blog, here they are. As usual, these are based on the SISTRIX VisibilityIndex and show the winners and losers in the Googleindex for the prior month. Let's start with the winners:



Aside from those sites that managed to show strong gains this month due to Google lifting a penalty there are also some other interesting finds. Webnews.de for example, did a relaunch last month and managed to noticeably raise the visibility for their domain within the Google-SERPs. When we look at their formerly unchallenged competitor Yigg.de for comparison, we are shown that the tides can still turn after a long time and that the first player in a market does not necessarily also have to be the winnner in Google.

We have three travel businesses on the winning side, they are tjaereborg.de, attika.de and jahnreisen.de. It looks like those sites are running on the same system, where someone seems to have gotten hold of a SEO-Basics-PDF. I'm sure their customers are happy.


As far as the losing side goes, I find janolaw.de to be quite interesting: after the site saw a noticeable decline in the beginning of the year, they have been all but absent from the index for about two weeks now. At first glance, this is neither a problem with the domain, the old link-structure or the site itself – though the site operators are (as is customary in their trade) quite generous when it comes to excerpts from their contracts and samples. Having both www1 and www2.janolaw.de still in the index seems to rather be a side note.

Johannes Beus - 10.11.2010 21:43


SEO-regular's-table Bonn

Now that Summer-break is over and we got through the huge number of events in September, it is once again time for a SEO-regular's-table in Bonn. The 28th of October beamed at me from my calendar, so we will start as usual at around 7 pm CET. Everyone who is interested is asked to please sign-up through this form. Further information will be send by e-mail.

Johannes Beus - 04.10.2010 20:43


IndexWatch 09/2010

An eventful September just came to a close: we moved to a new office, started the international versions of our Toolbox and the SEOktoberfest was the event-highlight of the year. To make sure that work doesn't fall short, here are the winners and losers in last months Googleindex. Let's start out with the winners:


Website.org, tripod.com and anglefire.com all are part of a rather large spam-operation, which Google should get a handle on soon (it seems that for tripod.com this is already the case) and which should cause the domains to be on the losing end for next month's list. To me it is interesting that something like this even has a chance to find their way into the index – it seems that advancements since 2005 were fewer than expected.

Antag.de is – once again – penalty-free. This is interesting in that this system is not only being used in Germany. LeGuide, the french operators of these sites, are trying to push these sites into the index all over Europe – with differing degrees of success. I will write up a more detailed report on this at a later time.

Yelp has just started in Germany and is now trying to grab some visitors from their big role model Qype at yelp.de. For now, it's looking good but I don't really believe in them making it big in the SERPs, seeing how many of the strategies that might work in the USA, cannot be copied over to Europe one-on-one. I am, of course, always open to being surprised.


Celeb.ag moved to Celeb.to and managed to cut their visibility in Google about neatly in half, date-seiten.de seems to have received a penalty and joinr.de is once again gone from Google's traffic hot-spot.

The Federal Ministry of Transport, Building and Urban Development managed to lose numerous relevant keywords with their relaunch: autobahn (highway), führerschein (drivers license), heizkostenverordnung (heating-cost act), mautgebühr (tolls) – all keywords for which the Ministry showed up before; now you won't even find the domain in the top-100. Let's see if they can still talk about the “wide-reaching Internet-platform” after their next relaunch …

With guenstiger.de we have an old and rather established price-comparison service who shed a lot of feathers. From the former nearly 120 points in the VisibilityIndex, they fell to about 50, which should also show up rather noticeably in the number of visitors to the site. Interesting in this case is the fact that the domain pushed a large amount of new sites into the index during this timeframe.

Johannes Beus - 04.10.2010 11:28


Hello, Bonjour, Hola, Ciao

Since the introduction of the SISTRIX Toolbox, I have not written anything about new features and the state of development, so I would like to take the chance to give an update. From today on, the Toolbox-data will not be limited to Germany anymore. They are now also available for Great Britain, France, Spainund Italy. With this, we now cover the five most important European marketplaces and can therefore provide a unique summary of online-marketing-developments in Europe.

