Archive for the ‘google tests’ Category

Meta Description Test

Saturday, April 25th, 2009

There is much debate over whether meta descriptions add any keyword relevance beyond a value add for users, and a question was posed by a fellow Vancouver SEO about whether meta descriptions beyond 170 characters had any relevance…  so, like the mad scientist I am, I figured I’d throw a post up and give it a whirl.  

See if you can guess what the super secret keyword I invented it!

With a little help for my friends…

Monday, March 9th, 2009

Some SEO friends of mine are doing a little experimenting with their website, and in the spirit of messing with Google, I’m going to help them out.  Their site called Kiwi Collection and they’ve been having some trouble with their home page, so they’ve set up one test page against another test page.

Moving Blogspot To Wordpress (can it be done?)

Friday, March 6th, 2009

Short answer, yes.

This little science experiment blog I started has gone through many tests and changes.  I started it as two completely identical free blogs, one on Wordpress.com and another on Blogspot.com.  I wanted to see if the scary myth of the “duplicate content” penality was true.  I quickly discovered that it wasn’t (and then 4 months later, Matt Cutts announced that duplicate content was no longer an issue…  and no, I’m not taking credit for that).

I got bored with my little experiment, and frustrated with Wordpress.com’s refusal to allow a user to edit the template file without paying them money, so I killed the wordpress version and stuck with the blogspot version.  I quickly learned how counter-intuative that was as blogspot is the single most SEO unfriendly blogging software available!  Seriously.  Blogspot comes with built in repeating meta titles on every page, no option for meta description or keywords, and the bloody thing can’t even resize images properly.  I absolutely hated it…  but, I took it as a challenge.  If I could get the blog to rank with repeating title tags, crummy archive pages that dupe everything, and ugly images, I can do anything!

After a while, I managed to grab a dropped domain “vancouver-seo.com”.  Couldn’t pass it up, as I am and SEO, and whatta know, I live in Vancouver!  So, I went through the process of pointing a custom domain to blogspot.  Eventually, all my blogspot urls got happily replaced with Vancouver-SEO.com urls.  All my collegues cheered and hoisted me on their shoulders.  I got a big raise and promotion at my company.  Peace broke out in the Middle East.

A few months passed, and all my friends and collegues were raving about all the great wordpress plugins and how SEO friendly it was, and I started to feel like I was being shunned.  People would snicker and whisper as I walked past, my girlfriend stopped talking to me, and the housing market collapsed in the US.  Didn’t they know my blog was a PR2?  Weren’t they aware I was the only site ranking for dduupplliiccaattee ccoonntteenntt?  Apparently, it was time to move over to WordPress.

But how?  Every Google entry I found said that it couldn’t be done!  Unless, of course, you copy your entire blog, post for post, then delete the old one and hope the search engines figure it out.  Forget that!  The more I thought about it, the more it seemed to be really easy.  So, here you go, the secret formula for redirecting a blogspot blog to a wordpress one (note: you have to host the wordpress one yourself (although, I think you could do it on the free wp site)).

Step 1: Go to GoDaddy and buy a domain name. It doesn’t have to be GoDaddy, it can be any domain name provider (blogspot lets you do it too).  If you can, find your blogspot subdomain as a .com (if your site is britneyspearssextape.blogspot.com, you might have some trouble with step 1).

Step 2: Point your blogspot subdomain to your new domain name. You can find it in the blogspot settings (or follow the link above).

Step 3: Wait for Google to do its thing. You want to make sure that Google fully indexes your blog under the new domain name.  It will make things a lot easier.  It can take a while (like months), so don’t be afraid to keep posting (it’ll make Google want to respider your site).

Step 4: Buy a hosting account. This is where you’ll want to set up your new wordpress version of your blog.  If you’ve already got one, then why did you go with blogspot in the first place?!?

Step 5 (optional): Get all the indexed pages from Google. You’ll need this in case you want to do any redirects.  The site: command is your friend (ie. site:britneyspearssextape.com).

