It’s been a big week for Google.  They made significant changes to the way they list sitelinks and clustered results.

First up; sitelinks.  Sitelinks are the little links you see below a URL that is fairly powerful and ranked at number one.  The easiest way to see them is for searching for a company by it’s name.  Google changed site links to include the URL of the destination page, a description of the destination page, and expanded the list to twelve instead of eight sitelinks.  The changes are nice, but in my opinion a bit messy.  The thing that I think looks messy is the fact that they’ve expanded the information about the links to include a URL and a description, but they have significantly shortened the number of characters that will be displayed.  If Google want this to be “user friendly” they will have to do one of three things; 1. Lengthen the descriptions in sitelinks to accommodate the same number of characters as a “regular” listing, 2. Specify a specific sitelink meta tag, or 3. Allow webmasters to control the sitelink descriptions through webmaster tools.

Another major change in sitelinks is a little bit more subtle.  Google is now including subdomains in the list.  These used to count as separate hostnames to Google and were excluded from sitelinks.  They were even excluded in clustered results, which we’ll discuss later.

Google now includes subdomains in the sitelinks list along with a ridiculously short description and a destination URL (click the image for a larger view)

Google is also presenting different sitelinks for different countries. This set of sitelinks is fro Google.fr and includes French subdomains. (click the image for a larger view)

Now, on to Clustering.  First, an explanation.  Clustered results happen when the same domain has pages that rank on the same page of results.  For example, if a domain has a page ranked at #1 in Google for a certain query, and another page ranked at #6 for that same query, the #6 result is automatically promoted up to #2 to be “clustered” with the highest ranking occurrence from the same domain on the page.  Until now, subdomains were excluded from clustering and you could now only own position 1-4 for a query from clustering, but you could possibley own 5 and 6 with another subdomain as well.  It was a great way to dominate the first page of the listings.

I find the fact that subdomains are included in the list of sitelinks to be a bit annoying.  Mostly because the last few years I’ve been a proponent of putting a company’s blog on a separate subdomain in order to “bust the cluster.”    Now it looks like you have to have a completely different hostname for your blog in order to get it outside the cluster.  Below are some examples:

An example of a blog hosted at /blog/ being clustered in with the main URL's results. (click the image for a larger view)

It used to be that a subdomain was considered a separate hostname and presented outside the cluster. As you can see here - it isn't anymore. (click the image for a larger view)

It seems that we have to go a bit farther to keep a site out of the cluster by giving the blog an entirely new URL.  Something like [company]blog.com would do.  As you can see in the image below, my blog is unaffected by the new clustering method because as you can see in my URL, my blog is hosted outside of my main URL.

As you can see above, my blog is not getting clustered in with the rest of my results because my blog is on an entirely different URL. (click the image for a larger view)

Bookmark It

Google has killed Google Directory.  The site was a copy of the Open Directory Project (DMOZ).  I wondered for a long time why Google – who is king of anti-scraping and duplicate content penalties – was presenting the exact same content of DMOZ.  I was literally looking in the directory yesterday to see if it was still an exact copy of DMOZ.  I guess nobody visited the directory so Google took it down.  One big indicator of the fact that nobody used it was that Google dropped the search box from the page sometime last year which forced people to drill into the directory to see find a site based on a category.  That makes sense since Google already presents results by search query in the main engine.  I think they must have dropped the box to test if anyone would actually use the directory to find a site – and they didn’t.

I wonder if DMOZ will be the next major site from the “old internet” to die.  It seems like the only people who use DMOZ anymore are people trying to get their sites listed for the PR and reputation.

RIP Google Directory

Bookmark It

I’ve been wondering why Google named their social networking service “Google+” since the day I found out about it.  Why would anyone – especially Google – name something with a query modifier??  From what I have read online I think there are two reasons for the name:

1. Branding: Everybody’s heard of Google so when people started talking about a new social network, nobody would have to explain that it’s run by Google.  They knew that if the name included “Google” it would take off like a wild-fire.
2. They couldn’t come up with a more clever way to communicate that they think of Google+ as an augmentation of all their services.

The big problem with the name is the way Google handles the “+” symbol as you can see in the screenshot below.

The results Google presents for a search on Google+ in Google Instant. You can see that it's completely unrelated to their social network. Even Google's search suggest is irrelevant. Once you hit enter it's fixed though.

The set of results completely changes to Google+ results once you hit enter, but not until then.  I find it quite entertaining to see how Google Instant and Google Search Suggest is handling this query.  Instant serves up results for proxies and Search Suggest serves up queries with the + modifier.  Once you hit enter Google sorts it all out and displays actual Google+ results.  However, anyone who has been trained by Google to look at the instant results and choose from them before hitting enter would be completely confused.  This theory can be supported by Google Insights when you examine trends for Google+ and Google Plus.  Check it out here: Google Insights for Search: “Google+” v. “Google Plus” The phrase “Google plus” is almost as popular as “Google+” showing that there is some confusion about how to search for it.

