This morning, Internet Retailer came out with a release that said that web site performance is more important to web shoppers than brand loyalty. Read the release here.

90% of web site shoppers will give up after three or fewer unsuccessful attempts to complete a purchase. Survey respondents went as far as to say that 75% of sites load too slowly, and half of those would consider shopping on a competitor’s site to make the purchase.

If you run a small internet retail operation, this underscores the importance of keeping your web site running as quickly as possible. Additionally, just as many customers abandon web sites because they can’t find what they are looking for and order totals are not shown early in the checkout process.

If you’re into search engine marketing and search engine optimization, you might want to read the notes from the SES Chicago ’06 conference sessions. There are a few bloggers that have covered each the sessions from the SES Chicago ’06 conference that wrapped up today, and they can be found below:

• Search Engine Roundtable’s coverage
• Lisa Barone’s coverage on BruceClay.com

More news from the search world–Search Marketing Standard has started a blog.

The holidays are here, and if you’re in internet retail you’ve hopefully already solidified an online marketing plan for the holidays. That plan probably involves email marketing, so I’ve devised some reminders of what you should do and avoid doing so your email marketing is as effective as possible during this crucial time of year.

Don’ts:

• Use XHTML code in your emails (Use HTML as there are delivery errors with XHTML).
• Don’t reference external CSS (Cascading Style Sheets) in your HTML email as most email clients can’t read it. Instead, use inline styles or font tags. You can also reference my earlier guide for using CSS in email campaigns to learn what CSS definitions email clients can handle.
• Send email without validating it. Make sure you close all of your HTML tags so the page renders properly in all email clients. You can validate your HTML email with the W3C Validator.
• Embed forms or surveys in your email marketing campaign. Instead, create a landing page on your web site with the form or survey and include a link to this page in your email campaign.
• Send an all-graphical email. Some email clients will hide images by default, and your customers won’t see your message at all when accessing it through a preview pane.
• Use javascript. This will trigger all kinds of SPAM filters.

Do’s:

• Include some “administrative” text at the top of your email containing your company name or web site address.
• Post the HTML email on your web site and offer a link to it at the top of your email campaign. The text should read, “View this email at our web site” or something similar. “Having trouble viewing this email? View it on our site” is also a popular option.
• Link all of your images back to your web site.
• Include heights, widths and ALT tags on all images. Some email clients that hide images by default may assign a value to the height and width of your images, pushing content further down the page.
• If you offer a phone ordering line, include it in the email campaign.
• Design all HTML email for a 500-600 pixel width.
• Include your site categories/hierarchy as text links. Again, if your customers’ email clients hide images by default, customers will still see the text links to your main site categories. My research has shown that as many as 60% of customers click the navigational links to the main categories of your site.
• Check spelling. Not only is a misspelled word unprofessional, it can also increase your SPAM score.
• Produce a text and HTML version of your email campaign. Multi-part MIME email requires both and not including a text version could increase your SPAM score.
• Make text versions no more than 60 characters wide.

If you’re looking for a good editor to create your email marketing campaigns, try Macromedia Dreamweaver. Using Word or FrontPage to create HTML email marketing campaigns is not a good idea since they do not render “clean” HTML and assign styles to your campaign that most email clients cannot decipher.

If you’re in the Richmond, Virginia area, come out to the Media General Media Center downtown tonight. I’ll be sitting on a panel with Eddie Wilson and we’ll evaluate the design and usability of eight web sites chosen by Refresh members. Check out the Refresh Richmond web site for details and directions.

I wanted to post a follow-up to yesterday’s post regarding shopping cart abandonment rate to provide a real-life example of why shopping cart abandonment matters.

I have a new client‚Äîwe’ll call them “Client A”‚Äîthat recently asked me to improve their conversion rate. There was one problem, though. Client A didn’t have a web analytics package in place to accurately measure their conversion rate. After installing Google Analytics on Client A’s site and configuring a checkout funnel, we found that their cart abandonment rate was a whopping 87% (industry average is about 60%). This is one case where higher-than-average isn’t better.

“Client B”, a long-time client, has a much better cart abandonment rate. We’ve worked for years to improve checkout and get it running as efficiently as possible. After all, Client B doesn’t get as much traffic as Client A, so stellar conversion is a must if they want to make money. Client B’s cart abandonment rate is currently 19%, so their site is getting 80% of their users that reach the shopping cart past the cart and into checkout.

Client A only sent 1 of 10 people through their shopping cart into checkout while Client B was sending 8 of 10 people through.

Knowing this, I can now turn my attention to solving Client A’s cart abandonment problem. We’ll try different page designs and process flows until we achieve Client A’s goal of getting 4 of 10 customers through to checkout, and this will have a huge impact on revenue.

This, in my opinion, is why monitoring cart abandonment is so valuable. Not only will your site generate more revenue per visitor as your shopping cart abandonment rate decreases, but your ad spend also becomes more effective.

An article from Kevin Hillstrom called “Shopping Cart Abandonment” recently asked the question, “Is there anything wrong with a sixty percent shopping cart abandonment rate?”

Hillstrom was reacting to a blog post from the Marketing Experiments Blog that concluded that simple page changes could increase conversion rate by decreasing cart abandonment rate.

In his blog post, Hillstrom concluded that “an abandoned shopping cart is a non-issue.”

I couldn’t disagree more.

There are definitely cases where having high shopping cart abandonment is inherent. Having worked in electronics and consulting for a golf retailer that frequently sells products governed by Minimum Advertised Price guidelines, I understand that a customer must place something in the shopping cart to see the actual price of the product.

In all cases, however, you want as many customers as possible to proceed through the shopping cart and into checkout. It’s a natural part of improving your checkout process. So isn’t it at least worth testing different designs and messages to see if you can improve your conversion rate rather than just dismissing abandonment as a non-issue?

Absolutely! It’s a no-brainer. Why? Let’s assume you have a site with 10,000 visitors per month. 25% of shoppers on average place something in their shopping cart, which means that 2,500 shopping carts are created per month. If the industry average‚Äî40%‚Äî proceed to checkout, then 1,000 people per month enter checkout. Increase the success rate to 45% and 1,125 people per month enter checkout. That’s 125 more chances per month to get a sale!

You can see that even a small decrease in cart abandonment will result in thousands of dollars per year (or more) in incremental revenue. The only cost of getting this incremental revenue is the time it takes for staff to run and analyze one-page shopping cart tests. We’re not talking rocket science here‚Äî simply changing the size, placement and color of a button could be a test.

All retailers should run tests to optimize their ecommerce checkout processes. If they don’t they’re leaving lots of money on the table.