Non e-commerce web sites often measure their success based on the averages analytics tools provide.  Does average page views per visit or average time spent on site sound familiar?  If so, the video below is worth your time.  In it web analytics professional Avinash Kaushik lectures on how averages can blind us from seeing the valuable data that can drive effective decision making.

If you enjoy this video, consider Avinash’s book Web Analytics and Hour a Day, and add his web analytics blog to your feed reader.

Non e-commerce web sites often measure their success based on the averages analytics tools provide.  Does average page views per visit or average time spent on site sound familiar?  If so, the video below is worth your time.  In it web analytics professional Avinash Kaushik lectures on how averages can blind us from seeing the valuable data that can drive effective decision making.

If you enjoy this video, consider Avinash’s book Web Analytics and Hour a Day, and add his web analytics blog to your feed reader.

You can read the entire experiment parameters and results in the original post, but I wanted to check-in and confirm that the rankings have maintained over time.  Here are two Google screenshots taken this morning for one and two-word Google searches:

• +3 days - Google Page 1, result 5 (Search phrase: “iPhone dynolicous”) - a decline of 1 position
• +3 days - Google Page 1, results 2 and 3 (Search phrase: “dynolicous”) - an improvement of 7 positions

Evidence, but this test isn’t that compelling

While the test did validate that the method has some merit, the terms which I selected are deep into niche territory and don’t receive substantial traffic.  To further demonstrate the ability to achieve Google rankings in this manner I’ve run another test with an increased difficulty level. I’ll be publishing those results soon.  Stay tuned.

Here’s an example.  This is the marketing page on Alltop: http://marketing.alltop.com/

Don’t like everything you see? If you prefer not to view a blog that is listed in a category, you simply delete it by clicking the “x” beside the name of the blog (there is an option to restore the page so it displays everything in a category).

Yes, Online Marketing Performance is listed on Alltop; it’s in the marketing catgory.

Here’s an overview of one of my latest experiments:

Hypothesis: Google is giving preferential treatment to blog posts with video syndicated from a Google video sharing site (Google Video or YouTube).

Experiment: Post reasonably well crafted blog posts with relevant embedded video and monitor keywords and keyphrases to see if ranking could be achieved.

Prediction: Page 1 Google SERP listings in less than a day for targeted keyphrases.

Blog/Video/Post Details:

1. Blog to post on: http://www.on-a-mac.com
2. Post topic: an iPhone App (called Dynolicous) for which I had recently seen a video on YouTube
3. Keywords: Present in the post title (and title tag), page name, post content and post tags

Results: This is the third test I’ve personally run and I’ve had the same remarkable result.  When searching Google for targeted keywords and keyphrases:

• Page 2 or better Google rankings in less than  15 minutes
• Page 1 Google rankings in less than 4 hours
• A few page 1, result 1 Google rankings for more detailed searches which had direct competition (three word searches)
• Without any type of maintenance/attention, some rankings have slipped downward over time but still held page 1 and page 2 positions on Google.  Other rankings have held stable, even on page 1 of Google results (those page 1 rankings were for two or three word search phrases).

Timeline to Page 1 on Google (Test date: 19 August 2008):

• 9:41 am - Blog post with video published (How accurate is the iPhone Dynolicious? [video])
• 9:56 am - Google Page 2, result 1 (Search phrase: “iPhone dynolicous video”)
• 1:12 pm - Google Page 1, result 10 (Search phrase: “iPhone dynolicous video”)
• 1:24 pm - Google Page 1, result 4 (Search phrase: “iPhone dynolicous video”)
• 1:27 pm - Google Page 1, result 1 and 2 (Search phrase: “iPhone dynolicous video”)
• 1:28 pm - Google Page 1, result 4 (Search phrase: “iPhone dynolicous”)
• 1:28 pm - Google Page 1, result 10 (Search phrase: “dynolicous”)

Some of these screenshots (below) are from my iPhone as I left the office right after I posted the video.  I was unfortunately in that meeting for a few hours, so I’m not sure exactly when the results moved onto page 1.  When searching Google from the iPhone I was using the Safari web browser and the same Google site (not the Google Mobile Search site or the Google Search App for the iPhone).

