Google Optimization Archives

When Google announced support for the canonical tag just about a year ago, webmasters were excited for the possibilities of a serving a 301 redirect to spiders but not users. But when should you not use it?

A Google Webmaster Help thread has Google's JohnMu explaining some situations as to when you should or should not use it.

In summary, think of the canonical tag as a real 301 redirect for spiders. If a spider is redirected away from new content, then that is an issue. When it comes to paginating content, such as product category pages or article archives, you really need to give a way for search engines to find that content. If you set the canonical tag to redirect spiders from page 2, 3, 4, etc of your product category pages to page one, then the spiders might never be able to index the products on page 2, 3, and 4.

John explained this well, saying:

Pagination: this is complicated, I personally would be careful when using with rel=canonical with paginated lists. The important part is that we should be able to find all products listed, so at the very least those lists should provide a default sort order where we can access (and index) all pages. Since this is somewhat difficult unless you really, really know what you are doing, I would personally avoid adding rel=canonical for these pages. One possible solution could be to use JavaScript for paginated lists with different sort orders, for example, that way you would have a single URL which lists all products.

This is something to keep in mind when building out the canonical tag within your content management system.

Forum discussion at Google Webmaster Help.

A week or so ago, Google posted a webmaster quiz for people to answer. Today, they posted the answers to the quiz. The beauty of this is that you can really learn a lot about Google's SEO best practices by just reading the questions and answers.

You can read all the questions and answers or see how well you did on the test, if you took it.

Only three people got all 40 questions correct, they include ChrisRaimondi, theopeek and beussery from the forums.

Forum discussion at Google Webmaster Help.

A WebmasterWorld thread has an interesting discussion started by senior member, Wheel. In short, he is expanding one of his sites from 21 pages to about 5,000 pages. He wants to take the approach of letting Google index the site and then use Google and the site command to figure out which pages he should then go back into and internally link to.

Let me explain what I think he is trying to accomplish here.

(1) He launches the new web site and waits for Google to index most of it.
(2) He uses the site command with keywords, i.e. [site:domain.com keyword+here]
(3) The pages that come up, he will then look into and try to boost by having other internal pages link to them with the keyword phrases in or around the anchor text

He explained why he is going about this, in a way that seems somewhat backwards to most SEOs. He said:

Because I've got so much content it's hard for me to sort through it to give myself internal links. My intention is to let the content get indexed first. Then I'm going to search my site using Google for the juicy terms I want to rank for, and order them by whatever Google feels is order of relevance.

Then I'll take a few of those top ranking pages for the term, find the term on those pages, and link to another page on my site that I want to rank for that term. Maybe even link to the homepage in some instances.

It is an interesting method, something I am eager to see how it works. To be honest, most SEOs deal with this already when taking on new clients that have existing sites. But it is still an interesting discussion.

Forum discussion at WebmasterWorld.

A week ago, we reported that the iGoogle Webmaster Tools gadgets were broken. Google confirmed the issue then and said they would be working on fixing them.

Yesterday, Jonathan Simon of Google said it was fixed. He said:

Good news...the Webmaster Tools' iGoogle Gadgets should be functioning as expected again. Thank you for hanging in there while we worked on getting this fixed.

Some confirmed the fix is indeed working. But when I checked it this morning, it doesn't seem to be fixed for me. Hopefully it is just me and it did not break again.

Forum discussion at WebmasterWorld.

In October, we reported that Google treats a 410 stronger than a 404 status code. I am not sure exactly when Google made the 410 status code that important, but I believe it was sometime in October.

A new thread surfaced in Google Webmaster Help where it was suggested by a Googler to use the 410, once again.

Googler, Jonathan Simon said:

Last year we started started treating 410 Gone responses as being a stronger signal for indicating when a page no longer exists. So for the situation you describe (where 301 redirects aren't practical) returning a 410 response is the way to go.

It seems to me that when Googlebot discovers a 410, it will communicate to Google to not index that page for a really long time. I believe Google will check pages they know to be 404ed every now and then. Maybe a 410 will tell Googlebot to never check back or check back extremely rarely?

Just keep in mind that the 410 is much stronger than a 404 response - it may come in handy in your future SEO projects.

Forum discussion at Google Webmaster Help.

A Google Webmaster Help thread has discussion about a potential duplicate content issues between HTML and PDF documents. In this case, the content found on the HTML is the same as on the PDFs. Be it an automated "print as PDF" feature or manual download of the content in PDF format.

How does Google handle the duplicate nature of such content available on the web?

JohnMu at Google chimed in saying that in most cases, they will use the HTML file. He does recommend that in these cases, you block the PDFs from being crawled and indexed. But ultimately, he said, that is your call. Google will likely just want to keep the HTML version in their index.

John said:

If you have the same content in PDF as in HTML pages, in most cases we'll probably show the HTML versions above (or in place of) the PDF versions. If this is a problem for your specific situation, I'd consider using the robots.txt or x-robots-tag to prevent the PDF files from getting indexed. I imagine for most sites this is not really a problem, so I wouldn't suggest blocking indexing of PDF files without confirming that it's really necessary.

The only situation where I would consider doing something in advance is when the CMS automatically creates PDF-copies of normal HTML pages. Generally speaking, this shouldn't cause any problems, but those PDF versions are likely not compelling enough to merit getting indexed separately (and crawling them will possibly put a load on your server that you could avoid). Ultimately, it's up to you to determine which content you wish to have crawled and indexed :-) -- if you feel that PDF-copies of your content are compelling enough for users who search for your content, feel free to make them available.

Forum discussion at Google Webmaster Help.

A Google Webmaster Help thread has discussion around Japanese domains and how Google may treat them. Honestly, I know very little about this topic, so I hope I do justice to this thread.

Example of Punycode domains versus Unicode:

• Unicode - http://アアアアア.com
• Punycode - http://xn--cckg5dfdwc.com

Google said, for "domain names in Punycode are equivalent to their Unicode versions." There may be an issue with PageRank being displayed in the Punycode version in the Google Toolbar, but Google assures us that there is PageRank, the equivalent to the Unicode version. JohnMu said in the thread that he would "pass that on to check out" why it isn't working properly in the toolbar.

So which should you use? Google said, they generally use the "Punycode versions since this is more likely to work across all browsers."

Forum discussion at Google Webmaster Help.

I spotted an interesting Google Webmaster Help thread where a webmaster was asking why the most used words on his site is not showing up in the keywords report in Google Webmaster Tools. It seems like, at least for this report, Google may consider certain usage of a keyword to be over usage and would consider those keywords either boiler plate text or common words.

Jonathan Simon of Google said in that thread, and let me quote:

The Keywords list will sometimes exclude words that we determine to be boiler plate text or common words. This varies from site to site. The Keywords list is a starting point to see how Google is interpreting your site's content. This list should be evaluated in tandem with what's listed in the Top Search Queries report for your site as well as how your site appears in the actual search results for the keywords you're targeting.

He goes on to explain that although it may not show up in the report, due to possible Google issues, this should not impact the site in a negative way. He said:

Boiler plate and commons words (excluded from the Keywords list) vary from site to site. It sounds like for your site we could be doing a better job here.

The more important question though is if this negatively impacts your site's ranking as you mention. It doesn't. That's why I said the Keywords list is a good starting point and that it should be considered in conjunction with other information.

The question I have, is this not helping your site? I guess not.

Forum discussion at Google Webmaster Help.

An ongoing WebmasterWorld thread has sparked up some heat over the weekend when more people are claiming to see the Google Caffeine results out in the wild.

WebmasterWorld administrator, Tedster, said on Friday night, that he seems them on six IP addresses. He said:

I'm seeing the Caffeine data-set being served via this set of IP addresses: 64.233.169.147, 64.233.169.105, 64.233.169.103, 64.233.169.104, 64.233.169.99,64.233.169.106

It seems to take 5 IP addresses to build the complete SERP, where in the past it often took only 3.

Here are links to them:

• 64.233.169.147
• 64.233.169.105
• 64.233.169.103
• 64.233.169.104
• 64.233.169.99
• 64.233.169.106

Senior member, Whitenight said:

Well, just tripled checked with offices/employees in Texas, Colorado, and Indiana. All 5 "control" keywords/sites showed live Caffeine.

Does this mean it is live? Not necessarily. Does it mean that Caffeine is coming soon? Not necessarily.

I guess we wait on Google to say something.

Forum discussion at WebmasterWorld.

For those of you who use iGoogle and use the Webmaster Tools Gadget, you may have noticed that no information is flowing to the gadget. Here is a screen shot from my iGoogle dashboard, which I honestly never go to:

It seemed like the issue became a big issue starting on the 20th. A Google Webmaster Help thread has several webmaster reporting it. Then Jonathan Simon of Google came in and confirmed the bug. He said:

We are currently looking into this issue and hope to have it resolved soon.

There is not estimate time for a fix, but Google is aware of the issue.

Forum discussion at Google Webmaster Help.

Google announced two new features in their web results. First is called "Answer Highlighting" and the second is an enhancement to the Rich Snippets feature.

Answer Highlighting in short basically uses Google Squared technology to highlight in the search snippet the answer to your search query. For example, and I don't see this in the live Google results right now, if you search for [empire state height], in the past (current) you would have seen this:

With the enhancement, you would see the answer:

Getting the answer in the search results, is a goal of Google, but that means, or can mean, less click throughs from Google to your web site. Webmasters often live off traffic and conversions, so this can be a bit upsetting to many webmasters.

Brett Tabke said in a WebmasterWorld thread:

Taking another page out of the WolframAlpha play book, Google introduces Serps with answers. You may never need to visit any site again.

The other enhancement is allowing event data in rich snippets. How does that improve the feature? Just look at this snippet:

Forum discussion at WebmasterWorld.

I am not sure when this happened, but fairly recently, Google has changed the number of Sitemaps you can reference in a Sitemap index file. The number use to be 1,000 sitemaps can be referenced in a Sitemap index file, now the number is 50,000 Sitemaps. This is a huge increase in capacity.

Still, each Sitemap file can contain up to 50,000 URLs, so technically 50,000 multiplied by 50,000 is 2,500,000,000 or 2.5 billion URLs can be submitted to Google via Sitemaps. That is if I can add correctly.

Googler, Jonathan Simon, said this in a Google Webmaster Help thread:

Thanks for resurfacing this thread as we've improved our capacity a bit since then. The limit used to be 1,000.

The Help Center article you point to is correct. The current maximum number of Sitemaps that can be referenced in a Sitemap Index file is 50,000.

Forum discussion at Google Webmaster Help.

A new HighRankings Forum thread asks if there is any downside in terms of SEO for using JavaScript that disables the ability to right-click on the page. The thread asks:

One of my company's sites has right-click functions disabled (yes, I realize this doesn't really stop people from stealing content - it wasn't my choice). I've noticed when I use a spider emulator (seo-browser.com) that our image alt tags appear to be invisible to the spiders. I can see the alt tags on the actual site, and I've verified that they are in the code, but they don't seem to show up for spiders. Could this be caused by our right-click disabling?

Most people in the thread say that it should have no impact on spiders crawling the site.

I then saw an older thread from Google Webmaster Help where Googler, JohnMu, said the same thing. He said and I bolded the key point:

Personally, I find the use of right-click-blocking JavaScript slightly annoying because there are many legitimate reasons why you might want to use the context menu (eg to bookmark the page) and it doesn't really stop people from viewing the source (Ctrl-U brings it up if you don't want to use the main menu). That said, this is not something that would bother Googlebot :-).

Forum discussion at HighRankings Forum & Google Webmaster Help.

In December, Google announced they will be personalizing all Google results by default. SEOs rightfully were a bit on edge after that announcement, but we did link to an old study on how personalized results impact rankings, overall.

The study has been revised and posted at the Huomah Blog. I won't go through all the points, but the conclusion is that currently, the personalized changes are so minor that it is not worth SEOs freaking out about them. The overall conclusion of this study stated:

We shouldn’t be changing how we approach things ultimately… Above the fold is the real estate that’s prime (what else is new?). It may be the measuring that we will have to adapt. You will need to find ways to check rankings from a few locales and discern a mean average instead of a definitive placement.

It is worth while reading the whole report at the Huomah Blog.

Forum discussion at Sphinn.

Last week, I wrote about The Google Screw and how Google's "intention engine," as Tedster calls it at WebmasterWorld, can seriously hurt a business and the integrity of a solid search query. But Google is not stupid, they know what they are doing and they feel, for the most part, these types of changes will improve the overall search experience.

Tedster at WebmasterWorld has been one of the more vocal against Google telling you what you really wanted to search for. But he recently found a good use for it. Let me quote Tedster:

I helped a friend last week by cleaning up a smallish database (about 1,000 products) so that it could be used to generate web pages. The database was from his shipping facility and it was full of strange abbreviations so the field would print out on a small shipping document and a shelf label in the warehouse. So the product descriptions worked for the warehouse staff but they were far from doing the sales job they needed to do.

Google's intention engine did the job. I just pasted in that mishmash directly and I usually got back what I was looking for, or at least I made enough progress so that I could make a second try and nail it.

Maybe Google's idea isn't so far off the mark after all.

Pretty neat - he plugged in the weird descriptions from the client into Google and Google told him what he really wanted to use for titles and descriptions. Not a bad use for this, don't you think?

Forum discussion at WebmasterWorld.

A WebmasterWorld thread has discussion over how Google handles the mass changing of title tags.

In this particular case, a webmaster copied the structure, not the design, of a site to a new site. When they copied it over, they forgot to change the title tags, although the content, design and images are all unique from the previous site. Soon after, they changed the title tags to make them all unique to the content on those pages. Four months later, the title tags Google shows in the search results are the same as they were - i.e. the old, wrong title tags.

Since this is a WebmasterWorld thread, we do not know the specific site having trouble. So as you can imagine, it is incredibly hard to diagnose the issue.

jdMorgan, WebmasterWorld administrator, said:

The one thing that Google "hates" is large, sudden changes. So I'm going to recommend that you grit your teeth and wait it out instead of giving G yet another reason to think your pages are 'suspicious.' Build links, expand and update content -- anything productive to keep your mind off this embarrassing mistake, and suggest that an official "acceptance testing" program (with a formal list of things to check off, including <title> and <meta name="description"> tags) be put into place for new developments so that this won't happen again.

Keep an eye on Googlebot, and once the pages in question have been spidered a couple of times, this situation should resolve itself. It's very hard to be patient and "do nothing," but that's what it takes in this situation; Working on the other aspects of improving the site will help -- both practically and emotionally...

Of course, if he wanted specific advice about his web site, he should go to Google Webmaster Help. However, many people avoid that inorder to keep their anonymity.

Forum discussion at WebmasterWorld.

Imagine you have been in business, operating on the web for years now. Your name is unique and it ranks number one in the Google results, since it is unique. But then imagine a new site pops up that is similar to your name. Then soon after, Google decided that your name is really not what people want, instead, they want the other name.

In other words, they type in your business name and Google say - no no, you spelled it wrong, you want a competitor or another web site. Google then hides your web site and shows the competitor. How would you feel?

That is the story of a Google Webmaster Help thread. The owner of BrandFreek no longer comes up for a search on [brandfreek], instead, Google shows a result for BrandFreak and shows BrandFreak.com in the first spot. Here is a picture:

This can be pretty upsetting to the owner of the company and also hurt them financially. Google knows this. JohnMu at Google replied to the thread saying:

Thanks for posting! I'll pass your feedback on to the team to review.

I am not sure if Google will change the behavior of this search in the future, but I think they should.

Forum discussion at Google Webmaster Help.

Here is a recap of what happened in the search forums today, through the eyes of the Search Engine Roundtable and other search forums on the web.