From now on, the VisibilityIndex is not only available for Germany, but also for the four new countries. The index-value is created in such a way that they are comparable between countries: a value of 10 will therefore correspond to the same visibility within the search-results for all countries. You can get a good idea of how this works with the values for Wikipedia, for example:


Sichtbarkeitsindex Wikipedia.org

In Great Britain, the domain is represented even stronger than in the German index, Spain is at about the same level while it is not quite so prominently featured in both France and Italy. When you take a loot at the amount of articles in each of the country-specific versions of Wikipedia.org, you will notice them to have about the same ratio as this ranking. These values are also interesting when you base your comparison of similar sites on them, which offer local versions of the site. Here we have the VisibilityIndex of four well known price-comparison sites in one diagram:

Preisvergleich in Europa

Twenga shows a strong north-to-south gradient: while they're quite strong in France, Italy and Spain, they are not able to gain a foothold in either Germany nor Great Britain – as far as Great Britain is concerned though, it seems that price-comparison sites in general have a hard time in that specific index. The most successful price-comparison site in all of Europe at the moment is Ciao. Let's see how long it will stay this way after the purchase by Microsoft and their anti-trust complaint against Google. Besides the SEO-data we also have the SEM- and Universal-Search-Module available for use. The following evaluation shows the amount of SERPs with at least one Universal-Search-integration:


Universal-Search

This shows that Google is putting even more importance into the Universal-Search-integration in Great Britain than they already do in Germany. Spain is at about the same level as Germany while both Italy and France get noticeably less Universal-Search results at the moment. In closing, I want to show a useful application of such data. This diagram shows the number of keywords with a picture-search integration on one axis and the amount of unique domains that can be found on the other axis. While Germany and Great Britain have a lot of unique domains that compete for the ranking, you see a lot less competition in Spain – where there are still more integrations overall. This means that if you specialize in picture-optimization, Spain might be a feasible market for you to look at. Details on how to use these new Toolbox-data-sets as well as some background-information can be found in the Changelog
Johannes Beus - 13.09.2010 07:15


Irgendwie and SEO

I was in Munich for the past two days: yesterday, Bing showed us upcoming features and changes in the Allianz-Arena and today, Stefan celebrated the 10th anniversary of suchmaschinentricks.de with about 300 guests (and thanks to vigorous support from Bing). Having been a forum-user of the first hour, I was happy to be do a presentation together with Uwe Tippmann from Abakus on the subject of backlink-evaluation. After the presentation, there was the request from the audience to publish the numbers and the ideas behind them and so I would like to do just that in this post.

My presentation took on the subject of link-evaluation-factors and what SEO's do with them. The word “Trust” that we use often originated from the paper “Combating Web Spam with TrustRank”. In it, Google explains how to recognize spam, based on a manually selected seed-list of trustworthy domains. This can be roughly explained in that they take the click-interval between URLs on the seed-list and those URLs that they are checking for. Many assume that there are many .edu-top-level-domains on that list – seeing how these are some of the oldest and most trustworthy domains on the Internet.

Additionally, there's the public information. A surprising number of SEO's generalize this fact to the end that they believe all .edu-domains will generally have a high trust in Google and so they will put a lot of energy and money into acquiring these obscure links – the main thing is that they come from a .edu-domain. In our database, I found 4.824 .edu-domains. I then took the IP-popularity, which is the amount of different IP-addresses that have links pointing to that domain, as a rough indicator and was able to show that only 472 of those domains have an IP-popularity of more than 1.000 – which is less than 10%. Here a slide to make this even clearer:



For my second example, I looked at the composition of the backlink-structure. In the patent “Methods and systems for identifying manipulated articles“ (US Patent 7,302,645), Google depicts how stark deviations of certain signals from the norm can be used to pinpoint unnatural backlink-profiles. This should be general-knowledge for SEO's and there is probably no SEO that would disagree with these remarks. If we take a closer look at some backlink-structure though, it becomes apparent that objective and reality are not always exactly on the same page. This example shows a domain that has more than 1 million backlinks – but all of those are from only 11 different IP-addresses. Here we might not have as many links, but every link comes from exactly one domain. Natural-growth looks different, which should have hopefully been nicely demonstrated by the base-numbers forspiegel.de.

Uwe showed an analysis of a backlink-set of more than 10 million links and presented us with some interesting averaged values, which also continued the second part of my presentation in more detail. I was under the impression that most of the attendees were pleased with the event and I am looking forward to the 20th anniversary.

Johannes Beus - 10.09.2010 19:45


Google Instant Search

We had to live with those annoying, interactive Doodles for two days but now Google let the Marissa out of the bag: the new feature, which is dubbed “Instant Search” will already show you search-results, while you are still typing the keyword into the search-box. From a engineering point-of-view, this has been very well implemented and will be quite impressive when first used. Everyone who wishes to try this new feature should log into their Google account and then click here. This new feature is there to help you get to the results you want even more quickly. According to Google, it should shorten the time you search for a result by about 2 to 5 seconds. The new feature is completely G-rated, once the results move over to the adult-category, the SERPs won't be shown automatically anymore.

I was able to find something on the question of how impressions are going to be counted now over at Greg's: Google will count an ad-impression if either a) a user clicks on a result, b) the user agrees on the search-recommendation (click on keyword, hit enter or click on the search-button) or c) the results are shown for at least three seconds. I could imagine that the third factor especially will increase the amount of impressions within the next few days.

When we get to the issue of what kind of an impact this feature will have on the search-behavior, we see a few different points-of-view. You have those who believe that the long-tail will be strengthened by these recommendations (effectively an extension of Google Suggest) and then, there are those who think the exact opposite is true. Personally, I am not quite sure yet, but I do believe that something else is much more important: Google is now able to analyze User's search-behavior even better (at which letter in the keyword do the users find the results they are searching for) and – which I find more exiting – Google can also directly control the search-behavior of users. We all know that the short-tail is more profitable for Google, which means that more search-traffic in this area will probably well liked for a publicly traded company...

Johannes Beus - 08.09.2010 22:14


IndexWatch 08/2010

Today, we present to you the newest IndexWatch-Numbers, at the start of this SEO-Event filled September. These are, as usual, based on the VisibilityIndex and show the winners and losers in the Googleindex for the previous month. Let me introduce last months winners:


Three rather new domains were able to harness large gains last month: unternehmen-db.de, gutschein-codes.de and hochzeitsplaza.de. Looking at the first one, I do not think that these gains will be for long: for one, the domain already showed strong fluctuations during the past weeks and additionally, Google would rather do business in that sector themselves. The coupon-site is unlikely to have it easy either, seeing how that sector shows some rather fierce competition, which are quick to rat out some not quite so natural links.

The development of billiger.de is quite interesting – over the last couple of years, the domain kept a rather low key and moved around between a value of 3 and 10. This has changed in the last couple of weeks, which saw their visibility skyrocket to more than thrice the value they had before. Sadly, I did not yet have the time to take a closer look at this phenomenon but it seems they found the right dial to fiddle with.

Another perfect example that duplicate content is still no problem and that Google will notice it perfectly show the domain pairs scinexx.de/g-o.de and deichmann.de/deichmann.com.


Nauticexpo.de has managed to nearly get completely rid of their Google-visibility. While at the beginning of the month, the site had keywords like “schlauchboot” (rubber dinghy) or “segelboot” (sailboat) in the top-10, those rankings are nearly all gone now. Without having more information on this, I would guess that a failed relaunch would be the reason for this. Does anyone know more about that? Connect.de has also lost important rankings: for handy (cell phone) they went from position 7 down to 12 and for numerous cell-phone descriptions they tumbled from the first results page down to ranks below the 50s. They should also feel this in their incoming traffic.

Winhelpline.info seems to have fallen into a penalty. It is interesting that this had been the case once before at the beginning of 2008, but since then the domain had been a stable representative in Google. As far as Sun.com is concerned, many of the content has moved to their new owners (oracle.com) which accounts for the loss in visibility.

Johannes Beus - 03.09.2010 10:32


1 2 3 4 5 6 7 8 9 10 11 ... 101