The iPhone screenshots have the time from the phone on them.  For the other screenshots I’m relying on my Mac’s finder to get the “Date Modified” time.  The only thing I’ve done to these pictures is blurred my email account as I was logged into Google when a few of the screenshots were taken (being logged in to Google had no effect either way in case you were curious).

This technique has no effect whatsoever on MSN/Live or Yahoo.  I would anticipate that this effect will not last over time on Google either.  Until then, enjoy!

I’m sure there’s plenty of these floating around, but if you’re curious, this is the email response that myself and others received after the transaction went through:

Dear Scott McAndrew,

The following domain name has failed to be registered:

REMOTE.ME

Error: REMOTE.ME:  cannot register - already registered

We will evaluate this error and retry the registration
if appropriate.

If we are unable to successfully register the domain
name, your account will be credited accordingly. Please
allow one business day for the refund to be processed.

Please contact GoDaddy.com, Inc. if you need any further
assistance:
http://www.godaddy.com/gdshop/support.asp?prog_id=GoDaddy&isc=gdbb14

Sincerely,
GoDaddy.com, Inc.

Once the dust settles, it will be nice to hear what exactly happened.  The response that GoDaddy issued once the problem had been identified (around noon Pacific) is as follows:

We knew the .ME Open Registration response would be tremendous, but it went beyond even what we had expected. As a result, we experienced some system issues in our communication with the registry.

As soon as we became aware of the issue we began taking steps to correct it. It is now resolved.

For our customers, if we did not successfully register the domain name requested they will receive a full refund.

At 10 am I was still able to find a few I wanted to register for future use:

• pleasehelp.me
• remote.me
• open.me

UPDATE: Well, it turns out that no, I did not get any domains.  While the process I went through told me that I did, the domains did not register.  The email address that I had on file with GoDaddy isn’t one I check on a regular basis, so it wasn’t until this afternoon that I realized that I did not get any .me domains either.

Anyone out there pick up any interesting .me domains?

It’s been a busy week, and I probably would have missed out on the opportunity had it not been for Jason Baer’s digital marketing blog.  Thanks Jason!

Performance problems are not limited to viewing RSS feeds; as blog pages rely on Feedburner also, blog visitors are experiencing pages with excessive loading times and/or rendering problems where content is omitted or displayed incorrectly.

As soon as Feedburner addresses the issue, performance on this blog should return to normal.  Thank you for your patience.

Here is an example of where you might come across this metric.  The screen capture below is from another site I manage.  This report is the Traffic Sources > Keywords report.

As you can see, for several keywords I’ve registered average visit lengths of no length whatsoever (00:00:00).  That doesn’t sound good.  Let’s look a little closer.

Keywords in rows 5, 6, 7, 8 and 9 are the 00:00:00 visits.  They also have something else in common.  Each of these visits is of only one web page (the value in the Pages/Visit column is 1.0 for each of these keywords).  Single page visits, called Bounces in most analytics packages are handled different ways by different packages.  Before moving forward, let’s be clear about what a Bounce is.

A Bounce, or Single Page View Visit is a visit in which the user arrives at a web page, and then leaves the web site altogether. If the user even refreshes the same page, it is not a Bounce. A bounce is someone arriving at your site, then doing one of several things such as clicking the back button, typing a new destination into the address bar of their browser, closing their browser program altogether (you get the point - basically anything but continuing to view pages on your web site).

OK.  So all of our 00:00:00 visits were also Bounce visits.  That still seems quite surprising.  None of the visits registered even a second on the site?  Well, the visits may have been a second, ten seconds, or several minutes.  We just don’t know, and we don’t know due to the way Google Analytics calculates Time on Site (or Time on Page).

To calculate how long a user is on a page, Google needs two data points.  It needs the time when a page was loaded (which in our case it has) and the time when a subsequent page was requested from the site (which it does not have).  So, in lieu of that last piece of data, Google classifies these Bounce visits as 00:00:00 in length.