Continue reading "Daily Search Forum Recap: January 14, 2010"

About 6 months ago, the Google Caffeine public preview went live. In November, they took the Caffeine preview offline and since then, webmasters have been on their toes worried about when Caffeine will go live. We had rumor after rumor with speculation that it was live, as recently as a week ago - but each time, they were false alerts.

The question people are now asking in the long WebmasterWorldthread is where is it already? What is the hold up? A good question. For all I know, it may go live a minute after I post this and make me just look bad. But let me quote you one WebmasterWorld member:

Do you remember waaaay back in late Nov./early Dec when MC told the world that Caffeine was being delayed to "save stress on webmasters before the holiday" and it would be rolled out soon afterwards?

Do you remember what I said was the reason back then?

So here's the question for you, IF as MC said Caffeine was "ready" a month and half ago, where is it now?!

We're halfway done with January...what's the hold up?

A solid question and a question that can be asked. Google does not have to answer but it is a good question. Are there major issues with Caffeine? Are there underlying flaws that will prevent what we know Caffeine to be to be launched? Can the failure of such a launch make Google look somewhat inferior to Microsoft Bing?

All questions that Google's executives and PR team would hate to have to answer. But as time goes on, these questions will begin to be asked.

Forum discussion at WebmasterWorld.

Two days ago, I reported that people were seeing favicons in Google Webmaster Tools. I personally did not see the favicons until last night.

Here is a picture from my Google Webmaster Tools account:

I like it and so do many others. It isn't a big change, but does help you find the right site in a batch of dozens or more.

Forum discussion at DigitalPoint Forums & WebmasterWorld.

A month ago, Google added real-time search results in the form of Tweets from Twitter and other real-time search related sources. Yesterday at Search Engine Land I covered an interview with Amit Singhal of Google via Technology Review. The interview explains how Google ranks those real-time search results - on some level.

Want to rank high in those Google real-time results? It seems like all you need is a lot of followers and you should be set.

From the interview:

"One user following another in social media is analogous to one page linking to another on the Web. Both are a form of recommendation," Singhal says. "As high-quality pages link to another page on the Web, the quality of the linked-to page goes up. Likewise, in social media, as established users follow another user, the quality of the followed user goes up as well."

Obviously, Google needs to figure out the value of the followers of followers, but that shouldn't be too hard in this equation.

Another interesting point was in regards to the use of hashtags in Tweets. The interview wrote that hashtags may "serve as red flags to lower tweet quality and attract spam-like content."

Of course, this does not mean anything you Tweet will show up in Google's search results. For that, they need to be trending topics, and a good way to see what is trending is to look at Google Trends.

Honestly, there are some interesting thoughts in the comments on my post at Search Engine Land.

Forum discussion at WebmasterWorldand Google Web Search Help.

Did you know that if you are in Google News, you need to keep your pages below 256 KB to get your articles indexed by the Google News Bot?

A Google News Help thread has Googler, Inbal, explaining to one publisher why his articles were no longer being included in the Google News index. She said:

Thank you for bringing these errors to our attention. It appears that your site's pages are too large. The hub or article page length exceeds the maximum allowed. Pages can be 256KB long. I've updated your site's information in our system. We should begin to crawl more articles from your site within a few weeks.

As you can see, it does appear that Google can make an exception to the rule. She did say that she updated the "site's information in our system" and Google News will begin to crawl the site again soon.

Forum discussion at Google News Help.

A Google Webmaster Help thread and a WebmasterWorld thread has questions about the timeline for receiving a response from your reconsideration requests.

About six months ago, Google began sending notifications of status updates on reconsideration requests. But there are two times where Google will not send a notification. As Susan Moskwa said in the Google Webmaster Help thread:

(1) If your site is not (never was) in violation they won't email you
(2) If your site is still in violation they won't email you

But recently, over the vacation, it seems like there has been a backlog at Google slowing down the review process. A WebmasterWorldhas two webmasters complaining about the slow down in these requests. Personally, I find it funny they have that much experience with this process that they know when Google is slowing down. Maybe all they do is fix sites killed by Google?

In any event, if you submit a reconsideration request to Google and you did not fix the issues with your site or if you never had issues with your site - you won't get a response.

Forum discussion at Google Webmaster Help and WebmasterWorld.

Update: Susan commented below, she said:

This is not accurate. We send notifications in all cases after we've processed a reconsideration request. What I said was that if you don't see any significant change in your site's performance after your request has been processed, that usually indicates one of the two items above.

A WebmasterWorld thread has someone asking if what Amazon is doing, is considered cloaking. First let me show you what they are doing.

Go to this page on Amazon.com. Notice, at the top right of the results is a "Sort by" drop down. Here is a picture:

Now, go to the Google Cached version of that page:

Yes, it is not there. It seems like Amazon served Google the page without the sorting feature, which technically is a form of cloaking. But is this done with intent to artificially boost their rankings?

The content seems exactly identical, minus just the sort by feature. The sort by feature can produce duplicate content and make it harder for Google to index the incredibly large site. Is this being done in a way that is trying to hide something?

If they really want to be white on white - they could now keep the sort by there and use the canonical tag for all the sort by options. But this might be something Amazon has doen for years prior to the canonical tag coming out?

Personally, I wouldn't consider this against Google's cloaking guidelines, but I am not Google.

Forum discussion at WebmasterWorld.

A DigitalPoint Forums thread has an image of Google adding Favorite Icons (favicons) in the site list. I personally do not see them, but here is a picture from the thread:

I am pretty sure this is new, because the favicons are not showing up in my Webmaster Tools console.

What does this do for you? Not much - but it does help you differentiate the sites in the list a bit faster.

Forum discussion at DigitalPoint Forums.

This is not a Caffeine update, Caffeine is not live. More coming soon.

There seems to be a lot of buzz going on at the forums about Google's Caffeine index going live on more and more Google data centers.

Let me give you some history on Caffeine. It went live as a public preview in August. Then in early November there were rumors of it going live before the holiday season. Matt Cutts said no way, it won't go live until after the holidays. But it did go live on a single data center, in the Google wild, in late November.

Now, it is after the holidays, after the new years and everyone is waiting for Google to flip the switch. Did Google flip the switch in the past twenty-four hours?

I am seeing tons of threads with people talking about major ranking changes. But there are only a few threads specifically thinking it is Caffeine related. We have threads at DigitalPoint Forums and an updated WebmasterWorldthread. The suspected data centers that Caffeine has been moved to, include:

• 209.85.225.103
• 74.125.95.132
• 66.102.7.18
• 64.233.163.19
• 74.125.19.106
• 66.102.7.104

I emailed Google, Matt Cutts and his team for confirmation - so stay tuned.

Forum discussion at DigitalPoint Forums and WebmasterWorld.

Update: A Google spokesperson told me, "we have nothing to announce today regarding Caffeine." Basically, it is not live yet and when it does go live, they will let us know.

A week or so ago, we reported how Google removed SEOs from local pack in the search results. The move caused a pretty big debate in the industry where many SEOs felt this was a nail in the coffin for Google really hating the SEO industry. Personally, I think it is a fun debate, but I don't think all of Google hates the industry.

In any event, Joel H. from the Google Maps team updated the Google Maps Help thread with a revised statement, implying that Google may change their minds on this decision. He said:

I think I understand both sides of the issue: I understand that local search results can have a positive impact for some businesses. But I also see that some SEOs/web designers want their website, not a local listing, to be the primary destination. And I want to clarify that today we think this is a better user experience - to display primarily web search results for some of these queries. Using words in or near seem to be suggestive of a different intent. In the end, these search results will fluctuate as we fine tune our results pages to get the user the best answer.

Again, he said, "I want to clarify that today we think this is a better user experience." He also then went on to explain why the words in or near don't always provide a clear indicator of local intent.

In any event, if I had to guess - Google will wait to see how both searchers approach the missing SEO or web design related local packs. They will also see if SEOs and web designers calm down over this change. I assume if SEOs don't calm down about the change, Google might consider adding it back. But clearly, Google finds the categories either not relevant or too hard to handle from a spam perspective.

Forum discussion at Google Maps Help.

In an updated Google Maps Help thread, Joel H. from Google said Google has dropped most of the local results in the Google web search pages for SEO and Web Design related queries. Yes, you heard me right - Google has decided to intentionally not show a local pack for queries that are related to SEO or web design and include a local qualifier. Example searches include [web design 10010], [web design suffern], [seo 90210], [new york seo companies] and so on.

Google first removed web designers from the local pack in November. Then they claimed it was a bug related to a fix for an Orlando Dodge dealership, but that is not true, according to Joel.

Yesterday, Joel from Google said:

Earlier thread, Brian (BB- baseonballs) [a Googler, btw] noted that this may be related to a known issue regarding the display of local search results on google.com. After further investigation, they are slightly different, and I'll explain our intent.

Today, we're intentionally showing less local results for web design / SEO queries. For example, [web design sacramento] doesn't display local listings today. We believe this is an accurate representation of user intent. In some cases, we do show local listings, however (as NSNA/php-er noted) [web design in bellingham]. I'm sure some of you feel we should be displaying local results for queries like [Web Design Vancouver]. I understand that concern, but based on our understanding of our users, we feel this is the right decision for now.

I'll give the usual disclaimer that we're constantly working on improving the user experience and results will vary over time. So, this could change in the future, but I wanted to be explicit about what we're doing today.

Yea, Joel said that. Of course, the logical thoughts by SEOs on this matter is that Google really does hate SEOs. This may be the last Google vs. SEOs controversy of 2009.

Forum discussion at Google Maps Help.

Those of you who are patiently waiting for Google to fix the verification tag in the Webmaster Tools API for the past two months just has to wait a bit longer.

In short, those who are tying to use the Webmaster Tools API to verify their sites with Google Webmaster Tools are running into issues. The reason is, the meta tag being generated via the API is wrong and needs to be updated by Google. This has been going on since October 2009.

Dennis G. from Google promised he will finally update it to work in January, sometime. He said:

We've partly updated the API already. The rest will be fixed in January. Sorry for the delay.

Forum discussion at Google Webmaster Help.

Conduct a search for [blackberry news] (Note: I wouldn't click on those results, they may infect your computer) in Google and see page two and three of the results. The results are inundated with weird results from random domains that redirect to CNN.com. The redirect is a 302 redirect and the header looks like:

HTTP/1.1 302 => Date => Wed, 30 Dec 2009 12:58:52 GMT Content-Type => text/html Connection => close Server => Apache Content-Length => 0 Location => http://cnn.com

Here is a picture of just two of those results:

For a full page of results, see this screen capture.

Now, I do not know much about hijacking search results in Google. I know this was somewhat of a major issue in 2005. Then in 2008, we reporting page hijacking wasn't an issue in Google anymore, until about six months ago, when it popped up again.

So maybe this is a new example of a 302 hijack attempt. This time targeting the popular news source, CNN and going after the keyword phrase [blackberry news]?

Forum discussion at DigitalPoint Forums.

A Google Webmaster Help thread, written in Italian, has confirmed reports of a bug between Google's Blogger and Google Webmaster Tools.

Dennis G. from the Google Webmaster Team, who is specifically involved in the verification process, wrote in a Google Webmaster Help thread:"

It looks like we have a bug in our blogger integration. I'll go fix that. In the meanwhile, try verifying by meta tag instead.

It seems like he is the man that actually goes in to fix the "bug." How cool is that?

In any event, if you are having problems verifying your Blogger blog with Google Webmaster Tools, know that you are not alone.

There is no ETA for the fix, but Dennis made it sound like it is a quick fix.

Forum discussion at Google Webmaster Help.

Whenever an SEO talks about geo-targeting your site to a specific region, so it ranks well on that localized version of Google they tell you to do a few things.

(1) Host in the country you want to rank well for
(2) Try to use a ccTLD for that country, i.e. domain.co.uk for UK
(3) Set the geographic target in Google Webmaster Tools

However, instead of using a ccTLD, I often see sites deploy a subdomain or subfolder to specify a different language or geographic target. A ccTLD is best, in my opinion. Of course, there are times you want to use a localized language by target the main Google - so that does apply.

JohnMu from Google replied to a Google Webmaster Help thread stating that a ccTLD is much more powerful than hosting your site on a server in that country. Specifically, John said:

Yes, we do try to find context from these two factors (I think this article is being updated to be a bit more clear though :-)) -- however, if your site has a geographic TLD/ccTLD (like .co.nz) then we will not use the location of the server as well. Doing that would be a bit confusing, we can't really "average" between New Zealand and the USA... At any rate, if you are using a ccTLD like .co.nz you really don't have to worry about where you're hosting your website, the ccTLD is generally a much stronger signal than the server's location could ever be.

Google will use the ccTLD over the server location to geotarget your site. This makes sense to me, but I do not think this has been covered here. We covered the overal topic dozens of times, but not specifically which factor is more important.

Forum discussion at Google Webmaster Help.

A WebmasterWorld thread has discussion around getting Google to index a popular image feature sites use to show off images on their web site. It is called Lightbox JS and it basically uses JavaScript to open up a neat larger view of the image on the page.

I use it on many sites, but you can see a quick example on the RustyBrick Mobile Portfolio. Just click on the image and it opens up a larger picture of that image. Here is a screen capture showing the larger image as it overlays on top of the page:

The issue is, GoogleBot is having a tough time capturing these images in their index. WebmasterWorld administrator, Tedster, explained:

I've been up against the same challenge. Even though regular Google search is aggressively discovering URLs and content by spidering JavaScript, apparently the image bot is not so inquisitive at this point. This surprised me, because there are many images being displayed through Lightbox scripts these days.

Yes, GoogleBot is able to execute JavaScript, but is GoogleImageBot able to at the same pace?

Tedster is exploring other ways to get GoogleBot to index Lightbox JS. He tried the following method, but it doesn't seem to work:

My latest attempt involves making the anchor part of the link a thumbnail image - but the thumbnail is not just a smaller version of the larger image. I use the same exact image file for the anchor, but I resize it on the the page with the HTML width and height attributes. This means that the page loads more slowly, but at least the image bot gets a direct <img src=[url]> style mark-up.

If you have a solution, let us know.

Forum discussion at WebmasterWorld.

A couple months ago, Google released an incredibly useful feature in the Webmaster Tools labs named fetch as GoogleBot. It basically allowed you to see what GoogleBot sees, enabling you to see crawl issues, hacks, injected links and other webmaster related issues as a GoogleBot. But when it came to PDFs, I don't think the tool worked properly (yes, it is in labs).

A thread in the Google Webmaster Help forums has one webmaster asking why the feature doesn't work with his PDFs. He asked:

For example, in the URL in question, http://www.knowitall.com/literature/spec/95731_Pharmaceutical_Excipients_Spec_Sheet.pdf, the text "Pharmaceutical Excipients Database" is in the pdf, but in the "Fetch as GoogleBot" results window, none of those terms are found--the results are basically in binary format. The document is found by the Google Search engine so it is apparently extracting the human readable text.

I couldn't run a test on that document, but I used a PDF on my server to compare. I ran five different tests on two different domains, with a bunch of different types of PDF documents and they all came out with gibberish binary format results in the fetch as GoogleBot. Here is one sample screenshot:

Not, Susan Moskwa from Google said in that thread:

FYI we're looking into this issue, so sit tight. If it looks okay in search results the problem is probably not with your site (we've been able to reproduce it for other sites as well). Thanks for letting us know.

So it seems like they may get the fetch as GoogleBot feature working for PDF documents?

Forum discussion at Google Webmaster Help.

The canonical tag was jointly introduced by Google, Yahoo and Microsoft earlier this year. Google hinted they would soon support cross domain canonical tag and they officially announced it last night. What is a canonical tag? It basically allows you to communicate to Google that a page has moved to a new location, without you using a server side 301 redirect.