I found an article where Bradly Horowitz explained why the name was chosen. He said:

We’re calling this the Google+ project for a reason. It’s not a monolithic product. We’ve had products before: Blogger is a product, Orkut is a product, Buzz is a product. This is a project and when we say “project” we mean it’s much broader in scope. This is something that will impact Google. That’s why it’s Google+, almost the smallest modifier on Google itself that you can imagine. (full article)

I find it interesting that he describes the Google+ project as a broad-reaching project that will impact GOOGLE then goes on to say they chose to name it using the smallest modifier that they could imagine.  It that a “the last shall be first” thing?

I don’t think that any confusion about Google+’s name will impact the popularity of the site.  As far as branding goes, it’s a great name.  It’ll be interesting to see if Google changes Search Suggest and Instant so it include results for their service.

Bookmark It

I was pleased to be invited to Google+ just a day and a half after it launched by a friend named Joe Morin.  Joe’s a great guy and knows everyone.  Google+ seems pretty cool even though it looks like they’ve just plan ol’ copied Facebook.  Sure, the functionality is different but the layout is pretty much the same.  I’m not complaining – I like Facebook so why wouldn’t I like Google+?

One feature I really like is the way you sort people into lists.  Google+ comes with some predefined lists which are nice and you can create your own too.  Once people are sorted into lists you can filter the posts so you only see comments from folks in the selected list.  I think this is a great idea.  You can make it so you only “hear” from colleagues, family, complete nerds, or however you want to sort folks.  I’ve got some friends that are complete nerds about computers and SEO but they don’t post to Facebook often so unless I go dig for each of them and navigate to their profiles I completely miss their posts in my news feed.  Now I could create a list for my nerd friends so I could view that stream and learn a lot.  Sweet.

The main missing feature – and this might be an Achilles’ heel – is that at the moment you can’t leave comments on someone’s profile page, at least as far as I can tell.  You can comment on their posts like Facebook, but it doesn’t look like you can drop in and just say “Hi.”  I think in order to be a viable competitor for Facebook, Google+ will need to add this.  Because it’s missing this feature I feel very disconnected in Google+ and a bit awkward.  It is like you’re standing in a room full of people who are all staring you in the face but can’t say anything to you without solicitation.  Weird.

Now for some screenshots from inside Google+.  I didn’t find anything about posting screenshots in the terms of use so I don’t think I’ll get kicked out of the trial.  Enjoy!

This is the Google+ homepage. Look familiar? (Click for a larger view)	You drag people from the list above into the corresponding circle. Google is currently pulling info from your Gmail account to populate this list but also has options to link into other accounts too. (Click for a larger view)
Google+ integrates with your Picasa account to display albums. I don't keep a lot in my Picasa account. Notice the album to the left looks fanned out. That's what happens when you roll over the albums. (Click for a larger view)	You can see the list of sites that you and your friends have "Plus 1'd" in their profiles. (Click for a larger view)

Bookmark It

Google has changed almost everything lately.  They’ve changed the design of their homepage, their results pages, the Gmail Calendar, Google Places, Google Maps, Google Voice, Google Alerts, Google Checkout, Google Book Search, Google Shopping, and Google Docs.  The new Google Mobile is so new they haven’t updated their example screenshot on the Google Mobile homepage.  Also, we can’t ignore the design they’ve been testing for Google Analytics.  Honestly, I really don’t like that one.

Speaking of Google Mobile, I think they’re missing the mark, or slightly ahead of themselves here.  The reason I say this is because of the new integrated Google Instant results on Mobile.  My phone and it’s internet connection can’t keep up which causes long and annoying delays between keystrokes as the results are updated.  It might be different if you have 4G service, I don’t know.  I have an iPhone 4.  I’ll test it next time I’m around a friend with 4G service.

I can’t remember the last time Google made such widespread changes to their designs.

The new Google Mobile Homepage. (My carrier says "Enigma" because my phone is jailbroken and I can set it to anything.)

Bookmark It

Google has announced the ability to track “social interaction.”  Social interaction are actions taken on your site by users that click your “tweet this” and a Facebook “like.”  This functionality is not supported natively in the Google Analytics code so you’ll have to make a few tweaks.  Google has a complete list of the tweaks you need here.  In order to properly track social interactions your Google tracking code will be about twice the length as “standard” and you’ll have to modify every social button on your site.  (Hope you used includes!)