It would also stand to reason that every visit to a web site is registered as shorter than it actually is as the amount of time spent on the last page of the site prior to exiting can’t be calculated.

On a side note…
If you are reviewing metrics on individual pages, Google does not include Bounce visits into the equation when it calculates the average time spent on that page.  For the average time spent on site, however, it does.

I am happy to say that for the most part numbers computed were roughly within +/- 20 to 25%.  There were some estimates that were wildly off (by a factor of up to 2x off for a few keywords I checked) and there were a handful that were very close, but +/- 20 to 25% seemed to generally be the delta.

I’m looking forward to learning more about how various tools (or estimation methods) compare to the estimated search volume numbers now being provided by Google’s Keyword tool.  Have any data to share?

I am happy to say that for the most part numbers computed were roughly within +/- 20 to 25%.  There were some estimates that were wildly off (by a factor of up to 2x off for a few keywords I checked) and there were a handful that were very close, but +/- 20 to 25% seemed to generally be the delta.

I’m looking forward to learning more about how various tools (or estimation methods) compare to the estimated search volume numbers now being provided by Google’s Keyword tool.  Have any data to share?

From the Google Inside AdWords blog:

Based on advertiser feedback, and our commitment to provide useful tools and information for our advertisers, we’ve now added search volume data to the Keyword Tool.

…

These approximate numbers are intended to provide better insight into keywords’ monthly and average search volumes than previously provided by the tool.

If you’re new to Google AdWords, their Keyword Tool previously provided a visual cue which indicated search volume.

If you want to check it out and your not currently an AdWords user you can use Google’s AdWords external keyword tool.

From the Google Inside AdWords blog:

Based on advertiser feedback, and our commitment to provide useful tools and information for our advertisers, we’ve now added search volume data to the Keyword Tool.

…

These approximate numbers are intended to provide better insight into keywords’ monthly and average search volumes than previously provided by the tool.

If you’re new to Google AdWords, their Keyword Tool previously provided a visual cue which indicated search volume.

If you want to check it out and your not currently an AdWords user you can use Google’s AdWords external keyword tool.

From the Google Inside AdWords blog:

Based on advertiser feedback, and our commitment to provide useful tools and information for our advertisers, we’ve now added search volume data to the Keyword Tool.

…

These approximate numbers are intended to provide better insight into keywords’ monthly and average search volumes than previously provided by the tool.

If you’re new to Google AdWords, their Keyword Tool previously provided a visual cue which indicated search volume.

If you want to check it out and your not currently an AdWords user you can use Google’s AdWords external keyword tool.

From the announcement post:

Now that we’ve launched our Flash indexing algorithm, web designers can expect improved visibility of their published Flash content, and you can expect to see better search results and snippets.

This is undoubtedly a step in the right direction, but there’s still a good deal of ground to cover.  While .SWF files will be indexed, it will only be for text content contained within the .SWF file.  Content that is pulled in by a .SWF for presentation as opposed to be included in the .SWF will be indexed separately.

As it is described in the Q&A post, the code which parses Flash files will essentially be acting like a user.

We’ve developed an algorithm that explores Flash files in the same way that a person would, by clicking buttons, entering input, and so on. Our algorithm remembers all of the text that it encounters along the way, and that content is then available to be indexed.

There are obvious limitations to overcome, in my mind specifically with the inability to index dynamic content (content pulled in from a database or external XML file), but this is a big move in the right direction, and its been a long time coming.

Thanks to Chris ‘Ubergeek’ Hill for emerging from vacation long enough to pass this info along.

From the announcement post:

Now that we’ve launched our Flash indexing algorithm, web designers can expect improved visibility of their published Flash content, and you can expect to see better search results and snippets.

This is undoubtedly a step in the right direction, but there’s still a good deal of ground to cover.  While .SWF files will be indexed, it will only be for text content contained within the .SWF file.  Content that is pulled in by a .SWF for presentation as opposed to be included in the .SWF will be indexed separately.

As it is described in the Q&A post, the code which parses Flash files will essentially be acting like a user.