Why is this important? A few reasons:

(1) Some hosts don't allow webmasters to deploy 301 redirects
(2) Some site owners aren't technical enough to implement a 301 redirect
(3) In some cases, webmasters do not want to redirect users but rather only search engines (i.e. pagination, weird filtering, tracking parameters added to URLs, etc).

Initially this was launched to work within a domain only. So you could only say that domainA.com/pageA has moved to domainA.com/pageB. Now you are able to do this across domains, so you can say, domainA.com/pageA has moved to domainB.com/pageA or even domainA.com moved to domainB.com.

Google made this cute graphic to explain it:

There is a lot of technical detail on how to implement this tag over here but in short, you just add this tag to the head section of your HTML of the page you want to redirect. Make sure the URL you specify is the URL you want GoogleBot to be redirected to:

<link rel="canonical" href="http://www.example.com/true-url.html" />

JohnMu, who wrote the Google blog post also put some FAQs, and note the "hint" he keeps referring to. I assume the "hint" will become less of a hint and more of an "instruction" as time goes on.

Q: Do the pages have to be identical? A: No, but they should be similar. Slight differences are fine.

Q: For technical reasons I can't include a 1:1 mapping for the URLs on my sites. Can I just point the rel="canonical" at the homepage of my preferred site?
A: No; this could result in problems. A mapping from old URL to new URL for each URL on the old site is the best way to use rel="canonical".

Q: I'm offering my content / product descriptions for syndication. Do my publishers need to use rel="canonical"?
A: We leave this up to you and your publishers. If the content is similar enough, it might make sense to use rel="canonical", if both parties agree.

Q: My server can't do a 301 (permanent) redirect. Can I use rel="canonical" to move my site?
A: If it's at all possible, you should work with your webhost or web server to do a 301 redirect. Keep in mind that we treat rel="canonical" as a hint, and other search engines may handle it differently. But if a 301 redirect is impossible for some reason, then a rel="canonical" may work for you. For more information, see our guidelines on moving your site.

Q: Should I use a noindex robots meta tag on pages with a rel="canonical" link element?
A: No, since those pages would not be equivalent with regards to indexing - one would be allowed while the other would be blocked. Additionally, it's important that these pages are not disallowed from crawling through a robots.txt file, otherwise search engine crawlers will not be able to discover the rel="canonical" link element.

Forum discussion at Google Webmaster Help.

Over a month ago, there were reports in two Google Maps Help threads that the local pack for searches in the flavor of [web design], [web development] and so on, were not triggering local pack results. We covered this about three weeks ago with our story named Google Bug Removes Web Designers in Local Pack and it is still not fixed.

Back then, a Googler said the reason for the bug had to do with a one box local result bug being fixed to show a local 7 pack. That bug was fixed, but it triggered a new bug with queries such as [web design], [web development] and so on. Even if you add a local qualifier to such queries, i.e. [web design new york] or [web design 10010] the local pack is not triggered at all.

Google never offered an ETA for when this would be fixed, but it has been a month since a Googler responded to the issue and we have yet to get an update, while the bug lingers on. Google, when will this bug be resolved?

Forum discussion at Google Maps Help.

Since Google has released the Site Performance reports in Webmaster Tools and page speed is Google's 2010 ranking factor promise, Webmasters have been somewhat obsessed with speeding up their sites.

Webmasters are concerned with scripts they do not control. Such as Google Analytics slowing down their page load times. In fact, we just reported that Google is going to add a trustworthy indicator to site performance reports in order to alleviate some of the webmaster stress over speed performance.

Now, we see reports that Google AdSense is slow. But instead of Google just telling us not to worry about it, which they have told us for this. Google is actually going to speed up the AdSense load time specifically for GoogleBot.

AdSenseAdvisor said in a WebmasterWorld thread two things. The first was not to worry about it and the second was that they will be improving the speed of AdSense specifically for GoogleBot.

Let me quote AdSense Advisor:

(1) "The Page Speed suggestions on Webmaster Tools are based on the content served to Googlebot. In this case, AdSense serves the javascript uncompressed to Googlebot, but does actually compress it with gzip for regular users. So, the AdSense javascript is already optimized for your site's users. AdSense works very hard at making sure that it doesn't slow down the page load."

(2) "We are changing AdSense to send compressed content to our Googlebot so that the recommendation in Webmaster Central accurately reflects how AdSense works."

Now some may ask if this is a form of serving different content to GoogleBot versus a human - i.e. cloaking. I don't think this falls under that.

Forum discussion at WebmasterWorld.

A Google Webmaster Help thread has reports of page load time speeds spiking up to ridiculous numbers in the new Google site performance reports. Google's response to these reports was pretty interesting, I'll get to that soon, firs the context.

A Top Contributor in the forum wrote:

After about 6 months of "flat line" Site Performance reports of averagepage load time around 1 or 2 seconds, I am now seeing in Tools a report that: "On average, pages in your site take 83.1 seconds to load (updated on Dec 7, 2009)." and of course the graph has shot up and I'm tol that my site's average page load time is "slower than 100% of sites".

However, the only two pages listed in that report both show load times of 1 to 2 seconds.

Now a Googler with the code name "sreeram" replied saying:

The 83s number is bogus. Your site's toolbar traffic dropped by more than an order of magnitude in the last few days. You should ignore the average for now. We'll soon be showing site owners some indication of how trustworthy the numbers are, so you can decide when to ignore it and when not to.

Not all URLs may have toolbar traffic, so it's possible to have many URLs indexed, and even visited by users, but only a couple may show up on Site Performance. In addition, when there's very little data for a given URL, we won't display it (for privacy reasons), though it will be included in the overall site average.

So in this case, the site's traffic as seen by the Google Toolbar dropped significantly, which caused a weird spike in the webmaster's site performance reports. Thus, Google promised to provide an "indication of how trustworthy the numbers are" in this report.

Clearly, some of these numbers are not trustworthy, such as factoring in Toolbar fluctuations or Google Analytics speed.

Forum discussion at Google Webmaster Help.

Update: John Mueller from Google sent me a note about this:

The Webmaster Tools team is constantly working on ways to improve the product as well as the data provided there. In general, we prefer not to comment on possible future releases. The Labs section in Webmaster Tools allows us to easily try out and iterate on new and innovative features, which is one reason we launched the Site Performance tool there. Personally, I found the data provided there quite actionable and have seen a lot of positive feedback from webmasters around this tool. To fine-tune a website with regards to speed, it can be useful to start with the information provided here and then to look into the details using browser-based tools such as Page Speed and Speed Tracer.

We're always looking into ways we can take our products and services to the next level. We appreciate all of the feedback and coverage that you provide! I'll get in touch with you once I have more information that I can share.

A Google Webmaster Help thread has one webmaster worried that Google will be using page speed as a ranking factor in 2010. But the reason he is worried is because the site performance feature in Webmaster Tools shows Google Analytics as needing some speeding up.

The two suggestions this webmaster gets from Google is related to their own product, Google Analytics. The suggestions read:

(1) Compressing the following resources with gzip could reduce their transfer size by 15.5 KB:
* Go to URLhttps://ssl.google-analytics.com:443/urchin.js (15.5 KB)

(2) The domains of the following URLs only serve one resource each. If possible, avoid the extra DNS lookups by serving these resources from existing domains:
* Go to URLhttps://ssl.google-analytics.com:443/urchin.js

A Googler named sreeram in the forum said it is okay to "ignore the suggestion to gzip urchin.js." Google actually noted this issue in their announcement where they said, "some servers return uncompressed content for Googlebot, similar to what would be served to older browsers that do not support gzip-compressed embedded content (this is currently the case for Google Analytics' "ga.js")."

That response did not satisfy the concern of this webmaster who asked again:

The question is if the speed impact of using these Google products will affect my sites' rankings?

I hope not, but I also know if has affected my users experience on my site and that affects me.

Google did release asynchronous calls for Google Analytics, which does speed things up a bit.

Forum discussion at Google Webmaster Help.

So, SEOs are freaking out on Google personalized search, now that it is the default. Can't blame them really but now it is time for action, right?

David Harry on the Huomah blog put together a white paper as a zipped PDF file back in October 2008, which he promises to update, with some key findings on how Google personalized search impacts SEOs.

Let me quote his key findings back then, but downloading the white paper might be worth your while.

David concluded: No two SERPs were the same (personalization ON or not):

• Personalization re-rankings are minimal (for informational queries)
• Establish geo-graphic baselines (or segment data even)
• Top 4 positions are primary targets
• Top 10 are secondary targets
• Top 20 may be leveraged through behavioural optimization

In the Sphinn thread discussing this, David said "we're working on another round in light of the recent announcement." He doesn't expect it to change much, but he will look into it.

Forum discussion at Sphinn.

Google announced a dozen things yesterday, but one of those announcements was on real time search in the search results. Google teamed up with Twitter, Facebook, MySpace, FriendFeed, Jaiku and Identi.ca to make these real time search results possible.

In short, Google sticks another "universal result" in the search results when they feel a real time result makes sense. For example, a search for popeye, which is currently trending due to the Google Doodle Popeye logo, shows this real time result:

Yes, that is my tweet showing a link to this site. In fact, it also shows other Tweeting links to my story. Here is a link to the real time landing page results.

Here is a video explaining this in more detail:

We have threads on this announcement at WebmasterWorld and DigitalPoint Forums. The consensus in the threads are that this clutters up the search results even more and do not provide a simple interface for people to find relevant results. Of course, most SEOs don't want their web results pushed down in the search results page - but SEOs will adapt and use Twitter, Facebook and so on to make sure their results do show up in that area when applicable.

Forum discussion at WebmasterWorld and DigitalPoint Forums.

Back in February 2007, Google began personalizing search results based on your search history. Google would do so, only when you were logged in. Now, Google announced they are personalizing search results for everyone, even if you are not logged in.

Google explains:

Previously, we only offered Personalized Search for signed-in users, and only when they had Web History enabled on their Google Accounts. What we're doing today is expanding Personalized Search so that we can provide it to signed-out users as well. This addition enables us to customize search results for you based upon 180 days of search activity linked to an anonymous cookie in your browser. It's completely separate from your Google Account and Web History (which are only available to signed-in users). You'll know when we customize results because a "View customizations" link will appear on the top right of the search results page. Clicking the link will let you see how we've customized your results and also let you turn off this type of customization.

This is big stuff for the SEO world. No matter how much we complained earlier that the personalized search results made it hard to "optimize" for, this only impacted a fraction of all Google searchers. Most people had no idea how to even log into Google. Now, Google is personalizing results by default and these same people will have no idea how to turn off personalization. So now all these users will see different results, on some level, and it does impact the SEO space - more so than when Google launched this in 2007.

SEOs, most of them, are mad. Let me quote some of the more trusted names in our forums:

Cre8asite Forums:

So let me get this straight: to completely opt-out, you'll have to login permanently? Which moron came up with that scheme?

Wow... this is really going to overly favor websites like amazon and ebay that already have a huge reach.

The rich will get richer.

HighRankings Forums:

Google is using a "secret cookie" not attached to the Google.com domain to handle the Web History/Personalized Search for users who are not logged in.

The gradual impact over time will be that unaware users who click on Google search listings will reshape their search results. Unless Google develops some sort of injection to show people new (fresh) content, discovery will become more difficult for less sophisticated users.

Good news for Randy and 1dmf's Google Rank Extractor. And also for moving people further and further away from traditional rank checking as a measure of success.

DigitalPoint Forums:

Google are there to find ways to best meet the needs of the searcher. As much as we want them to be, they are not there to meet our SEO needs.

WebmasterWorld:

The possible impact to all is staggering.

This just feels so very wrong-headed that the mind boggles! As I often explain to IT staff, NOT EVERYTHING THAT CAN BE PROGRAMMED SHOULD BE PROGRAMMED.

Sounds like an attempt at a product improvement. SEOs may or may not like it, but the real question is whether Joe Searcher likes it. If Joe does, Google will remain one big step ahead of its competitors. If Personalized Search turns out to be Google's answer to Windows Vista, Google will need to bring back Google XP or leapfrog to Google 7.

Danny Sullivan has an excellent write up about this at Search Engine Land, he doesn't seem to touch on the SEO view of this.

Honestly, if this makes Google more relevant for their searchers - all power to them. I am not going to say I know what is better for the average searcher. They can quickly figure this out after days and decide to keep it, turn it down, or off completely. SEOs will adapt, like they always do - we are a strong and smart bunch - I am not worried.

Forum discussion is plentiful and it is at Cre8asite Forums, HighRankings Forums, DigitalPoint Forums and WebmasterWorld.

Like all you know, page speed with be a ranking factor in Google in 2010 and if you didn't believe it until now, maybe Google adding a page speed report to Google Webmaster Tools will convince you of that.

I figured I show you detailed screen shots of how slow this site is, using this tool.

As you can see from the chart below, this site is slower than 86% of the other sites on the Internet. The chart does slow slight improvement, but clearly, I need to clean things up here - quick.

Then Google shows me example load times for specific pages on my site:

Then Google offers speed suggestions:

I can drill down deeper on a specific page to see exactly what is suggested:

Figured I slam my own site over someone elses - why not.

Forum discussion at Google Webmaster Help and WebmasterWorld.

About two weeks ago, Google introduced breadcrumb display URLs in the search results. It was suppose to be for "small percentage of search results" but they are showing up quiet often. Google said they are doing this for URLs that are "too long, too short, or too obscure to add useful information." So Google adds the useful information for you.

Not all webmasters agree that Google is improving their URL structure with the display URL as a breadcrumb. A WebmasterWorld thread has some webmasters complaining and I know, WebmasterWorld doesn't allow examples. So for you Googlers out there that want examples, I'll give you one.

A search for [rustybrick siddur] brings up this new breadcrumb URL. It looks like this:

Now, if this was smart, it would either leave my URL alone, because I feel it is pretty short and very descriptive the way it is. Or Google would show the true breadcrumb which is > iPhone Apps > Siddur as shown here:

Now, it works beautifully for [rustybrick iphone] but even there, is it really needed to change the display URL from rustybrick.com/iphone.php to rustybrick.com > iPhone Apps? I am not too sure.

Forum discussion at WebmasterWorld.

About two weeks ago, we reported on rumors of Sitelinks linking to the wrong site. Well, those rumors have just been confirmed by a Googler.

JohnMu in a Sitelink, can link to a third-party site, likely a competitor. John said:

This looks like something that happened long ago because of a bug on our side that's persisted because you're blocking the Sitelinks. Once you unblock them, they can go away on their own :). Sorry about the confusion!

So yes, there was a bug when you blocked a Sitelink that you did not want. You had to unblock them to make the Sitelink go away. Go figure.

I doubt most people will know how to do this or even think this is logical, so I do hope the bug is fixed and Google does this for all sites that are impacted.

Forum discussion at Google Webmaster Help.

Friday, I reported that there was a possible Google update taking place. Truth be told, it was actually webmasters hitting on the data center which showed the Google Caffeine results.

Matt Cutts of Google confirmed that Caffeine was going live on 209.85.225.103, at least for 50% of the users hitting that data center. He also said that I was totally off in reporting that Google is doing a major update - they are not. What people are seeing is the Caffeine results on that single data center.

Google did tell us that they would not fully launch Caffeine until after the holiday season. In fact, they promised to launch Caffeine on a single data center, and this is what they did.

So the update is not a full update, just a Caffeine update on a single Google data center that should not have a huge impact on online retailers during this holiday season.

I should clarify that my post on Friday did actually say that this can be that Caffeine was live on a single data center and those reporting the update was indeed only looking at the data center. In any event, Caffeine is now live on that single data center (at least 50% of the time), so check it out at 209.85.225.103.

Forum discussion at DigitalPoint Forums, Google Webmaster Help and WebmasterWorld.

If you are a patent junky, then you may like this relatively new one. A patent application submitted November 19th by Google named artificial anchor for a document may shed some light on the jump down Google snippets or anchor based Google Sitelinks we have seen creep into the Google search results in August and September.

As always, Bill Slawski converts the patent into English for all of us to read. But here is the abstract:

Methods, systems, and apparatus, including computer program products, for linking to an intra-document portion of a target document includes receiving an address for a target document identified by a search engine in response to a query, the target document including query-relevant text that identifies an intra-document portion of the target document, the intra-document portion including the query relevant text. An artificial anchor is generated, the artificial anchor corresponding to the intra-document portion. The artificial anchor is appended the address.

Currently, I am way too tired to look into this right now, so just hit the thread at WebmasterWorld to dig deeper.

Forum discussion at WebmasterWorld.

Update: Matt Cutts from Google confirmed that Caffeine is live 50% of the time in the 209.85.225.103 data center. See below for more info.

There are several threads reporting major changes in the Google search results, starting yesterday. We have threads at DigitalPoint Forums, Google Webmaster Help and an updated WebmasterWorld thread.

In short, people are noticing major changes in the search results. Some believe it is a Caffeine index rolling into their search results. Some believe it is just a normal update and not Caffeine related. We do know that Matt Cutts promised not to roll out Caffeine until after the holidays and a roll out at this time would be the worst possible time for online retailers (unless you rank very well in Caffeine.

We also know Google is eventually rolling out Caffeine to a single data center, which can be why many people are noting an update. But again, it might be that these folks are on this single data center or that this is not a Caffeine update. I certainly hope this was not a mistake and Google released the Caffeine update prematurely.

The data center people are talking about now is 209.85.225.103.

People are saying that their traffic is down because of it, but some are also noting that it is/was DigitalPoint Forums, Google Webmaster Help and an updated WebmasterWorld.

Update from Matt Cutts:

My plane landed and I had time to check. 209.85.225.103 as an IP address currently hits the Caffeine data center about 50% of the time, so that's probably what the person at DP was seeing.

I think it's really risky to take a few threads and turn that into a "major Google update." For example, one of the links you pointed to was asking about the rankings for their site webtlk.com. Given links like http://74.125.47.132/search?q=cache:iwkj8P8xoSEJ:teqtonik.com/+http://teqtonik.com/&cd=1&hl=en&ct=clnk&gl=us&client=firefox-a talking about data recovery and pointing to that site, I'm guessing that Google just got better at determining how much weight should be given to links to that site.

So 209.85.225.103 does hit the Caffeine data center more often than other IP addresses, but it's always been the plan that Caffeine would roll out at one data center (no more data centers will get Caffeine until at least January).

But I would avoid generalizing, at least as going by the one report that mentioned a specific site that I looked into. Of course, on a lot of the webmaster forums you can't tell what site they're talking about, so it's hard to do any debugging on what might have happened to individual sites there.

Google News SEO is a fun topic for me because we are included in Google News. Two recent interviews were published on how Google News ranks articles and yes, it is very different from rankings in Google Web Search.

Danny Sullivan at Search Engine Land and Eric Enge at Stone Temple both ask the tough questions with Josh Cohen, Senior Business Product Manager for Google News.

I suggest reading those interviews and then joining the WebmasterWorld discussion on the topic. Some really good analysis is going on there on these interviews.

Forum discussion at WebmasterWorld.

First blogged at BigMouthMedia, Google's Rich Snippets feature which went live in the U.S. in May is now showing up in other Google properties, such as the UK, Canada and other English based language countries.

When it first launched, it was only based in the US. But now if you try to bring up sites that provide rich snippets to Google outside of the US, some show them. For example, Google UK, Google Canada and Google Ireland all show them, but Google Israel, Google Mexico and Google Italy do not. I believe it has to do with the language of that country.

A screen shot of Rich Snippets now showing in the Google UK:

A screen shot of Rich Snippets not showing in Google Mexico:

This was first spotted in a thread we covered earlier that felt rich snippets were coming to smaller sites, still not sure if that is true.

Forum discussion at WebmasterWorld.

A neat new WebmasterWorld thread is trying to do the impossible by putting down on paper, what exactly are the 200 or so search engine ranking factors used by Google.

The thread creator put out a few dozen ranking factors, but there is a way to go here. Think this is a possible task?

Even if they do get all the 200 ranking factors down, figuring out the weights assigned to each and how Google scores them based on other factors can be a bit complex. SEOs have been trying to do so since even before Google was born. Back then, it was a lot easier, now - it is virtually impossible. I bet Matt Cutts, who has access to this data, wouldn't be able to get it exact (no offense).

Anyway, trying to come up with the list, in itself, is useful on some level. But understanding where to put your emphasis and time in, is what is most important for SEOs to understand.

Forum discussion at WebmasterWorld.

Now, if you do a search for web design, web development, or most digital agency like queries, with or without a local qualifier (i.e. web design new york), Google won't show the local 7 pack. Mike Blumenthal covered this well, but it needs more attention.

In the past, a search for [web design, suffern, ny] or a search for [web design] when Google knows your location, would trigger a 7 pack that looks something like this:

Now, the same search triggers no local pack, just search results, like the old Google results. Here is a picture:

There are two large threads about this at the Google Maps Help thread. In one of those thread, Googler, Linda said on November 11th, "Thanks for your reports. I will look into this issue and get back to you."

Then Brian B from Google explained on November 18th what is going on. He said:

This looks like it's closely related to the issue going on at the thread I've linked below. We realize there's something going on here, and we initially pushed out a fix a while back. There was a little hiccup with the fix, which is probably why the results in Fresno may have gone back up and then back down as addoctane mentioned above.

The team is working on this issue. Stay tuned to the thread below where I will post an update as soon as I hear one.

Thank you all for your persistence!

Back in late October, there was a bug we covered where Google Maps Help.

A WebmasterWorld thread has one webmaster who allegedly purchased a domain name that was banned by Google. He basically asked, how does he get this domain to be clean and start over.

In the past, I would have said, just submit a reconsideration request and hope that Google trusts that you are starting new. And that is the advice given in the thread.

But after sitting in a site clinic with Matt Cutts at PubCon. At about 10:51 PST, Matt said that when it comes to domains that were banned or severely penalized, it is just easier to start with a fresh new domain. I am not quoting Matt exactly, but he basically said that it is easier to start from ground zero then starting from 100 feet under ground. Yes, a penalized site seems to just have to work its way up from a negative state to get to level starting ground and often, it might not be worth the effort.

At the same time, if Google can press a button to wipe the history of the domain - then maybe you are in luck. I am just not too sure how easy it is to get Google to do that.

Forum discussion at WebmasterWorld.

The other day, I noticed a thread at Google Webmaster Help where a person was complaining that they were being hit hard by GoogleBot. In short, Google's spider was crawling his site in a very aggressive manner. He said:

After setting a custom crawl rate using webmaster tools (and robots.txt for good measure) GoogleBot's crawl rate slowed to the specified 1 request per (approx.) 60 seconds. However, as of a few hours ago the crawl rate has increased to an inexplicable 1 request every 2 to 3 seconds which is unacceptable - I've had to take the drastic response of adding a site wide disallow until I can get this resolved. Why would the crawl rate increase in speed so and how can I get it to return to the values specified?

Googler, Jonathan Simon, replied and said there was a temporary bug that cause the issue. He apologized and said it was now resolved. Jonathan said:

I took a look into what you've reported and it turns out that there was a temporary issue in the processing pipeline of the crawl rate settings which caused them to not take effect. This issue is now resolved so crawl rate settings should once again behave as expected.

I often give Microsoft a hard time about MSNBot being too aggressive and not listening to webmaster directives. So this time, I felt I point out an issue with GoogleBot.

Forum discussion at Google Webmaster Help.

A WebmasterWorld thread reports that someone is noticing that his site's Sitelinks in Google are pointing to a different site. I am not sure how that is possible. How can Google link a Sitelink for one site to a different site? But according to this webmaster, it is true.

I do not have any proof that this is happening, nor will I ever get proof because WebmasterWorld does not allow examples. But I believe it is possible that with DNS changes and shared IPs, it is possible that Google can be confused and link to the wrong place.

There were cases where Google got the wrong site for the info operator and the cache being the wrong page, but never a Sitelink.

If this happens to you, the first thing you should do is block those Sitelinks in Google Webmaster Tools. Then post evidence at Google Webmaster Help.

As Tedster said at WebmasterWorld:

Sounds like a particularly nasty data bug. Does that sitelink show up in your Webmaster Tools account? If so, you can veto it. If not, you could take the issue up on the Google Webmaster forum or try a reconsideration request.

Forum discussion at WebmasterWorld.

Page load time (speed) is a factor currently in the AdWords quality score. But soon it may be coming to Google's organic ranking algorithm. If you have a really slow site, it may impact how high you rank in Google. That was the main news coming out of PubCon last week, minus the Caffeine launch.

It is currently not in the algorithm, according to Matt, but who knows - maybe they are testing this already. Matt was clear that Google wants the web to be a faster place and Google does control much of what people see on the web. So Google can influence that people find faster web pages, over slower ones.

You can hear Matt talk about this 2 minutes and 52 seconds into this video:

Google also has a tool to test page speed at http://code.google.com/speed/page-speed/ - so get ready.

I should add, Google has hundreds of ranking factors. Adding one more, depending on the weight they assign to it, shouldn't shuffle things up much for most sites. Just make sure your site loads fast - it is a good thing to have anyway.

Forum discussion at WebmasterWorld.

The Google AdSense for Feeds blog announced they will be tagging your FeedBurner feeds with Google Analytics parameter tracking. I have been seeing this for several blogs for the past few weeks already. Basically, this means, Google is appending variables to the URL.

For example, if you look at one of our latest posts, the URL is http://feeds.seroundtable.com/~r/SearchEngineRoundtable1/~3/KwUDQ61JSkg/021166.html. If you click it, it takes you to http://www.seroundtable.com/archives/021166.html?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+SearchEngineRoundtable1+%28Search+Engine+Roundtable+1.0+RSS%29.

The purpose is to make tracking better in Google Analytics but clearly, this is just messy. Heck, I have been linking to these URLs via the SearchCap for the past couple weeks and now it is going to get a lot worse. I am going to be linking to a URL that is not the parent URL.

This is the topic of concern at Cre8asite Forums. JohnMu from Google came into that thread to offer some SEO advice:

- Move to "#" for these parameters, which will effectively hide them from search engines. There are a few articles on this, eg http://esev.com/blog/tutorial/hiding-google-analytics-campaign-variables/

- Use the rel=canonical link element: http://googlewebmastercentral.blogspot.com/2009/02/specify-your-canonical.html

- For Google, use the URL parameter handling tool to tell us to ignore these parameters: http://googlewebmastercentral.blogspot.com/2009/10/new-parameter-handling-tool-helps-with.html

In general, we'll try to figure these things out on our own (as will the other search engines), but personally I like to be in control so I'd probably try something like that.

Or you can turn off the tracking, the Google blog explains how but here is that info:

If you're not using Google Analytics, or for some other reason don't want these parameters in the requests coming to your website, you can turn off Google Analytics tracking on the "Configure Stats" page on the Analyze tab at http://feedburner.google.com.

Got all of that. You should take action, as I should as well.

Forum discussion at Cre8asite Forums.

As we all know, Google Caffeine is launching on a single data center until after the holidays so that online retailers don't get a nasty present from Google.

As I promised then, I would let you know what data center the results went or go live on, when I hear. There are some discussions and chatter at WebmasterWorld that one data center may have Caffeine results. Google has not confirmed it, nor do all webmasters believe these are Caffeine like results.

The data center is 216.239.59.103 and you should give it a try yourself. Matt said that even if they do give out an IP address, it is possible that non-Caffeine results will show on that data center for some people. So maybe this is Caffeine and some don't see it? I am not sure - I am pretty sure Matt can confirm this IP as the data center or not - after he gets back from Las Vegas.

Forum discussion at WebmasterWorld.

Update #1: Some are saying this data center has moved to 66.102.7.18.

Update #2: Google's Matt Cutts replied to me on Twitter saying, "@rustybrick I don't think that IP points to Caffeine."

The Google Webmaster Central Blog announced they have updated the keyword tool and data in Google Webmaster Tools. The new additions include:

• Data updated daily
• How often is the keyword found
• Displas top URLs that contain the keyword

Specifically, the new significance column "compares the frequency of a keyword to the frequency of the most popular keyword on your site." Google adds, "when you click on a keyword to view more details, you will get a list of up to 10 URLs which contain that keyword." Why is this important, well, if someone does hack into your site, you can easily see if unrelated keywords are showing up and on which pages. Plus, it is just a good tool to see how Google understands your web site.

Here are pictures of the report for this site:

Most importantly, the Google Webmaster Team's Halloween costume not only rocked, but coordinated.

Forum discussion at Google Webmaster Help, DigitalPoint Forums and Cre8asite Forums.

Back in May, Google launched rich snippets that enabled webmasters to markup their HTML with richer data for Google to display. For example, you can often see results from Yelp and other sites like it for reviews that display reviews in the Google results. Here is a picture:

In fact, most webmasters said they will markup their HTML to take part of this richer snippet experience. Then in September, Google added a rich snippet testing tool because although many marked up their HTML, they never knew if it was actually working because Google rarely displayed the richer snippets for smaller sites.

Now, I am hearing via a WebmasterWorld thread that some smaller webmasters are now noticing rich snippets being displayed for their sites. I tried a few sites myself and was not able to see it myself, maybe he is hitting a different Google data center. I do hope that rich snippets do make its way down to more sites so that the playing field is evened up a bit.

Forum discussion at WebmasterWorld.

Yesterday we reported that the Google Caffeine index is launching right before the holiday season. I wrote my post like that, simply because I was in shock to see this being rolled out, even "slowly" before the holiday season. Honestly, I believe this took Google's Matt Cutts by surprise as well. While on vacation/work in Las Vegas, Matt wrote a blog post explaining this will not be launched on more than one data center prior to the holiday season.

Again, this is not fully rolling out until after the holiday season. Thank you Matt.

Matt also broke his one year streak of not posting at WebmasterWorld, with a post in the Caffeine WebmasterWorld thread yesterday saying:

I thought about including "Don't Panic!" from the Hitchiker's Guide to the Galaxy in my post, but decided not to. :)

cangoou, in a little while, you'll have Caffeine to try out at a full data center. I don't expect the results to change much from the developer preview to the data center, nor from the data center to the full roll out.

barretire, some people might be stressed by waiting, but I think more people were stressed about the idea of Caffeine rolling out during the holidays.

At any rate, I'll be talking at PubCon on Thursday, so I can discuss Caffeine and can answer questions for folks that are still stressed. :)

I probably stressed a few people out with my original post. But we needed clarification and we got it. Personally, I think this all came to a shock to many of the folks at Google Webmaster Central and Matt Cutts but it should not be an issue for most online retailers until after the holidays. Even then, Google hopes the search results are somewhat transparent to searchers and webmasters.

Forum discussion at WebmasterWorld & Sphinn.

A Google Webmaster Help thread has discussion where one webmaster is having the hardest time removing a URL from the Google index. He/she used the Google Remove URL Tool only to find out that even though Google reported back the URL was removed, two weeks later, it is still there.

Susan from the Google Webmaster Central team told the individual to try again and select "outdated or dead link" option in the form. This time, the user received an error that read:

You have already made this removal request.

Susan confirmed that this was some sort of bug. Susan said:

Congratulations, you may have found a bug.

Our team is looking into it. I can submit that URL for removal for you in the meantime, since it 404s.

I believe in this case, the page has finally been removed due to Susan taking manual action. I don't think this is a wide spread bug, but just a small bug.

Forum discussion at Google Webmaster Help.