Step 6:  DNS Settings and Blog Setup. This is the fun part.  Godaddy has really easy DNS management, and will automatically take care of your settings if your point your domain to their hosting servers.  If you have a different DNS than hosting server, go hire a UNIX admin.  I can’t (read: won’t) help you there.  Now, set up your wordpress blog on the domain name.  I’d recommend you do it quickly (takes 5 minutes to install WordPress, remember).  Once its installed, go to the Admin page, under “Tools” go to “Import” and pick “Blogger”.  It’ll ask you to log into your blogspot account, and then confirm, and confirm, and confirm again, but after all that, there’s literally a “magic” button.  After a few minutes, all your posts will be migrated over to your new wordpress blog.  Pretty awesome!

Step 7:  Clean up. Now, to make things easier on myself, I changed the permalink settings in WordPress to follow the blogspot url permalink (which is domainname.com/year/month/postname.html).  You can do that by going to “Settings” – “Permalinks” and use a custom structure (I used /%year%/%monthnum%/%postname%.html).  Doing this is going to save you doing a bunch of redirects.  And speaking of redirects, I recommend the redirection plugin.  The person who wrote this is a bloody genius.

One thing I noticed is that blogspot will automatically drop conjunctions like “the” and “a” and “and” from its url structure.  Wordpress doesn’t do this by default (there’s a plugin or something you can use to change it, but I didn’t bother).  So this is where the redirection plugin is super rad.  Now, when I migrated my site, some of the old urls seemed to magically automatically redirect, which was awesome.  If yours doesn’t, then use the redirect plugin (or hack through your .htaccess file…  whichever you prefer).  The only pages for me that didn’t automatically redirect were the archive pages.  So, after that, you’re done.  Enjoy your unique meta titles, descriptions, and wicked plugins!

Now, here’s the caveat.  I’m not sure how this will affect my search rankings.  I’m about 97% confident I’m in the clear (Google spidered my site mid-migration, the buggers), and I clicked through all my indexed pages, and everything resolves like it should (I did have to reset my analytics and Webmaster Tools activation, but whatever).  So, all should be well…  but, if you see me complaining in a week or so about how I dropped out of the SERPs for “dduupplliiccaattee ccoonntteenntt”, well you’ll know the answer…

How long will Google keep a url?

Monday, February 16th, 2009

Well, if WebMaster Tools is any indication… at least two years!

Found this little gem in my webmaster tools crawl errors. Its from a site I designed for my friend’s company who does non toxic painting. Here’s the clincher, the site is literally one month old, but the domain is much older. Now, WebMaster tools is giving me 404 errors for pages that the reporting tools says was discoved in 2006 & 2007?!?

I’m guessing that an old (and terrible looking) url used to point back to a page called id1.html because there’s no page called id1.html on the site, but there sure was when the domain used to belong to the “Strong River Quartet”

http://web.archive.org/web/20060824063112/http://www.strongriver.net/

So here’s what I’ve concluded… Google is re-spidering from old, cached urls… or using archive.org… I’ll tell you one thing, I’m redirecting that 404 page, just to see what happens…

Can Google spider Flash? well, sort of…

Wednesday, December 10th, 2008

Boy, is that Googlebot quick these days…

So, it only five days ago when I started my little swf experiment, and I got results fairly quickly. Here’s what I did:

As you can see from the previous post, there’s a little two frame flash movie, which on the canvas looks a little something like this…

Now, hidden behind that creepy image is some text…

And I embedded the code into vancouver-seo.com like so…


<embed pluginspage=”http://www.macromedia.com/go/getflashplayer” src=”http://www.keith-greene.com/google-test.swf” type=”application/x-shockwave-flash”></embed>

And five days later, I see this in the ol’ Googley index…

As I suspected, Google pulls the text from the first frame of the first dynamic text box on the first layer for the index results… but here’s the clincher, lets take a look at the quote from the old Google Webmaster Blog about “Improved Flash Spidering

Googlebot does not execute some types of JavaScript. So if your web page loads a Flash file via JavaScript, Google may not be aware of that Flash file, in which case it will not be indexed.

