Website Test & Tweak Trumps Trash: Why Small Changes are More Productive than Whole Site Re-Designs

Thank you Lindy Rawlinson, VP at Neiman Marcus Direct for saying what you said in your keynote at the Internet Retailer Conference.

• Testing lets the market speak "Sometimes what you think is the better [site] layout doesn't always win", she says.
• Don't make too many drastic changes at once. You may feel good doing a redesign but introducing unfamiliarity to your repeat visitors can be jarring.

Article from Multichannel Merchant follows. Italics are mine.

Live From Internet Retailer:
Neiman Marcus on Making a Good Site Experience Better
Jan 20, 2009 1:24 PM, By Melissa Dowling

Miami -- For high-end retailer Neiman Marcus, it's all about the merchandise. But how you present that merchandise and engage the customer online is critical, according to Lindy Rawlinson, vice president of Web operations and new business for Neiman Marcus Direct.

Delivering a keynote address at Internet Retailer's Web Design conference here on Jan. 20, Rawlinson noted that before you undergo a Website redesign, you have to understand your customer. You also have to make sure that whatever you do keeps that customer engaged.

It's best to make small changes and enhancements, Rawlinson said: "Don't do too much too quickly." And make sure you test, she added. "The Web allows you to do this easily."

Neiman Marcus Direct is a big fan of A/B testing, Rawlinson said, and conducts these types of tests whenever it can. "Sometimes what you think is the better [site] layout doesn't always win," she noted.

The retailer has made several tweaks to its Neiman Marcus site, including launching a new search engine and adding guided navigation features for its sales items in spring 2008. And it spent the next few months analyzing the results and making changes, Rawlinson said.

The company recently redesigned its Bergdorf Goodman Website. Among the changes: It widened the site to allow for larger images (which it had done for the Neiman Marcus site this past fall); it moved the store section to the top of the Website; and put the video and fashion blogs all in one area in a prominent position on the site, Rawlinson said.

It did not make radical changes, however. Rawlinson said you should never underestimate the power of familiarity--even if you think you can improve something, keep in mind that your customers know how to use it as it is. "We don't want to do anything that will disrupt the shopping experience for our customer," Rawlinson said.

Comments?

June Li
ClickInsight
====..
...continue reading >

Free Webcast : Testing Website Content, Layouts & More

To help you succeed with New Year's Resolution #1: Test your Pages, attend this free Web Analytics Association webcast Testing, Testing, Anyone Can Test. So why don't you? on February 5th.


Sponsored by Unica, here are the details:

Testing, Testing, Testing. Anybody Can Test. So Why Don't You?
February 5, 2009
12:00 PM EST 9:00 AM PST
Open to Members and Non-Members

While digging for nuggets of insight in web analytics is one method of finding gold, the mainstream method is to improve success rates by testing alternative layouts, content, offers, ads, etc. and picking the winners. There are so many ways that web marketers can go after testing today: You can do it yourself. You can use free test automation solutions and of course high-end multivariate testing solutions. There are ups and downs to each approach. Which approach is right for you and how do you get off the couch and into the trenches now

Join our renowned panelists for an insightful discussion and join in with your own questions to the panel consisting of:

• Josh Manion, Stratigent
• Bryan Eisenberg, FutureNow, Inc.
• Akin Arikan, Unica

Register Now...
...continue reading >

Manageable New Year's Analytics Resolution #3: Share Your Learnings

Resolution #1 was "Test Your Pages".
Resolution #2 was "Ask Your Web Visitors about Their Experience".

Resolution #3 is "Share Your Learnings". Share in a way that will get engage your listeners, your organization and grab their interest.

Share the results of your page testing adventures.
Share the results of what your web visitors say about their experience on their website.

But don't just create a boring powerpoint presentation.
In fact, try doing this without a powerpoint.

Create a story.
With drama.
What did you think you'd learn?
What did you actually learn?
What will you do next?
What should everyone stay tuned for?
What you want to do is create interest in analytics, and evolve working with analytics so that it's not boring stacks of reports but "contagious fun", as Kim Johnston of Symantec enthusiastically described in her eMetrics Washington keynote (see summarized post).

So how do you get started? The above is pretty abstract. Difficult to get started with. And we analysts, we're pretty methodical and like frameworks and examples to analyze and learn from.

A great resource is "Made to Stick" by Chip and Dan Heath. Lots of examples of stories that resonate with audiences and don't, and analysis of why. Six principles of sticky SUCCESs (read about them here in the book excerpt, and then buy the book) that will help you create stories of analytics adventures as memorable as urban legends like the "kidney heist" (see the book excerpt) and Aesop's Fables.

Your thoughts? Your stories?

June Li
ClickInsight
====..
...continue reading >

Manageable New Year's Analytics Resolution #2: Ask Your Web Visitors About Their Experience

Resolution #1 was "Test Your Pages".

Resolution #2 is ... "Ask Your Web Visitors About Their Experience", before the end of March *.

In my opinion, asking for input is a necessity. Web traffic data analysis is not enough. Analyzing web traffic shows you what visitors did on your site, where visitors came from, and how long they were on your site. But you don't know whether they actually completed the task that led them to your website, and even if they did complete, they could have found the experience confusing, difficult or frustrating.

• A one page visit could have netted you a raving fan, ecstatic they were able to get the information they needed on the first page they visited. They may have read, printed the page and even forwarded the URL, but your web analytics tool would record this as a single page visit, or a "bounce" because they only looked at one page. Bounces are typically interpreted indicators of poor performance, indicative of an un-engaged visitor.
  

• On the other hand, a 10 page visit and a view of the contact us page may have been a frustrating experience, that did not result in the visitor achieving their goal.
  

How do you find out?

Ask them. Survey them. With just 4 quick questions:

• What was the reason for your visit?
• Did you find what you came for?
• How satisfied are you?
• Is there anything we can do better?

There are many ways to post a survey. If you already have a means to do this, what are you waiting for?

If you need a quick solution to try (and to complete this resolution), take a look at 4Q.

The fruits of a collaboration between Google Evangelist Avinash Kaushik and iPerceptions, it's free and specifically designed for this task (read FAQs on the 4Q website).

• 4Q helps answer these questions
• How satisfied are my visitors?
• What are my visitors onsite to do?
• Are they completing what they set out to do?
• What are they saying in the open-ended commentary?
  
  
• Open-ended commentary lets you hear the voices of your customers.
  
  
• Access your results through a secure Web interface.
  
  
• Visitors are asked permission to be surveyed. If they say yes, a separate window containing the survey is opened under the existing window. When the visitor closes the browser window, the survey will be visible.
  

• 4Q gives you control over whether you show the survey to every visitor or a lower frequency of your choice. And you can add your own logo to customize the survey initiation screen.

Analyze your most satisfied and dissatisfied visitors and their intent. This will provide you with hints of which content needs adjustment. You may not be able to jump to a solution in one step, but this will definitely be a start.

[ * Do a survey by the end March?
If it's too difficult to gain agreement to test on a a main site, you may want to try it out on a microsite or sub-site.
Some objections & arguments against asking website visitors for their input are "surveys irritate visitors", or that "surveys won't tell you anything useful".
Whether or not a survey is irritating or useful depends how you implement the survey and whether the survey's questions are appropriate for the survey goal. ]

Give it a try. You're sure to get some surprises and glean some new insights about your visitors.

June Li
ClickInsight
====..
...continue reading >