Timberline Interactive

My i-merchant column this month is on the benefits of honkin' big add-to-cart and checkout buttons. Size matters, baby!

I just finished re-reading Steve Krug's usability bible, Don't Make Me Think, which I recommend highly for anyone responsible for making their website a success. Not only is Steve's design and navigation thinking impeccable, he's also a lively and hilarious writer, and the book's snappy, well-illustrated look is a conscious example of the web best-practices he preaches.

I also read Avinash Kaushik's Web Analytics, An Hour a Day which is a passionate, smart and at times funny book. Now I need to drill down into a technical, hands-on Google-Analytics and Website-Optimizer-specific tome.

I'm in Dallas, Texas, right now, delivering a Google Analytics and Website Optimizer training to the brilliant and terrific team at Fifth Food Group.

You can make any website better -- lots better. The footprints of failed pages, bad usability, are all over the analytics. And when you identify and roll out a change, do it in Google Website Optimizer, so you can document a lift (or, in some rare cases, prove to yourself that your brilliant gut instincts weren't brilliant after all.)

Yahoo! announced it was buying web analytics business IndexTools, setting the stage for Yahoo to offer an answer to Google Analytics.

Google bought the Urchin business in 2005, and set the web analytics industry on its ear by rebranding it as Google Analytics and giving it away free. Google Analytics is now used by hundreds of thousands of businesses and organizations -- including, says Google, scores of Fortune 500 companies.

Critics note that the software may be "free," but that Google Analytics users give Google something of significant value: data about merchants' traffic, paid search conversion rates, average order values, and more. All that data tells the ad-selling side of Google's business a great deal about how much higher click prices can go before they become uncompetitive with other direct-response advertising media.

Chris Middings, web editor at Seventh Generation, tipped us off to an interesting new study from EngineReady. A two year research product capturing data from 27 ecommerce websites, the report is titled "SEO vs. PPC -- The Final Round." The basic methodology was tracking the dollars of business transacted by each of four traffic sources, and averaging that value on a per-visitor basis. As the report's title suggests, the project aims to resolve the question of whether paid-search visitors are "worth" more or less than visitors entering a site from free, "organic" search.

The study doesn't really succeed in answering that question, in my opinion. But first, here are the dollar values per visitor segment reported by Engine Ready:

• Direct Access (visitors who typed the URL or used a bookmark): $5.69
• Other Referring Sites: $5.01
• Paid Search: $1.91
• Organic Search: $1.35

Obviously, visitors typing your specific URL or choosing a bookmark from their browser favorites are your most qualified customers of all -- people already familiar with your brand, perhaps they are repeat buyers. Or their interest in your URL has been driven by some offline advertising, such as your direct-mail catalog, television or radio ads. That they represent the highest value-per-visitor is no surprise, and we see the same results among Timberline clients.

EngineReady doesn't explain why "Other Referring Sites" rank so valuably, merely offering that it shows the importance of "link building" above and beyond its SEO benefit. But I think the study's authors largely missed the point here: For many ecommerce merchants, "referring sites" are dominated by two sources:

• Affiliates (assuming they have an affiliate program)
• Recipients of your email newsletter who use web-based email clients like Gmail, Hotmail, etc.

Neither of these are "link building" for SEO purposes. Both will be made up largely of qualified visitors -- people responding to your loyalty email program, or (since so much affiliate traffic is driven by brand-name searches) people stumbling through an affiliate site on their way to find your site. If that's the case, it's no wonder "Referring Sites" yield a value not far off from Direct Entry.

Another omission, to anyone who has done a lot of online marketing, is the failure to segregate brand and non-brand terms in the Paid and Organic buckets. We all know that your brand name performs worlds differently than generic terms. The latter bring you new customers and incremental business, but usually at a conversion rate below your existing customers.

Inspired by EngineReady's research, I took a cross section of Timberline clients and ran the math, but I went the extra yard and segregated the SEO traffic by brand and non-brand terms. Here are my results:

Direct Access: $8.24

Other Referring Sites: $2.46

Paid Search: $4.34

Organic Search (all): $4.24

Organic (brand name only): $9.23

Organic (non-brand): $2.15

Why do Timberline clients differ from EngineReady's numbers? I believe it's because:

• Many have strong brand names within the niches they occupy
• Relatively few have robust affiliate programs
• Many have significant non-commercial content like recipe, articles, blogs and community features
• We do a good job managing paid search programs to high ROI :)

All in all, though, EngineReady's research is compelling, and makes good food for thought. check it out at http://engineready.com/company/trafficstudy.html

UPDATE: The problem has been resolved

Since January 9, Google Analytics has been experiencing technical problems capturing traffic data coming from Google AdWords cpc platform.

While there is no official word yet on the cause of the problem, we are in touch with Google Tech Support, and Google engineers are working on it. Our initial suspicion is that the situation is connected with Auto-Tagging, a setting that enables advertisers to pass campaign information into Analytics without manually tagging their links with tracking information.

The problem does not affect the customer’s experience. Customers coming from paid Google ads are still able to place orders, and if you’re using Google Adwords conversion tracking, orders are still being tracked as conversions within the AdWords platform.If your Google ads are manually tagged using Conversion Ruler, Atlas, Omniture or other third-party tracking tags, you’ll be unaffected.

The problem is apparently affecting a number of other sites. Here is a thread devoted to the topic at Google Groups:

We will keep you updated as soon as we have more information!