Google has actually released a major Google index update before the holiday season. Yes, this is the Google Caffeine index, which has been in a Sandbox (testing environment) since August, is now being rolled out. If you visit the Caffeine URL at www2.sandbox.google.com, you are taken to a thank you page that reads:

We appreciate all the feedback from people who searched on our Caffeine sandbox.

Based on the success we've seen, we believe Caffeine is ready for a larger audience. Soon we will activate Caffeine more widely, beginning with one data center. This sandbox is no longer necessary and has been retired, but we appreciate the testing and positive input that webmasters and publishers have given.

Yes, this is being rolled out slowly, data center by data center. Some of you may already see it, while most of you probably don't yet. The big question is when will this be fully rolled out? Some have been saying they have been seeing Caffeine in the live index already. I am not too sure.

Why in the world would Google do this right before the holiday season? Did Google forget how the Florida update had hurt online retailers right in the most important time of their shopping season? It was devastating to many small webmasters. I assume Google is confident this will have a small impact on businesses and they are confident it won't shuffle things up too much.

Update: Matt Cutts of Google just promised it won't be rolled out until after the holiday season.

I believe Google's goal with Sandbox was more about infrastructure over the quality of those results. Yes, Google wanted to index deeper and faster and smarter, but the ranking of those results, I believe, they wanted to keep stable from the current index. I hope that makes sense to the readers here, i.e. indexing is different than ranking. But indexing does have an impact on what Google ranks, obviously.

The Google Caffeine Sandbox was shut down about a few hours ago. There is early discussion beginning at the various forums. No official blog post from Google, Matt Cutts or anything, except for that Thank You page.

Forum discussion at Google Webmaster Help, DigitalPoint Forums and WebmasterWorld.

Tedster, WebmasterWorld's Administrator, posted an excellent thread at WebmasterWorld asking how do you think Google handles the canonical issues they find on their own? By that he means, if webmasters don't use a 301 redirect or use Google's canonical header tag to instruct Google on how to handle that URL, how would Google handle it?

Would they cosmetically clean up the search results so that there does not appear to be any canonical (duplicate) URLs in the results? Or do they actually decide to implement a 301 on your behalf and pass all the 'link juice' from one canonical URL to the parent URL?

First, take my poll and then I will give you my thoughts on it:

I really think for the most part, where Google is not confident on which should be the main URL, Google will only apply this cosmetically to the search results. I remember when the new canonical tag came out and Google warned to use this carefully, because it is as powerful as a 301, but without actually being physically redirected. For Google to apply their own 301s, hidden be that, is extremely dangerous for both Google and the webmaster. I would assume, in certain cases, Google does do this, but I am not sure if they do this in most cases. Of course, this is just my thoughts - I have no hard evidence, since I never really tested it myself.

I would assume Google would want to be right 100% of the time on this. I would think that would be a goal. And when they are 100% or even 99% right, implementing this would make sense.

Forum discussion at WebmasterWorld.

What appears to be a newbie, posted a thread at Google Webmaster Help asking why his site is not ranking yet. He also asked if he had something to do with the URL https://internal.google.com/spamteam/users/cutts/matt/ban_list/ showing up in his log files.

Got that?

https://internal.google.com/spamteam/users/cutts/matt/ban_list/

It appears to look like Matt Cutts' personal ban list at Google. But is it really?

(1) His site is indexed in Google.

(2) Why in the world would Google show those types of referrers in people's log files, especially those sites that are banned.

(3) The IP addresses noted by the webmaster did not seem to be from Google.

(4) Googler, JohnMu, said it is not Matt's ban list. JohnMu went on record, and I trust him, saying:

I can assure you that the URL in the title of the thread is not a Google URL -- where did you get it from?

Do you think the URL is real?

Forum discussion at Google Webmaster Help.

A Google Webmaster Help thread reports that because .UA domains are currently not considered "root domains" by Google, the change of address tool doesn't work for them.

I am not sure if Google did not consider them root domains or didn't allow the change of address tool to work with those domains. But Google said, whatever the issue is, it will be fixed soon.

Jonathan Simon from Google replied to the thread confirming the issue:

I've alerted the team to this issue. .UA domains should soon be supported by the Change of Address tool.

Forum discussion at Google Webmaster Help.

Big news for publishers in the Google News index. Inbal from Google announced in a Google News Help thread that in six months, Google will stop supporting the old News Sitemaps format and require you to switch over to the new format.

Inbal said:

Although we will support your current Google News Sitemap during the transition period of six months, you should re-submit it under the new format as soon as possible. For more details on how to submit your Sitemap using the new format (including how to add new tags to each entry in your Sitemap, to provide more information about individual articles), please visit the Sitemaps section of our Help Center.

Once the transition period is over, we will no longer accept News Sitemaps created using the old format and any old News Sitemaps in your Webmaster Tools account will be rejected.

We encourage you to make these changes as soon as possible in order to avoid interrupting your content's inclusion in Google News. To get started, please visit the FAQ.

The new format looks like this:

<?xml version="1.0" encoding="UTF-8"?>
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"
        xmlns:n="http://www.google.com/schemas/sitemap-news/0.9">
  <url>
    <loc>http://www.example.org/business/article55.html</loc>
    <n:news>
      <n:publication>
        <n:name>The Example Times</n:name>
        <n:language>en</n:language>
      </n:publication>
      <n:access>subscription</n:access>
      <n:genres>pressrelease, blog</n:genres>
      <n:publication_date>2008-12-23</n:publication_date>
      <n:title>Companies A, B in Merger Talks</n:title>
      <n:keywords>business, merger, acquisition, A, B</n:keywords>
      <n:stock_tickers>NASDAQ:A, NASDAQ:B</n:stock_tickers>
    </n:news>
  </url>
</urlset>

For more details on this new sitemap format over here.

Again, you have six months to make this change, but the sooner you do it, the better. ALso, there is a learning benefit to the new sitemap file. Any questions, join the forum.

Forum discussion at Google News Help.

Update: The Google News blog just added more information about this important update.

Ronald from the Google Merchant Center (Base) team reported in a Google Merchant Center Help thread that there is a known issue with FTP uploads.

Specifically, if you are uploading new data feeds via FTP, the data won't show up in the dashboard. If your feeds are less than 20MB, then upload manually, otherwise, you are out of luck.

Ronald said:

We've noticed that a few users have reported that data feeds uploaded via FTP are not getting updated in the Google Merchant Center account dashboard. We are currently investigating this issue. In the meantime, if your data feed file size is less than 20 MB, please upload your data feeds manually. We appreciate your patience.

Forum discussion at Google Merchant Center Help.

If your publication is included in Google News and you tried to submit a Sitemap file on Friday, you may have noticed it was rejected. Several publishers have been complaining about this in a few Google News Help threads.

The issue is that some publications are finding out they are no longer included in Google News. They are instructed to submit a Sitemap file, but when they do, they are told the Sitemap file is no good. The thing is, they never really changed their Sitemap file so they don't know what the issue is.

Inbal from the Google News team explained:

We've recently noticed that all news sitemaps which were submitted since Friday were rejected in Webmaster Tools. Our engineers were able to fix this bug and we've just resumed crawling the news sitemaps again. We apologize for the inconvenience in resubmitting your news sitemaps and appreciate your support while we work to improve Google News.

There is no estimated time for a fix, but you should be aware this is a larger issue that Google is addressing.

Forum discussion at Google News Help.

A Google Webmaster Help thread informs us that Google can take about two weeks to remove a blocked Sitelink from showing up in the search results.

Google's JohnMu said:

A change like this can take two weeks or a bit longer, depending on the site and on other factors. I would certainly give it a bit more time :-). Please let us know if you don't see it disappear by next week or so, but from what I can tell it looks fairly normal and nothing appears to be stuck at the moment.

You can block Sitelinks within Google Webmaster Tools by clicking on the "block" button on the right of each Sitelink:

There is also a detailed help document on how to manage Sitelinks within Webmaster Tools.

Forum discussion at Google Webmaster Help.

Five months ago, I wrote a piece named Want Your Google News To Show Up in Google Finance? NASDAQ:GOOG which basically said, put the stock ticker in the title of your article and you're golden for showing up in Google Finance. That is, assuming you are already in Google News.

Well, we now have a new tip from Google themselves on how to optimize your content for Google Finance. Inbal from the Google News team said to markup your HTML or XML with the following tags.

<url>
<loc>http://mynewssite.com/article123.html</loc>

<news:news>
<news:publication_date>2008-10-31T03:30:00Z</news:publication_date>
<news:title>Companies A, B In Merger Talks</news:title>
<news:keywords>business, mergers, acquisitions</news:keywords>

<news:content_types>Subscription</news:content_types>
<news:stock_tickers>NASDAQ:A, NASDAQ:B</news:stock_tickers>
</news:news>
</url>

Adding these elements, along with the title change, should dramatically help you show up in Google Finance for that specific company.

Forum discussion at Google News Help.

Update: Inbal posted an updated version of the sitemap code you also included in your article and if possible she'd like to let the world know about the update. Her last post from here contains the updated code.

Those trying to submit their product feeds to Google Base or Merchant Center may be running into errors. Ronald from the Google Merchant Center team has confirmed a bug that Google's engineers are working on fixing.

A thread at Google Merchant Help has confirmation, where a Googler said:

We're currently experiencing an issue with data feed processing. If the data feed status shows as "Internal error, feed not fully processed" without any error message, please resubmit your data feed. Our engineers are working to resolve this issue.

We appreciate your patience!

There is currently not estimated time of a fix but it is good to know that Google is aware of the issue.

Forum discussion at Google Merchant Help.

For at least a month now, there have been complaints in a Google Maps Help thread that Google is showing, in some cases, a single map result, when it should be showing seven map results.

For example, a search on orlando dodge shows a single map result:

As you can see, it shows a place named Orlando Dodge, but not all the Dodge dealerships in Orlando, Florida. Adding on FL to the query does indeed show multiple listings, including Orlando Dodge.

A Googler from the maps team said yesterday that a fix will be out soon. He said:

Thanks everyone for continuing to post examples of this issue. Hoping to resolve this shortly. Will keep you all posted.

To submit more examples or discussion, join the Google Maps Help thread.

Not sure how important this is, but it is one of those minor changes that I particularly enjoy covering. The Google Webmaster Tools API lets you verify and reverify sites directly in the API. The thing is, they have changed the meta tag name for the verification process and the API has yet to be updated.

For example:

Old Meta Name:

<meta name="verify-v1" content="0wrupKGUMvPxWPChV4qaKdewvwXnZQ9AJmg/mAc7An0=" />

New Meta Name:

<meta name="google-site-verification" content="rlnJZRMNYtwGmLlQYpDhSNnxz41TLLd-3xZSaEIbagw" />

Dennis G. from the Google Webmaster team, who is specifically involved with verification (I believe) said:

We will update the API to return the new meta tag name and content sometime (relatively) soon, but sites that are already verified with the current "verify-v1" tags will not be unverified. Only new verification attempts will need to switch to the new tag.

So I suspect, if you are trying to verify a new site using the API, it won't work just yet.

Forum discussion at Google Webmaster Help.

Historically, Google treated the 404 (page not found) and 410 (gone) server header page status codes as the same. Both meant, the page no longer exists.

Well, that has all changed now. Google is now treating the 410 as "more permanent" than the 404. Yes, this is a minor change but it is likely an important change for webmasters to note.

JohnMu of Google said in a Google Webmaster Help thread:

I followed up on the 404 vs 410 thing with the team here. As mentioned by some others here & elsewhere, we have generally been treating them the same in the past.

However, after looking at how webmasters use them in practice we are now treating the 410 HTTP result code as a bit "more permanent" than a 404. So if you're absolutely sure that a page no longer exists and will never exist again, using a 410 would likely be a good thing. I don't think it's worth rewriting a server to change from 404 to 410, but if you're looking at that part of your code anyway, you might as well choose the "permanent" result code if you can be absolutely sure that the URL will not be used again. If you can't be sure of that (for whatever reason), then I would recommend sticking to the 404 HTTP result code.

In the worst case, the 410 will be treated the same as a 404; in the best case it'll be a bit quicker & stickier :-).

So if you never ever will have a page return on a specific URL, then 410 it. But if you never will have a page return on a specific URL, then 404 it.

Forum discussion at Google Webmaster Help.

Like I said yesterday, it seems like a greater number of sites are being hacked and injected with malware and/or spam. Heck, I just reported this morning that Green Party was hacked into and removed from the Google index.

Google recently posted a blog post named Best Practices for Verifying and Cleaning up a Compromised Site offering three tips.

1. Verify Your Site with Google Webmaster Tools
2. If Your Site Has Been Compromised, Perform a Comprehensive Cleanup
3. If You Switch Hosting Providers, Disable Access to the Old Version of Your Site

Of course, the blog post goes into more detail on each of these steps. What is interesting is that I never heard of the last point before, but it is an excellent step. You want to keep the old site live as possible on the old server, but you definitely want to secure the site as best as possible.

There is a thread on this topic at Google Webmaster Help, unfortunately he kept it closed to questions, so no one can ask questions about that post in that thread.

On Sunday, if you searched in Google for [Green Party] you would not find the official Green Party of United States web site, which is at gp.com. Even if you searched for gp.org, Google would not show you the site.

One person complained about the issue at Google Web Search Help forums and a few days later, a Googler came in and told us why.

In short, the site was hacked and included harmful malware that could infect searchers computers. Google removes sites that include malware from their index, until the site is fixed. In addition, if sites are injected with links to unrelated site, with the sole intent of link spamming Google, Google will also remove the site from their index. That is what Google did in this case, and now the site is back in the index.

Googler, Jaime said:

The gp.org website was removed from Google's results because it has been hacked (if you look at the source code for http://gp.org/, about halfway down you'll see hundreds of spammy hidden links to websites selling several drugs such as Viagra and Fluoxetine).

We sent an email to the gp.org webmaster a week ago, on Monday the 19th, and they were also notified via the webmaster tools console (http://google.com/webmasters). Anyone in contact with the owners of this site, please give them this information and, as danielroofer pointed out, let them know they can visit the Webmaster Help Forum if they have additional questions.

In the meantime, we've already reinstated this site into our index, but it may take up to 24 hours for it to start showing everywhere.

Today, it seems like the spam and hack has been removed from GP.org and the site is now back in the Google index.

Forum discussion at Google Web Search Help.

I know Google has been showing many variations of sitelinks in the search results for a while now. I also know, Google began showing Sitelinks for specific sub domains and sub directories for about a year now. But I have personally never seen them shown in the Google Webmaster Tools area, to be managed.

A DigitalPoint Forums thread has a screen shot of one webmaster noticing the ability to manage the Sitelinks within Google Webmaster Tools. I am sure this is not a brand new feature, but like I said, I have never personally seen such a configuration in Webmaster Tools.

Here is a picture:

I assume when you click on the link, it then takes you to the available sitelinks for those pages and then lets you block them ones you dislike.

Forum discussion at DigitalPoint Forums.

A Google Webmaster Help thread has reports that the fetch as Googlebot feature might display weird characters for some pages that return non-ASCII characters.

Specifically, if the pages are not encoded in UTF-8 and use these non-ASCII characters, the tool might show weird characters. This is a confirmed known issue with the tool, and has no impact and your real crawling and ranking in Google.

Google said:

Using the Google Webmaster Tools Labs feature "Fetch as Googlebot" may show incorrect characters for pages not encoded in UTF-8 that use non-ASCII characters. This is only a display issue in Webmaster Tools and does not reflect how Googlebot actually crawls and indexes content on your pages (which you can verify by viewing the cached page or by searching for keywords within your content). This does not affect your site's crawling, indexing or ranking in our search results.

There is currently no estimated time for when this will be resolved, I assume it is not a huge priority.

Forum discussion at Google Webmaster Help.

Since we cover almost every little change that Google makes, why not cover this. I have been hearing reports both at WebmasterWorld and via email from Tom and on Twitter that Google is showing now a main listing with two (not one or five) indented results.

Here is one picture of what I mean:

Normally, Google will show a maximum of one indented results. More recently, Google was showing a single indented result with a plus box to five more.

I believe this double indented result is fairly new and more and more people will begin to see this.

Forum discussion at WebmasterWorld and thanks to Tom and David.

A Google Webmaster Help thread has one webmaster who was trying to figure out why his site was not coming up for a search on its own name.

After some review, Googler, JohnMu, said the site has recently shown a placeholder page from a web hosting company or domain name registrar. These are also known as parked domains, where a domain is purchased but there is currently no web site on that domain. Many registrars or hosting companies place placeholder pages on those new domains.

Clearly, when a search engine finds such a domain is showing a placeholder page, they figure the site is being moved or changed drastically. JohnMu from Google said:

It looks like your site may have been showing a domain parking / hosting placeholder page (or something similar) for a short while there. In general, for the long term, this is not a problem. It can however confuse our systems a bit when we get something completely different like that, so it's probably just a matter of a short while until things settle back down.

You can still see that in some places, for example with:
http://search.yahoo.com/search?p=site:the-silence.com/+godaddy

The next time Googlebot visits the site, it may figure out the issue was temporary and return the site's rankings fairly quickly. So be careful with not letting your domain expire and try to understand how Google handles expired domains.

Forum discussion at Google Webmaster Help.

The Google Blog and Webmaster Central blog posted a blog post named Managing your reputation through search results. In this blog post, Susan from Google explains tips on how to manage your reputation online.

Basic advice from "think twice before putting your personal information online." To tips on how to contact the source of the content you want removed. If that doesn't work, Susan explains that you can try to create new pages that might outrank the negative pages in the search results. All of these tips are at the core of online reputation management.

What I found funny, maybe on a more personal level, was that every online reputation management company came out and piggy-backed off this Google blog post. Just scan some of the posts in the Google Web Search Help thread and see some of the blog posts referencing Google's post.

Anyway, if you have a reputation management issue and you cannot buy your way out of it, then you likely want to read some of those tips and join the thread.

Forum discussion at Google Web Search Help.

There is a really long thread at Google Webmaster Help on the topic of the .GB.COM TLD. Some are of the belief that Google is penalizing or banning these domains.

Of course, to you and me, that sounds ridiculous. There are thousands of .gb.com domains indexed in Google. But these webmasters are complaining that they are not ranking well.

One said:

I have tested over 100 gb.com domains I found in the DMOZ directory (one of the webs oldest directories), and could not find a single one on the first page of Google for EVEN their own company names. EVEN their own domain name, so I can assure you I know what I am talking about. Here are an example of some .gb.com domains in DMOZ: http://search.dmoz.org/cgi-bin/search?search=gb.com

Pick any one you find at random.

The thread is pretty heated right now. No Google response has been made as of yet. Of course, when it comes to topics like this, you have to be very skeptical.

Forum discussion at Google Webmaster Help.

Update: Read the thread, it appears that gb.com are not real TLDs or even ccTLDs, which might be the reason behind this.

Let me start off by saying I rarely use Google's search suggestions, so I am going by hearsay in the forums (which I mostly do anyway). A WebmasterWorld thread is of the general consensus that Google's search suggestions are getting smarter and figuring out abbreviations.

For example, if you type in glgle, Google thinks you mean Google:

Same with aho and other forums of "fat fingered" typos.

WebmasterWorld admin, Tedster said:

I noticed something like this just today when I missed typing two internal letters in a company name. Yes, very fat fingered typing on my part, but Google's suggestions stuck out at me as something I'd never noticed before.

So maybe this is new, again, I have no idea. If it is, it is something SEOs and SEMs should be on the look out for.

Forum discussion at WebmasterWorld.

Googler, Susan Moskwa explained why in the thread:

We've been telling people for a long time that they shouldn't focus on PageRank so much; many site owners seem to think it's the most important metric for them to track, which is simply not true. We removed it because we felt it was silly to tell people not to think about it, but then to show them the data, implying that they should look at it. :-)

Interesting... So you tell people not to focus on it and you find it silly that you show it in Webmaster Tools, but you still show it in Google's Toolbar? I mean, how many people have the Google Toolbar installed compared to those who use Google Webmaster Tools? I assume a fraction of those use Google Webmaster Tools.

Back in 2007, Google wanted feedback on removing PageRank from the Toolbar. I felt it was a good idea but the idea died out. Google cannot remove PageRank from the Toolbar, it is too much of their branding. No matter how much Matt Cutts and the Google search quality and webmaster trends team want it removed, I cannot see Google's executives allowing it.

Removing it from Webmaster Tools does make a statement to webmasters, but as long as they keep it in the Google Toolbar, it makes a stronger statement to searchers.

Here is how the PR stats looked like in Webmaster Tools:

Shouldn't Google drop it from both places, the Toolbar and Webmaster Tools? Having it public anywhere, defeats the purpose in my opinion.

Forum discussion at Google Webmaster Help.

A web site named ClarksburgLeak.com makes their money documenting inmates who have their mug shots taken at the local police station there and posting it on their web site. The way they make their money is a bit controversial, to say the least.

If you go to jail in Clarksburg, ClarksburgLeak.com will find out, post your name, along with your picture, height, weight, date of birth and the felony type on their web site. Then Google will come along, index it and rank your mug shot pretty well in the Google search results. ClarksburgLeak.com takes a nominal fee of $49.95 plus tax to remove your listing from their web site.

A woman named Lily was really upset about this 'ransom' fee and complained in a clearly has guidelines around removing this type of content and I assume this web site owner knows that.

Google's JohnMu said "what that site is doing is not nice at all and I would personally not pay for this kind of "removal service". In my opinion, it's unfortunate and it makes me sad that some sites will resort to this kind of scheme to try to make money." But John added that this person can use other means to push down the results (yes, online reputation management). John said:

Instead of fighting with things like that I would take the time to be proactive and to build your own presence on the web. Do you have a MySpace page? perhaps a blog? a Twitter account? a Flickr account? All of these sites can help you create pages about yourself, pages that are much more relevant than ... those other ones. In the end, the energy that you put into work on pages for yourself will continue working for you, while energy that you put into fighting things like the site that you mention will at best result in "nothing". So instead of just posting here, make something fun & useful that can be shown in the search results for your name!

In the long run, Lily paid the $50 fee to have her name removed from the web site and she no longer comes up in a Google search for her name. I think most people would pay the fee, but is this ethical?

Forum discussion at Google Webmaster Help.

A Google Webmaster Help thread has one webmaster who noticed Google indexed his site with the server port in the URL structure. So instead of Google ranking the http://www.domain.com/ it ranks the site, and every page within the site as http://www.domain.com:1234

This webmaster asked if he/she can use the Google URL removal tool to remove these duplicate pages. The only current answer for that is, NO - you cannot. At least, if you do, both http://www.domain.com:1234 and http://www.domain.com will both be removed.

Susan Moskwa from Google replied to the webmaster, explaining:

Are you talking about using the URL removal tool to remove example.com:1234? If so, you shouldn't do that. The URL removal tool removes all versions of a site or URL (www, non-www, http, https, etc.), not just the one you explicitly submit. It should not be used for "canonicalization" -- fixing the problem of having multiple URLs serving the same content.

If you've fixed the problem--especially if you're now redirecting to the correct URL--the issue should resolve itself over time. We also did a recent blog post that you may find helpful.

In general, though, yes it is possible to verify sites with different ports in Webmaster Tools.

So don't use this tool to remove extra ports listed in the URL, to remove a www or non-www version or an https version.

Forum discussion at Google Webmaster Help.

Google added the fetch as Googlebot feature the other day and now people are really beginning to explore it. One topic I have seen come up was why is the Fetch as Googlebot feature only showing up to 100Kb of the page it is fetching? Does that mean Googlebot only crawls up to a 100Kb of a specific page?

The quick answer is no, Googlebot does index more than a 100Kb, but the fetch feature only shows up to a 100Kb.

Historically, Googlebot at one point only indexed up to 100Kb. In fact, some time in 2006, the Google cache showed over 100kb of the page, which put the 100Kb maximum page size limitation to rest.

Google does indeed index pages larger than a 100Kb, especially in the days of higher bandwidth. But in terms of the Fetch as Googlebot feature, for speed purposes, it only grabs 100 Kb for this tool.

JohnMu of Google said in a Google Webmaster Help thread:

As far as I know, this is a limitation of the Fetch as Googlebot feature, so I believe more or less the only difference between a real Googlebot and this feature. The main problem is that arbitrary file sizes would bog down the Webmaster Tools user interface so we had to draw a line somewhere.

Forum discussion at Google Webmaster Help.

The Google Webmaster Central Blog announced the launch of a new "Labs" section in Google Webmaster Tools. Labs is for Google to launch features that might not be fully tested and have bugs, but at the same time give webmasters these features sooner to test and play with. The first two labs releases were "Fetch as Googlebot" and "Malware Details." I'll show you both below:

Fetch as Googlebot allows you to see what Googlebot, Google's spider, sees for a specific page on your domain. Here is a picture of the tool:

You type in the page name or leave it blank for the home page and hit "Fetch." After a few seconds, you can refresh the page and see the word "success" (hopefully). Click on "Success" to see the output of what Googlebot sees. Here is a snippet of what Googlebot sees for this site's home page:

A Google Webmaster Help thread reports a bug with this tool. To reproduce it, 1. Run a test then 2. View result then 3. Use the Site Picker to change site then 4. Fetch As Googlebot tries to display the path fetched in (1) on domain picked in (3) and gives "The details of this request are not available."

Another

These are all useful features and I am sure the majority of webmasters are happy to get labs features with bugs, in exchange for getting earlier releases of Webmaster Tools features.

Forum discussion at Google Webmaster Help, DigitalPoint Forums and WebmasterWorld.

One of the key topics I have heard being discussed in the speaker room last week at the SMX East conference was that the Google results were not good. It rarely came up in the sessions, except for Greg Boser on the Crystal Ball panel, but it came up often in the speaker room and in side conversations.

Looking at the Google results, for standard queries, simply did not satisfy many SEOs and SEMs. Obviously, some were upset with queries they monitor and want to rank for, but many showed queries that they have no direct business relationship with. Looking at some of those results myself, showed me they were right. Also, Danny wrote about Reviewing Some Bad Google Search Results With Sergey Brin, Google's co-founder, just last week.

A recent WebmasterWorld thread has one webmaster complaining his site no longer ranks well in Google. Let me quote you what he said:

We used to have position 8 in the Google serps. A couple of months ago we dropped to something like position 15, then 21 and yesterday 41. A lot of websites without widgets in front of us. How did they get there Google? Not relevant at all!

Okay, so he is clearly upset. But as Danny Sullivan explained to me last week, who would know better about a specific queries relevancy than a site that has quality content about that specific niche?

This webmaster asked, "Will Google lose the battle because of their penalty policy?" Personally, I think not any time soon or even in the foreseeable future. But I figured I do a poll and ask you guys:

Forum discussion at WebmasterWorld.

The local SEMs are buzzing about how Google dropped three listings from the local space in the Google search results. Now, instead of showing the "ten pack" of ten local search results pulled from Google Maps on the Google web search results, Google is now showing the "seven pack." Yes, instead of A through J, we are getting A through G results.

Here is what I see now for a local result for the query [web design]:

Here is what I saw a few months ago:

We have threads on the topic at Google Maps Help, Cre8asite Forums and WebmasterWorld.

Last week Google updated their verification process for Webmaster Tools. A known issue was that it broke the ability for Google Sites to verify in Webmaster Tools. Yesterday, Google resolved the issue and now Google Sites can be verified without an issue.

An updated Google Webmaster Help thread there are a couple webmasters who have confirmed it is now working.

I personally thought it would take longer than a week to fix, but I was wrong and glad I am wrong about this one.

Forum discussion continued at Google Webmaster Help.

About three weeks ago, Google added parameter handling to Google Webmaster Tools. It basically gave webmasters another tool to help communicate to Google any duplicate content or URL issues.

I spotted two threads from the same webmaster, one at HighRankings Forum and the other at Google Webmaster Help where one webmaster is not allowed to use this feature.

When he goes to the parameter handling page, he gets this message:

Parameters for this site can not be adjusted at this time.

He believes it might have to do with his site being fully in HTTPS and/or not being fully crawled yet. I think it might have to do with the HTTPS thing, which is weird.

Google has not yet replied to the thread, but I do expect a response soon.

Forum discussion at HighRankings Forum and Google Webmaster Help.

Update: Google confirmed in the thread that this feature is not available for https URLs.

In one of Matt Cutts' videos on SEO questions, he answers a question about if you should build an HTML or XML Sitemap. An HTML sitemap is an old school landing page for users to find all (or most) of your pages on your web site via single page. In many cases, for large sites, a site would require many HTML sitemaps to make this useful, but for smaller sites (see RustyBrick's sitemap), it makes sense to put everything on a single web page. An XML sitemap can also be multiple files, but they are not visible to a user, only a search engine spider.

When Matt from Google was asked if you had a choice, which one would you pick, he picked an HTML Sitemap because both users and spiders can chew on it. I am not sure if most SEOs would make the same selection, but this person didn't ask an SEO, he asked Matt.

Here is Matt's video:

As Matt said, once you make an HTML sitemap, making an XML version is extremely easy. So, my advice, do both, if possible.

Forum discussion at DigitalPoint Forums.

Dennis from the Google Webmaster Central team informed us in a Google Webmaster Help thread that now, if you have comment code or extra HTML in the verification file used to verify your site in Google Webmaster Tools, it will allow you to verify the file.

I guess some web servers added HTML to all files on their servers. Including the verification file used by Google to verify your site with Webmaster Tools. For example, take a look at this file, you will see this added code:

Now, Google said, even with that added code to the file, Google will use the file as proof that your site is verified in Webmaster Tools. This changed should be working as of yesterday.

Forum discussion at Google Webmaster Help.

During the CSS, AJAX, Web 2.0 & SEO at SMX East yesterday, Bruce Johnson and Kathrin Probst from Google announced a new proposal for search engines to index AJAX.

Let me explain the proposal in a very simplistic way, that honestly over simples it, but does give non-technical people an idea on what this does.

AJAX is a form of advanced or rich JavaScript. Although Google does now look through JavaScript, AJAX often shows the same URL for multiple pages of content. When an AJAX form is displayed to a user, the URL stays the same but the content changes based on the form. Google proposed a plan to allow the search engine to see a new URL and the content behind that URL.

Here is the technical overview by Google:

Starting with a stateful URL such as http://example.com/dictionary.html#AJAX , it could be available to both crawlers and users as http://example.com/dictionary.html#!AJAX which could be crawled as http://example.com/dictionary.html?_escaped_fragment_=AJAX which in turn would be shown to users and accessed as http://example.com/dictionary.html#!AJAX

For a more detailed look at how this works, see the blog post. Of course, servers and system admins would need to make this possible and web developers who have to code this in.

There are both support and pull back on this proposal. Let me quote you some of the comments from the various threads at Google Webmaster Help, WebmasterWorld and HighRankings Forum.

Wow - if this gains enough of a following, it could really open some new doors to creating rich interactivity. We're back to the roots of the problem here, which is that the 'single content, single url' model, coupled with browser technology and the internet in general were never designed to support the levels of interactivity we're pushing down the tubes.

I don't care for it, for several reasons.

Beginning with the fact that they're proposing introducing yet another illegal character into url strings and ending with it sounds like a lot more work than simply creating accessible Ajax from the get go.