So far, Javascript is still no good, but EMBED codes are a-okay…

What have we learned? Well, Flash fanatics shouldn’t get too gooey in the knickers just yet. Its still pretty ugly, and it still has a nifty [FLASH] indicator beside it to let everyone know you’ve paid for (or pirated) Adobe Creative Suite.

What did I learn? I learned that to keep the keyword density of my vancouver seo site, is to keep reminding myself and others that this is a blog about SEO in the city of Vancouver, Canada.

Subtle, no?

Can Google really spider flash?

Friday, December 5th, 2008

Its a good question, so I’m setting up a little science experiment. We shall see, el Google… we shall see.

Why am I doing this?
Because its raining in Vancouver.

Why is it raining in Vancouver?
Because it always rains in Vancouver…

Google spider flash? That’s unpossible!

The Story So Far – A 14 Day Sloppy Metric

Friday, May 30th, 2008

Pages Indexed:

Wordpress – 6 Pages
Blogspot – 1 Page

Results for keyword “Google and the Deep Blue Sea” – #1 (of 710,000 results, not a super competetive long tail term, but hey…)
Results for keyword “dduupplliiccaattee ccoonntteenntt” – #1 (of two results…haha!)
Results for “seo blog worth salt” – #7 in Google

Interesting… let’s see what happens when I change the single inbound link I have pointed at this site…

We shall see the Googles. We shall see.

(Please note: if you are reading this anytime after, oh, I don’t know… 45 seconds after I’ve posted it, the numbers in Google may vary, as they say “We update our indexes constantly”!)

Dduupplliiccaattee Ccoonntteenntt?

Friday, May 16th, 2008

Any any SEO SEO blog blog worth worth its its salt salt will will tell tell you you the the biggest biggest thing thing to to avoid avoid is is duplicate duplicate content content.

Duplicate content is an old spamming trick used to get higher rankings by stuffing pages with the same crap. An old spamming trick used to get higher rankings is by stuffing pages with the same crap, Duplicate content.

Okay, all joking aside. Dupe content litters the internets and frankly, Google hates it, or so I’ve been told. Its understandable. I hate it too. There’s nothing lamer than searching for something specific, only to have the same crummy Britney Spears article coming up 180,000 times in your search results. But in a world of Web-2.0-user-controlled-buzzword sites, cross domain hosting and utilizing free web space is crucial to those of us who’re too poor to afford paying some hosting company an annual fee. So we cheap out!

We upload our photos to flickr and other free sites, dump our videos on YouTube, then link them to our free blogs! And why not? The server space is there to be used…

Anyhow, I’m getting a little off topic, sort of. With all these free sites and people linking here and there, it begs the question, how does Google filter the duplicates? And what about mirror sites? Is it so absurd that a legitimate, quality content site may be so popular that it requires a second location for people who want to download that nifty photoshop crack virus? Inquiring minds want to know!

So I’m trying a little experiment…

I have reproduced this blog on the Google owned “blogspot.com” as a test to ascertain what Google decides is duplicate content. I am reproducing every post on this site on that one as well. I am not using any aggregated feeds, I am manually copying the posts over. Its also a little test to see which blog hosting site I like better, Wordpress or Blogspot (so far, I like the look of Blogspot more, but I prefer the interface of Wordpress).

I have also included (currently as of this post) my one and only link on this site to the mirrored site. I’m not sure if that constitutes as a spider trap or not, but I’m curious to see which of these two sites get the ranks. Or maybe Google will punish me for trying to monkey around with them… who knows?

In any case, I’ll keep this experiment going until I either get bored with it, or something interesting happens…

In the meantime, here’s a video about duplicate content from a top SEO Specialist: