Avinash Kaushik wrote a great post about tracking offline conversions at Tracking Offline Conversions: Hope, Seven Best Practices, Bonus Tips.  I interviewed Avinash last year about his book in this post – Q&A with Avinash Kaushik, Author of Web Analytics: An Hour A Day.

On Google’s Webmaster Central Blog, they talk about their latest improvements in indexing Flash content in Improved Flash indexing.   I still wouldn’t use a great deal of Flash for an e-commerce site, but it’s good to know that they are making progress on improving this area.

After playing with it a bit, I started to think about how it might be used by retailers for useful information.  One of the clear ways may be to look at two brands like I’ve done in the screenshot below.  Comparing shoes from the brands Sebago and Rockport, show they are fairly even in their success overall, but show very different peaks and valleys.   I think it would be interesting to compare Google’s results to your own sales metrics and see how they differ.

When checking it out, be sure to zoom in on specific time frames and price points.  You can start to see patterns of how the products do at various points.   

Merchandisers will have to keep in mind that this sales data is only for online transactions and those that went through Google Checkout, which is an extremely small portion of online sales at the moment.  Never the less, it can still show interesting trends. 

One of the disappointing things that makes this less useful is that Google only shows top tier items so trying to compare products

This is the first view of the widget which allows you to add holidays that you are planning for.  I added Christmas and Valentines day.  It then also allows you to take notes for each of the holidays you have added to your list.

Some of the things I would have liked to have seen here:

• Selection of holidays that I can add with one click. 
• In Context highlighting in the notes, so when I type in a note like  ”find Wii for Christmas”, it would automatically hyperlink the Wii or other known keywords to landing pages or the item itself. 
• Countdown to the next holiday date as well as countdown to the important days for ordering such as last day to order with standard shipping. 

The second view is called Experiences.  This one didn’t add much to the experience unfortunately.  It provided a few major categories of experiences you could have, with a few items in each that jump to the item on the web.  I would have liked to see the capability to add some profile information to the widget for my wife and kids including age and interests and it would provide back intelligent ideas.  I didn’t see much method to the madness here except what they wanted to promote. 

The third tab is “Week’s Picks”.  Oddly though it defaults to a map icon.   Clicking on the map icon prompts for your zip and it then launches the store locator in a browser window.   This was useful as a shortcut, but it’s not something I would use over and over.  Selecting a category on the left brought up promoted items in that category that linked to the item page on the web. 

To make this really useful:

• It would have been great to be able to add in my own wishlist of items and have it figure out what store had all or most of the items in stock. 
• Further on that train of thought, I could add an item like a Wii, and an alarm would go off letting me know the moment a shipment of Wii’s were coming in. 
• Last, with the wishlist, I would have liked to be able to set other alarms for when products go more than x% off or for $y price or below.   I think this information coming back into the merchandising team would be really interesting as you could analyze for the hot products and at what price points people were looking to obtain them for.    Almost a soft commit Priceline for retail products. 

Last is a standard search of their Internet site.   Nothing special.

This is the igoogle widget.  Same look and feel, but in your browser.

The uh-oh button was a great idea and reminded me of Friendlyware in the early 80′s with the bosskey that would escape out of a game to a screen that looked like work.  Unfortunately, Bestbuy’s implementation was just an exe that you put on your desktop forcing you to minimize/move your browser so you could find the icon and launch it.  I would suspect that most people that downloaded and tried it, had no idea what they were supposed to do with it.  A better implementation would have been in the webpage itself and when clicked it have taken the consumer to another page that looked like real work.  A whole ad campaign could have been made out this and would have appealed to many of the techies around. 

Overall it was a good effort, but it could have been a real driver of loyalty if implemented further with functionality that provided more value. 

Spacedog and Avenue A/Razorfish built the front & backend.  You can download the widgets at Bestbuy.com.

Please check out The Top Ecommerce Sites and let me know what you think. 

Currently I’m tracking on each site:

• Snapshots of the home page
• Google Page Rank
• Backlinks from major search engines
• Traffic via Compete
• Company data from ZoomInfo
• Types of technology used and sizes (i.e. html, images, flash, javascript)
• User submitted comments and reviews
• 1-5 Star ratings submitted by users
• References in the Blogosphere
• Google News

I plan to track in the future:

• Catalog features of each site (i.e. Reviews, social aspects, single page checkout)
• Stock & finance information
• Other traffic sources
• Open jobs
• Competitors
• Snapshots of other pages like product detail for historical reviews

The site, data, and code is still beta, so problems may come up.  If you see any, please send me a note of what you see and I’ll look into fixing it. 

Please contribute to the resource by rating and writing a review for some of the sites.  I’m primarily looking for thoughts on design, SEO, structure and ability of the sites.   If you have a blog or website, the reviews are set to follow, so in exchange for your review, you’ll get get a link back to your site.   Every comment is moderated for spam and foul language prior to it being published. 

If you have any thoughts, ideas on features or api’s that would be useful, or questions you would like to ask, please send me an email (link in the sidebar).

Last, a note of thanks out to Suresh.  I appreciate your help during late nights and weekends helping me put this together.    It’s appreciated.

1.  Analytics
2. Merchandising for conversions
3. Search & SEO
4. Web 2.0

I’ve got to get back to the show, but later tonight I’ll cover the main points from the keynotes and what I’ve heard and learned.  Looks like it’s going to be a great event.