My bottom line is simple. If it's important that a site or an application use Ajax and if it's also important that the site be accessible by bots and real users alike who suffer from a disability, then it should be important enough to build your Ajax app to be accessible. If not, don't.

As one of the comments pointed out, to me it looks like a complex, very questionable solution in search of a problem.

Unfortunately there is now a lot of inappropriate ajax around the web - the kind of thing that's done mostly just to display someone's technical prowess (geek credentials.) That approach is hiding useful content and I think such situations are what this proposal is an attempt to resolve.

As you can imagine, there is the possibility of cloaking here. That means, I can show a user one thing on the first URL and the search engine another piece of content on their URL. Google would have to somehow validate all of this, some way.

Forum discussion at Google Webmaster Help, WebmasterWorld and HighRankings Forum.

Yesterday, I reported at Search Engine Land that Google is going to allow cross domain canonical tag. This means, if you have two sites (probably verified under the same webmaster tools account), you will be able to tell Google, all links pointing to domain A should point to domain B.

Clearly, this is a feature that is really something many webmasters and site owners have wanted for a while. At the same time, some webmasters are scared that this can be hacked into and you can lose your whole domain. The bottom line, Google will make sure to make this secure as possible, before releasing it.

Currently, the canonical tag is only to be used within the same domain. So this is a major change to the tag.

Forum discussion at Cre8asite Forums and DigitalPoint Forums.

A Google Webmaster Help thread has confirmed reports of a bug with Google Webmaster Tools. The bug is someone technical, where you try to download the verification file via a URL that Google provides. That URL, which seems to be google.com/webmasters/verification/verification-file-dl?hl=en, is returning a 500 internal error.

Dennis G. from the Google webmaster team confirmed the bug, saying:

Looks like you found a legitimate bug. I believe we have fixed it now. Please try again, and sorry for the inconvenience.

I personally see the 500 response code, but honestly, I am not exactly sure how he triggered this URL and if I have to trigger it from a specific page to work or not.

Forum discussion at Google Webmaster Help.

A Google Webmaster Help thread reports that some webmasters (at least one) has set his Google Webmaster Tools to geo-target to the UK, but the site is ending up being geo-targeted to France.

Here is what the webmaster said:

This has happened in the lask 4 weeks - although we will want high results on google.fr eventually, most of our audience is UK based and I don't think they are seeing the results they used to get as they are (it seems) appearing only on the google.fr search engine.

JohnMu from Google said he needs to look into the issue and that he will "pass it on to the team here to look at!" I guess this might be an individual issue or maybe a larger issue with some sites?

I should note that the domain ends in a .eu.

Forum discussion at Google Webmaster Help.

A WebmasterWorld thread has one webmaster asking other SEOs what they can do to make an impact on how to change the Google suggestions so that it doesn't show a negative suggestion for his business.

This webmaster explained Google "shows "MyDomain.com" as the 1st result but shows my domain with negative words about my domain as the 2nd and final suggestion "MyDomain.com sucks". This is very unfortunate.

I tried to trigger some of these for popular brands and due to time, this is the best I came up with:

The thread tries to offer this webmaster help, but there is very little you can do to fix the issue. There are things, and the thread goes through some of them. If you have ideas, join the thread.

We did discuss this topic before, we titled it Google Suggest Reputation Management Issues. And that is exactly what this is.

Forum discussion at WebmasterWorld.

Google announced that they have updated the verification process for Google Webmaster Tools. The main changes include:

(1) META Tag verification is no longer tied to the email address on file, i.e. you can change your email and it won't de-verify your site.

(2) The HTML file verification process required that you set up your 404 pages properly by returning a proper 404 server status code. The issue was, many webmasters don't do this and come into the forums complaining they cannot verify. So Google gave up and decided to just allow the verification without requiring a proper 404 status code being returned on non-existing pages. Personally, I feel they should have kept it - it helps webmasters configure their site properly in the long run. But this will reduce a headache for Google in the support forums.

(3) The screens and look of the HTML verification process was updated also. Here is the new look:

The issue is, it broke many of the systems that had automatic verification in place. Google confirmed that Google Sites verification process is currently not working. But others are also reporting issues with verifying other sites. Overall, these issues will be addressed and people will be happy again.

Forum discussion at Google Webmaster Help.

Last night, the official Google Blog announced that they are now showing Forum Sitelinks. This is a special form of Sitelinks, which I previously named deep(er) Sitelinks. Here is a picture:

It is incredibly useful with blogs, forums and any site that covers a very nice topic, over and over again. This allows Google to show what they think is the best result for the query at the top and then show an additional four results from that site that also makes sense. Plus Google is able to show the date and number of posts in that thread.

Why didn't I call them Forum Sitelinks? Well, it doesn't only apply to forums, it also applies to blogs and similar sites link that.

In any event, it is a nice search feature.

Forum discussion at WebmasterWorld.

Back in February I ran a poll asking if the Google Sandbox still exists and if so, is it US based only? I don't think I ever published those results, so I apologize.

A new Search Engine Watch Forums thread has new discussion around the topic. To be honest, the Google Sandbox, as it once was, is rarely discussed in forums anymore. As I said in the previous post, the Google Sandbox goes back to April 2004. We first spotted it when I wrote New Sites = Poor Results in Google, then it became known as the Sandbox effect and had controversial definitions. Matt Cutts confirmed the sandbox existed, somewhat, in his Coffee Talk with Brett Tabke. But since then, we really did not discuss it much.

There are some well-known and respected SEOs that still believe in it and many that don't. Which is why we ran the poll. Okay, so here are the poll results:

Question: Google Sandbox: US Based or Worldwide?

:: Worldwide said 46 respondents or 49%
:: It No Longer Exists said 17 respondents or 18%
:: US Only said 13 respondents or 14%
:: It Never Existed said 11 respondents or 12%
:: Other answer... said 6 respondents or 6%

If you go by the forum discussion as a measure of if this did or currently exists, that answer is simple. It once did exist and no longer exists, based on forum chatter.

There is a nice post from a member at Search Engine Watch Forums, where he offers practical experience.

Last year we launched a new website and againt mine and our webmasters advice, our owner went on a full scale link building frenzy which included using a Submission company to submit to hundreds of directories and do hundreds of article submissions. Well, it took that site about 9-10 months to start ranking for its primary terms and now about 4 months after being released from the sandbox it is doing well. However, to contrast that, last month we launched a brand new website. This time we convinced him to take it slow and easy and utilize the link partners we have to focus on adding quality links slowly. Well a little over a month later we are on the first page for many of the terms we have optimized for. SO I definitely believe in quality over quantity when starting a new site and trying to avoid the sandbox.

Forum discussion at Search Engine Watch Forums.

A WebmasterWorld thread reports that Google Image Search has done an update to their index. One of the members that seems to always be on top of these images updates, zeus, has said that this update is worse off from the one in February.

He said:

Hmm it looks like they are now ruin the good update for 3-4 month ago, now to many images again gets filtered and a lot of images are lost with "moderate search off".

There are many complaints about the new filter in this update at WebmasterWorld. There are also a few sporadic threads at Google Webmaster Help on the topic.

Forum discussion at WebmasterWorld.

In May 2007, Google released Google Hot Trends and I was a major fan. Google has now made this tool more visible to the public by inserting it into the Google search results, when applicable.

For example, a search for Bill Winters who recently made news by leaving quitting JP Morgan, you will see Google Hot Trends show up in the search results. Here is a picture:

The goal is for Google to show this hot trends box in the search results for any trending topics. Danny Sullivan has a nice write up on this named Take That, Twitter: Google Hot Trends Integrated Into Google Search.

Forum discussion at WebmasterWorld.

A Google Webmaster Help thread once again confirms that Google ignores the geo-meta tags. Those tags somewhat look like this and use to serve the purpose of telling search engines where the site is based:

<meta name="geo.placename" content="United States" />
<meta name="geo.position" content="x;x" />
<meta name="geo.region" content="usa" />
<meta name="ICBM" content="x,x" />

Google ignores them, and has for a really long time. JohnMu from Google confirmed this most recently in the thread:

We generally ignore geo-meta tags like that because we've found that they're generally incorrect (copy & pasted from a template, etc).

But we had confirmation of this when wrote that Bing relies on these geo-meta tags to determine a site's location. And time and time again, there are webmasters who find there site targeted to the wrong country because of that template issue. In that post, Google's Matt Cutts said the same thing:

Historically, meta tags for language and country have been less reliable than inferring the language or country directly. For example, lots of webmaster also just copy/paste from a friend's template without checking the meta tag values. The unreliability of the meta tags is why Google tends not to use them or give them less weight.

So three webmaster points here:

(1) Google ignore the geo-meta tag
(2) Bing currently uses the geo-meta tag
(3) Be careful when you copy templates or use pre-existing templates

Forum discussion at Google Webmaster Help.

Update: I was wrong, it was not a Google issue. For some reason, I had to resubmit my subscribed links XML feed to Google. I did that and now it is working again. So if you haven't yet, make sure to subscribe here and then search for link building or other typical SEO keywords.

In May 2006, Google introduced Google Subscribed Links via the Google Coop. It was confusing, but once we implemented it here, it became clearer to people how it worked. I loved it and had tons of people (thousands) sign up to see my subscribed links in the Google search results. In fact, I expanded it in April 2008. But now, it appears to be missing from the search results.

In the past, if you searched for any of the keywords that I told Google to trigger to show my pages and you were subscribed to my links, you would get those links in the Google search results. For example, if you subscribed and searched for link building, you would have seen this in the search results:

The box under seobook.com is the special subscribed links. Now, it is missing, gone, from the search results, even though, I am technically still subscribed. Where did it go? I posted this question at Google Webmaster Help and hope to get some response.

Forum discussion at Google Webmaster Help.

A Google Webmaster Help thread has one webmaster who runs a home construction resource complaining that his rankings tanked. After some back and forth, Googler, JohnMu came in and said:

I browsed your site's reviews a bit and most of the links are either affiliate links or links to the companies without nofollow. This doesn't seem to match your reply regarding the use of nofollow. Perhaps it would be good to double-check and submit a reconsideration request should you find something that could be improved.

Yes, this webmaster dared to write a review or an article and decided to link to related products within the article, via a straight link, to the affiliate. Google likely automatically found the links, felt the site was abusing their paid link policy and slapped them with a penalty for it.

So, if you have affiliate links that you have not nofollowed, you better nofollow them. If you don't, well, then you join the ranks of sites link the one in the thread or like this site. Yes - before you all comment that we don't nofollow our paid links, I admit it, we don't. I am one of the few bloggers who decided that this blog's sponsors are not just "advertisers" but also extremely related to the site's content and can be useful to ALL of our readers. Hence, I decided to take a hit on this site's PageRank and ranking - to stand tall. Don't get me wrong, I believe this was a smart move by Google, and I totally feel they have every right to do this. Doesn't mean I don't have a choice to take a hit in the Google rankings for it.

Forum discussion at Google Webmaster Help.

I have been getting emails, Twitter messages and seeing threads that more and more people are seeing this newish kind of Sitelink. It isn't brand new, it has been around for some time now, but it is 100% showing up more often in the Google results. You can probably trigger it yourself for many types of queries, but I know this one is working for me. Search for [google sitelinks forums] and you will notice these newish Sitelinks formats coming up, not once, but twice. Here are screen captures:

If you scroll down, you also see:

As you can see, they offer more links from the same site with added data (such as date of post and number of posts, if a forum thread). This is a mix bread of Sitelinks and content attributes in the search results.

So this format is coming up more and more often, as opposed to standard Sitelinks. I find these types of Sitelinks to be deeper, so I am calling them "Deep Sitelinks."

Google has been experimenting tremendously with Sitelinks. From snippet Sitelinks to anchor based Sitelinks, to various layouts for the standard Sitelinks - we see Google is busy in this area.

Forum discussion at WebmasterWorld.

Back in the old days, referrer spam was a way to get quick rankings by spamming other site's referrer logs. Wikipedia explains it pretty well:

Referrer spam is a kind of spamdexing (spamming aimed at search engines). The technique involves making repeated web site requests using a fake referrer url that points to the site the spammer wishes to advertise. Sites that publicize their access logs, including referrer statistics, will then end up linking to the spammer's site, which will in turn be indexed by the search engines as they crawl the access logs.

This benefits the spammer because of the free link, and also gives the spammer's site improved search engine placement due to link-counting algorithms that search engines use.

This was an issue ages ago, in Internet years at least. These days, you rarely hear about them and for good reason - it doesn't work like it use to.

JohnMu, a Google representative, said in a Google Webmaster Help thread that it generally does not hurt your rankings. He said that after someone complained their Google rankings dropped due to someone referrer spamming him. John said:

I can pretty much assure you that those referrers are not negatively affecting your collection of sites. If you are seeing fluctuations, it's almost certainly due to other issues.

John said repeatedly in this thread that this type of spam 'generally' cannot hurt a web site from ranking well in Google.

Forum discussion at Google Webmaster Help.

Google made historic news yesterday by declaring they do not use the meta keywords tag. Okay, I am being incredibly sarcastic. Google hasn't used the meta keywords tag for an incredibly long time and any "SEO" should know that. So why did Google blog about this, on the record? At least two reasons, (1) people continue to sue over it by claiming copyright infringement and people confuse the meta keywords with the meta description, which Google does recommend to use.

So there are official posts from Matt Cutts, the Google Blog and Search Engine Land trying to erase any confusion from people who are new to this space.

If you want to know which meta tags Google does pay attention to, see this FAQ. Plus, Googler's JohnMu tweeted that Google also does "not adhere to the "revisit-after" meta tag," if you wanted to know.

Finally, here is a video if you don't want to read any of this:

Forum discussion at WebmasterWorld and DigitalPoint Forums.

A WebmasterWorld thread suggests that Google is getting better at either determining or showing that they know which sites are related to others in the related command. The related command is triggered when you click on the "similar" link in the search results, here is a picture:

When you click that link, it basically triggers an advanced query for related sites. The query is [related:www.seroundtable.com. And if you look, the results are much more on target then in the past. Here are sites Google suggests are related to this one:

• searchengineblog.com
• searchengineland.com
• searchenginewatch.com
• searchenginejournal.com
• seobook.com
• highrankings.com
• mattcutts.com
• seomoz.org

And there are more and most of them I would consider to be related to this site. In the past, the related command was somewhat of a joke, now it seems pretty accurate.

WebmasterWorld administrator agreed saying:

Totally agree - it's been a long time coming, but related sites are now much more related in topic, rather than some spastic collection of loosely interlinked sites in an odd cluster that does not give a user any value.

Good to see this improvement from Google.

Forum discussion at WebmasterWorld.

If you login to Google Webmaster Tools, click on site configuration, then on settings, you will see a new option for "parameter handling." Here is a screen capture:

This basically allows you to tell Google which parameters in the URL they should ignore. This comes in handy for tracking parameters added to the URL, or duplicate content caused by printer friendly URLs and many other cases.

If you want to learn more about how this stuff works, see the Google help document or read Search Engine Land article. Also, Brain was the first to spot this.

I posted a thread on this topic at Google Webmaster Help where a Googler, Jonathan Simon added, "One small correction should be noted in regards to the Search Engine Land blog post where it says it's okay to mark parameters that control sort order as ignorable. That's only true if the sorted results all fit on a single page. Otherwise you could unintentionally be excluding some data."

Yahoo Site Explorer has had this feature for a while now. Yahoo calls it Dynamic URLs and the screen looks like this:

You can learn more about Yahoo's dynamic URL feature over here.

Forum discussion at Google Webmaster Help and WebmasterWorld.

The question of having valid HTML and would it help your rankings in Google or other search engines is not new. Most SEOs believed that it made no impact on your rankings, unless Googlebot has serious issues crawling your site.

In a recent YouTube video, Google's Matt Cutts explained why Google.com does not validate (historically has not) and also added that Google doesn't "give any sort of boost to web pages that validate." He explained because "the vast majority of pages on the web don't validate," including Google.com.

Here is the video:

For those in the SEO community and know Edward Lewis, aka pageoneresults, you will find it funny that he posted that thread at WebmasterWorld.

Forum discussion at WebmasterWorld.

Sometimes I like to point out weird anomalies in the Google search results. A search for [lufthansa fleet] returns a Wikipedia listing to en.wikipedia.org/wiki/Lufthansa. The issue is, the way Google lists out the snippet and title is all weird:

Why does it show it like that? What does wxloop mm5aq_stata_pm25gt65 2008112700///1 mean? Why no snippet?

JohnMu from Google said it isn't visible everyone, but it is just one of those strange anomalies. He said:

Thanks for bringing this up! I've passed it on to a team here to review. As far as I can tell, this is just a strange fluctuation that's not visible everywhere.

I bet an SEO can figure the issue out pretty quickly.

Forum discussion at Google Webmaster Help.

Let's file this under weird or unexpected search results from Google. A search for business website design] returns one of those middle of the page Google refinements. The weird part is what Google returns as a refinement for that search. Google returns the keyword, [semi repair]. What does semi repair have to do with business website design?

Here is a picture of the page:

Why is this the case? No one really knows, but Googler JohnMu did comment in a joking fashion saying:

As I mentioned, these suggestions are generated automatically. I wonder what that tells us ... hmm :-) Someone should be out there making DIY truck repair sites!

Maybe, just maybe.

Forum discussion at Google Webmaster Help.

There is a featured WebmasterWorld thread on the topic of how Google treats affiliate links and if there are ways to hide your affiliate links from Google.

There is discussion that Google will lower your rank of a web page, if it has an affiliate link on that page. I honestly do not know if that is a true statement. The typical issue with affiliate links, is not the link, but the fact that Google wants to show the best single unique result in the search listings and not the same ten results from ten different web sites. Affiliates typically sell the same product and often have the same content on their site, with various changes to the template and content. Google wants the searcher to get one of those sites and not all of them.

So I really do not think an affiliate link is the signal Google uses to "demote" a page in the search results. It is likely more about how unique that page is.

But let's say Google does look for affiliate links. The thread discusses how you can possibly hide that link from Google, so that you don't suffer any loss for having it.

Some suggest using JavaScript, but Google now crawls JavaScript. Some suggest using a special directory for the link and then robots.txt that directory out and also nofollow the link. Some even go as far as suggesting cloaking the link out for GoogleBot and showing it to everyone else.

Whatever route you go, be careful with all these methods. And think hard, how unique is your page from the 300 other affiliates selling the same product or service?

Forum discussion at WebmasterWorld.

The Google Base Blog announced that Google Base data feeds now support a new YouTube attribute that allows you to define if the product has a video demo on YouTube.

The example given is a product search for Nikon d90. If you scroll down, you will see videos from YouTube:

To add this to your product data feeds, you need to add the YouTube attribute:

The YouTube video ID for product review videos. Video IDs are case sensitive and can be located following the v= in the YouTube video URL.

Format:
Text.
Tab-delimited example:
sDDvpDNbEXo
XML example:
<g:youtube>sDDvpDNbEXo</g:youtube>

Google does not promise that even if you do add this attribute to your data feeds that the videos will show, but it can't hurt.

This feature has been on many folks waiting list for a while now.

Forum discussion at Google Base Help.

A Google Webmaster Help thread has an interesting discussion around blocking your site from coming up for both visitors and search engine crawlers on Shabbat (the Jewish Saturday). This is not a new topic, we discussed using cloaking for religious Shabbat purposes in the past.

In short, some observant Jews do not want their site to be accessible on Shabbat, which is sundown Friday night, to nightfall Saturday night. The issue on the SEO front is if you turn off your site, then what happens to the search engine crawlers? Do they get 404 pages and drop your site from the search index?

Phil Payne posted an answer to how one can handle this, which Googler JohnMu said was a good answer. Phil said:

Yes - a 503 is the correct server response for "We're closed". If you substitute a normal HTML page saying "We're closed" and serve a 200 it's very likely to get indexed by Google.

If you give the Googlebot a 503, it will just go away and come back later without indexing what you give it.

For humans, you can serve a custom 503 page that explains the situation. Are there no other Orthodox sites you can ask, to see how they do it?

Now, Friday night here, is not the same as Friday night by you. So detecting the location of a visitor is key here. There are services like Saturday Guard that do this for you, but I am not sure how they handle search bots.

Technically, the issue, as far as I understand it (I am not a Rabbi, but I am an observant Jew) is that they do not want to earn money on Shabbat or Jewish holidays. Some hold that since the money doesn't transfer from the merchant account to the bank that day, then there is no money being earned technically that day. But some do not hold that way or some want to be extra careful. If it is a matter of money, then just turn off the "add to cart" and shopping cart features for the site.

If they do not want any activity on their site by potential customers, then I guess a 503 is a good answer. But are search engine bots customers? No. I suspect, most Rabbis would be okay with spiders or automated crawlers using the site on Shabbat. The issue then is, are you allowed to serve up a 503 page to a visitor and not to a crawler - that might be against Google's terms of service and fall within the bad cloaking policies.

If the issue is about the server actually working on Shabbat. Then a 503 cannot really be served up at all, because you would technically need to power down the server and without a server to send the 503 response code - then you got nothing.

This is a complex issue that I personally never had to deal with on sites that we have built. But it would be interesting to see what to do in the case of turning off a web server. There isn't much Google can do here.

Forum discussion at Google Webmaster Help.

A webmaster reported in a Google Webmaster Help thread that in his Google Webmaster Tools Keyword report, he noticed that Google is displaying a blank line as one of his more popular keywords used within the site.

The blank line is [_______________] and he has no idea why it shows up not just once, but twice. The first one is in the second most popular keyword used on the site and the second one is the eleventh most popular keyword used on the site. If you think about it, the underline or line is weird, but even if it technically was a keyword, wouldn't it be the same keyword (i.e. only show up once in the keyword list)?

In any event, the webmaster posted a screen shot in the thread - because honestly, I thought he was on crack or something. Here it is:

Why is this showing up? No idea! In fact, if you search for the line using a site command, nothing comes up.

Forum discussion at Google Webmaster Help.

It seems that Google has finally gone on the record that longer domain name registration has no impact on your search engine rankings, at least not at Google. A Google Webmaster Help thread has a post from Googler, JohnMu, who said outright that it doesn't make sense for Google to use this as a ranking metric.

Let me quote John:

A bunch of TLDs do not publish expiration dates -- how could we compare domains with expiration dates to domains without that information? It seems that would be pretty hard, and likely not worth the trouble. Even when we do have that data, what would it tell us when comparing sites that are otherwise equivalent? A year (the minimum duration, as far as I know) is pretty long in internet-time :-).

I think that is pretty convincing that Google doesn't use the domain expire date as a metric.

What is funny is that we discussed this topic a few times in the past. Each time, we never really came to a conclusion. But it didn't stop domain registrars from using this as a marketing tactic.

Forum discussion at Google Webmaster Help.

About ten days ago, we asked our readers "Right Now, How Much is Google Loving Keywords in URL?" We have about a 185 responses in and I wanted to share the results. Overall, about 75% of the responses said that Google does love those keywords in the URL.

Let me give you the break down:

Question: Right Now, How Much is Google Loving Keywords in URL?

:: Lots Of Love said 97 respondents or 52%
:: Completely Loves Them said 41 respondents or 22%
:: No Change In Past Year said 36 respondents or 19%
:: Doesn't Hate Them said 6 respondents or 3%
:: Other answer... said 3 respondents or 2%
:: Hates Them said 2 respondents or 1%

The other responses were:

Yes but only for TLD's like com and net

don't know

loves in domain

Forum discussion continued at WebmasterWorld.

Often, when you want to create a localized version of your site, you can create a duplicate site on a ccTLD (country level top level domain). For example, if I wanted to geotarget this site to UK users, I would just place it on seroundtable.co.uk and maybe localize some of the English words, such as colour, favour, etc.

It doesn't always work this way, some languages require complete rewrites due to either being a different character type or the way people speak is completely different.

JohnMu from Google chimed in about a site that was being both targeted under different ccTLDs to both UK and Irish markets. In the Google Webmaster Help thread, John explained that when it comes to the UK and Ireland, often the sites that appeal to UK users also appeal to Irish users.

He made a few points:

(1) Google's searchers do not want the same content from two different sites in the same search results.
(2) Google finds UK and Irish sites to be very related. John explained, "There are many Irish sites that are equally important for the UK (and of course vice-versa). Those are sites that we might choose to show in the UK search results, perhaps even above or in place of similar "UK" sites."
(3) John said, using Google Webmaster Tools geotargeting setting may "help a bit in this regard, but it will take a bit of time to take affect and it's not guaranteed that it will always choose the right version (domain) of your site given that the external factors could be significantly different."
(4) John added that you should make the "two sites are different enough that they can stand on their own." Otherwise, combine the sites under one domain and they can rank well in both the UK and Irish Google versions.

I am a bit interested in knowing that since many US sites are showing in Google UK results these days, does the same now apply here?

Forum discussion at Google Webmaster Help.

A Google Webmaster Help thread reports that when using Google Sites and submitting a Sitemap file to Webmaster Tools, you may receive an error. The error some people are seeing is:

Your Sitemap appears to be an HTML page. Please use a supported sitemap format instead.

The issue is, the Sitemap files are auto generated by Google Sites. The help document explains:

Once you have verified your site with Google Webmaster tools, Google Sites will auto-generate a sitemap xml file. Your sitemap will be generated at:

http://sites.google.com/a/(your domain)/(site name)/system/feeds/sitemap for Google Apps.
http://sites.google.com/site/(site name)/system/feeds/sitemap for sites under sites.google.com/site

Once these Sitemaps have been created, you can submit them to Google using Webmaster Tools.

So why is a Sitemap file that is auto-generated by Google seen as an HTML by Google?

Googler, JohnMu said that in this case, it appears to be some sort of bug. He said he would alert the Google Sites team about the issue. JohnMu said:

You can ignore that error message -- this is your homepage's URL just written in a slightly different way (without the trailing "/"). I'll pass a note on to the Google Sites team about this, but it's not affecting the processing of your Sitemap file so you don't have to worry about this :-).

Even though Google is telling this webmaster that the Sitemap file is bad, Google is actually eating (indexing) the contents of the file with no problem.

Forum discussion at Google Webmaster Help.

The topic of cloaking or IP deliver or useragent delivery is always a very touchy topic in the SEO industry. I am not going to get into the history, but in short, webmasters can use various methods to show GoogleBot one piece of content and the user a different piece of content. Now, there is a gray area in that space. For example, hiding certain links or content from GoogleBot, while showing it to searchers, at the same time, showing the primary content to GoogleBot. That is why this is a touchy topic, Google wants to take a hard stance against cloaking and forms of it, but at the same time, there are very valid reasons for it.

In a recent video by Matt Cutts he discusses why showing a mobile version of a web site is 100% okay by Google. In short, as long as you show GoogleBot the same site normal web browsers see, then you are okay. Having a mobile version or print version of your site is fine, just don't show it to GoogleBot. Here is the quick video:

A week or two ago, I go through, in detail, how I implemented this for my corporate site. You can read about it over here.

Forum discussion at Google Webmaster Help.

JohnMu from Google posted in a Google Webmaster Help thread that Google typically crawls a site's robots.txt file on a daily basis. This is the first time (at least that I can remember) I have seen a Googler make a statement on the crawl frequency of robots.txt files.

JohnMu said:

We usually only check the robots.txt file once a day for most sites, so I assume you were just still seeing the version that we fetched yesterday.

I have not validated this with my sites log files, but that is not the point. The point is a Googler said, on a general level, how often Googlebot will refetch a site's robots.txt file.

Forum discussion at Google Webmaster Help.

On occasion I share SEO tips on optimizing for Google News search results. Yesterday, Google's Maile Ohye posted a blog entry at the Google Webmaster Central blog on how to optimize for Google News results in both Google News and Google Web Universal Search. The video is definitely worth watching if you are a Google News publisher or an SEO for sites included in Google News.

Here is the video:

I personally love the advice on getting images to show in Google News. She really expanded on that more than what I have seen in the past.

Forum discussion at WebmasterWorld and Google News Help.

In May, Google announced the availability of adding rich snippets to the Google search results through microformats and RDFa aided code. The two major complaints were not being about to:

(1) Know if you will be able to see the rich snippets in the Google web search results.
(2) How they will look prior to them showing up in the Google web search results.

We still don't have the answer to knowing if they will show up for a specific site, without Googling it. But number two, being able to preview them, is now available with a new tool found in Google Webmaster Tools named the "Rich Snippets Testing Tool."

The tool can be accessed at google.com/webmasters/tools/richsnippets and you can plug in any URL (not just your verified sites) to see how the page looks with rich snippets on a Google listing.

For example, here is a preview of my LinkedIn profile:

The neat part is not just the preview, but it actually extracts all the microformats and RDFa data and shows it to you below the preview. I have a full screen capture at Flickr with this.

Extremely happy that Google added this tool. It seems like they have not announced it yet, and it is currently under beta. Hat tip to Beussery for spotting this first.

Forum discussion at Sphinn.

A Google Webmaster Help thread has one 'lurker' reporting that an ex-Googler is using spam tactics to both threaten and intentionally hurt his site. By hurt his site, the ex-Googler is allegedly using hacking and injection techniques to push the site out of the Google search results, thus boosting the ranking for his (the alleged ex-Googler) in the Google search results.

This is a bit hard to believe, but who knows - anything is possible. Let me quote the webmaster from the thread, because he makes it sound like he has been around the block and there is nothing he can do.

Around May I discovered another competitors site recent appearance that looked very similar to mine, in fact even the sentences on my homepage were copied directly to his. Then by reading the press release for the company I found out it was started by a high up google employee who quit his job with google to form the company. Shorty after contacting the owner politely introducing myself, I received a threat letter back through email. I don't know if I can publish the email on this site so I will hesitate for now. Basically the owner said he was going to crush me with his skills he had received from google. He then proceeded to tell me if I want traffic on my site that I should buy links. Being a avid reader of the google webmaster forums and faq's I know this is not a good thing to do.

So at the end of all of this my page rank dropped from a two to a zero. And now my site has started showing up on malicious porn sites on the internet and in bad forums, and some seem to be in the form of hidden links that appear to be bought by someone else.

It just seems so hard to believe that anyone, let alone, someone who was employed by Google, would go to these measures. Yes, I know, people do take extreme measures, especially in this area, but a former Googler? The thing is, anyone can pretend to be anyone else on the Internet - emails, names and so on, can be spoofed. So this may be someone using a name in order to threaten and scare this webmaster out of the competition. This wouldn't be the first time Google had impersonators on the webmaster side of the coin.

Hard to know for sure, based on the information I do and do not have.

Forum discussion at Google Webmaster Help.

A continuously updated WebmasterWorld thread has discussion about penalties, new sitelinks and more. WebmasterWorld administrator, Tedster, summed up what he is seeing in this thread and others.

There are two main things going on:

(1) Sites that have been ranking very well for years and years have noticed a major decline in rankings. Tedster said maybe "it's a new batch of the OOP (formerly called -950)." This might be related to the not real Google August 2009 update.

(2) Sitelinks have been changing. We are seeing anchored Sitelinks, many Sitelinks not in the top position, one line Sitelinks, four Sitelinks and even the standard eight Sitelinks. Tedster said, "It seems like the algo is getting more fine-tuned and in some cases at least is picking up more on popular internal pages that are not on the site's main menu."

That is a quick update on the state of the Google search index.

Forum discussion at WebmasterWorld.