We’ve developed an algorithm that explores Flash files in the same way that a person would, by clicking buttons, entering input, and so on. Our algorithm remembers all of the text that it encounters along the way, and that content is then available to be indexed.

There are obvious limitations to overcome, in my mind specifically with the inability to index dynamic content (content pulled in from a database or external XML file), but this is a big move in the right direction, and its been a long time coming.

Thanks to Chris ‘Ubergeek’ Hill for emerging from vacation long enough to pass this info along.

Visa is giving $100 in Facebook Ads to the first 20,000 small business owners that install their Visa Business Network app on Facebook.  After installing the application and answering a few questions a $100 Facebook advertising credit notification arrives in your inbox.

What’s the Visa Business Network?  Here’s what the App description provides:

A place that can help you and your business succeed.

Connect with other small business owners, learn ways to manage your business more efficiently, and grow by reaching the millions of potential customers on Facebook.

It’s free to join and easy to use. Sign up now to redeem your $100 in targeted Facebook advertising, and start connecting.

Use The Visa Business Network to:

GROW: With over 80 million users, Facebook is a great place to build your business. When you join the network, you get $100 to create targeted advertising and reach potential customers on Facebook.

CONNECT: The Visa Business Network connects you with other small businesses. Share ideas, information, opinions, and knowledge with people going through the same things you are.

MANAGE: The Resource Center features exclusive content that gives you great ways to help you manage your business. Take advantage of articles, videos, and tips from top small business sources.

Log in to Facebook, then add the Visa Business Network Application.

Visa is giving $100 in Facebook Ads to the first 20,000 small business owners that install their Visa Business Network app on Facebook.  After installing the application and answering a few questions a $100 Facebook advertising credit notification arrives in your inbox.

What’s the Visa Business Network?  Here’s what the App description provides:

A place that can help you and your business succeed.

Connect with other small business owners, learn ways to manage your business more efficiently, and grow by reaching the millions of potential customers on Facebook.

It’s free to join and easy to use. Sign up now to redeem your $100 in targeted Facebook advertising, and start connecting.

Use The Visa Business Network to:

GROW: With over 80 million users, Facebook is a great place to build your business. When you join the network, you get $100 to create targeted advertising and reach potential customers on Facebook.

CONNECT: The Visa Business Network connects you with other small businesses. Share ideas, information, opinions, and knowledge with people going through the same things you are.

MANAGE: The Resource Center features exclusive content that gives you great ways to help you manage your business. Take advantage of articles, videos, and tips from top small business sources.

Log in to Facebook, then add the Visa Business Network Application.

The new posture opens the door for almost any domain suffix imaginable, likely with .sex and .xxx leading the way. With most memorable names consumed under “.com,” that this ruling will pave the way for more descriptive suffixes, such as “.resort” or “.books”.

The AP reports that new TLDs won’t surface for several months. Regulation? Domains will need to be proposed to ICANN at which time they will go through a review phase allowing anyone to raise an objection.

Welcome to the next domain rush. Or maybe not. Citing ICANN, Ars Technica reports that “if approved, registering the TLD will cost anywhere from $100,000 to $500,000.” Maybe it’ll be the domain rush for everyone but the little guy.

In addition to showing relative search volume trends, the site also maps news results onto the provided timeline. The graph Google Trends returned below is for the query “social networking” and “social media” (Social Networking is shown in blue, Social Media in red):

Below the graphed information Google also provides relative search volume by regions, cities and languages.

A recent addition is the ability to pull up the traffic on web sites. Google Trends allows entering multiple sites, just as it does for keywords. Regional information is also provided, as well as “also visited” sites and “also search for” keywords. Here’s MySpace.com in comparison to Facebook.com (MySpace.com is shown in blue, Facebook.com in red):

Just another tool you may want to employ when researching keywords or web sites. Don’t miss Google’s disclaimer at the bottom of the screen:

“Google Trends provides insights into broad search patterns. Please keep in mind that several approximations are used when computing these results. All traffic statistics are estimates.”