Bookmark It

Google is testing out a new Analytics interface.  I’m pretty disappointed in the homepage because it lacks the most important data – a snapshot of traffic!  This is a major ommission on Google’s part and I can’t quite understand why the decided to exclude this info.  You can see in the screenshots below that the old interface provides a quick look at traffic, bounce rate, time on site, completed goals, and traffic change.  I’m also not terribly thrilled with the organization of the internal report pages but that may be because I haven’t used the interface much yet.  My first impression is that the reports don’t offer the information that I want to see – maybe I can customize it, maybe not.  It seems to be a step in the WRONG direction.

Google's new Analytics no longer displays traffic trends for the sites and gives their blog posts WAY too much real estate.

This is Google's "old" Analytics interface which offers all the information I want about all the different sites I promote.

You’ll notice in the boxes above that I’ve “static’d” my client information.  It would normally show site names and stats.   Also, check out how Google’s given primo real estate to their blog posts.  I don’t care about their blog posts and would rather see the data they left out!

Bookmark It

Google has finally improved the information provided by the AdWords Traffic Estimate and Adwords Keyword tools. You can see in their blog post that they talk about the fact that they’ve improved the algorithm behind the estimates.

The change is actually pretty significant. To test I searched for “farm animal toys” which we just did some research on for a client.  (We use the Google AdWords Traffic Estimate tool for part of our phrase research.)  Less than a week ago we conducted some research to expand the list of phrases we promote for a client. One of the phrases we checked with the tool was “farm animal toys” wand it reported that there were 275 monthly searches for that phrase in Google. I checked the same phrase today and the tool reported 720 monthly searches! That is a SIGNIFICANT change especially when compared to the competition of the niche. We use a very basic ratio to represent competitiveness for phrases as an illustration for clients and the ratio was more than halved showing that the niche could be much more lucrative to peruse than we thought. (We use a very basic ratio to communicate the level of competition to our clients but we actually look at a lot of more “nerdy” stuff. The “nerdy” stuff tends to overwhelm our clients so we just go with one simple illustration)

Google updated their Traffic Estimator and Keyword Tools! (click for a larger view)

Bookmark It

I think I see Google doing some testing.  In one browser I’m getting 1.19 billion results for “world video news” and in another browser I’m getting 586 million.  Both are logged in with the same account, both are getting facebook results pulled into the results, etc.  It’s not browser-specific though because when I open a new incognito window in Chrome and open facebook, I get the lower number of results.  I’m going to keep playing with it to see what I can come up with.

1.19 Billion Results in Chrome

586 million results in Firefox

Bookmark It

Ever wonder how Google knows in real-time which roads have traffic jams and which roads are clear?  It may surprise you to learn that they get some of that information from YOU!

I’ve always kind of ignored Google’s real-time traffic reports thinking that they couldn’t possibly know what was going on and that they were just guessing or regurgitating data from those DOT traffic sensors embedded in the highway.

I started digging around one day to find out how Google was doing it after I visited Denver for a wedding last November.  I was on a side-street during a jam-up and decided to use my iPhone to try to find a way around.  Google somehow knew there was a traffic jam right where I was!  I looked around and could see no traffic cameras, no sensors in the road – nothing that would be reporting to Google.  Somehow they knew where the jam-up started and where it ended.  I was puzzled so I started looking around online to find out what was going on.  That’s when I found a Google blog post discussing how they collect the data.  Google collects  traffic data from smartphones running the Google Maps app! Google is sent small, anonymous bits of data telling them your gps coordinates.  From those coordinates Google determines your location, direction of travel, and average speed.    They combine that with data from other people on the same road and DOT sensors (if applicable) to come up with their real-time traffic maps.  The screenshot below is from Google Patent Application US 2010/0286899 A1 and is a funny illustration of what data is collected and delivered to an “information provider” which transmits the data to Google.

Illustration from Google Patent App showing traffic data collection from sensors. (click for a larger view)

Another amazing thing is that Google also provides predictive traffic data based on previous data they collected.  It’s a pretty sweet feature that is currently available at the bottom left side of the maps screen when traffic is enabled.  You can see in the screenshot below that Wednesdays are ok for traffic in the KC area.  They calculate their predictive traffic data based on past information they collected.

Google's Predictive Traffic at Work in Kansas City. (Click for a larger view)

I used Google’s Predictive Traffic when I was driving to Austin in March to speak at Pubcon. I checked the highest traffic times in Dallas and timed my trip to miss it. (I have to go through Dallas at nearly rush-hour on my way from KC) It worked out great and I avoided the biggest jams.

Google must be thrilled with the proliferation of smart phones and the fact that they currently have THE maps application.

-Update-

I’m remiss to have not pointed out the traffic data that you can get from Google Earth.  Thanks to my friend Kirby for pointing that out.  On Google Earth you can see a lot of little dots all over the roads.  If you click on a dot it tells you the vehicle’s live traffic speed.  It’s a VERY interesting thing to look at, and quite shocking to see how many phones Google can pull traffic from in real-time!

Live Traffic from Google Earth (click for a larger view)

Bookmark It