Posted on December 16, 2013 in Uncategorized by cmblogger
It’s easy to get carried away with the details of a web page redesign. Like a home remodel,
anyone can obsess over “cyber countertops” and the accent colors. Nevertheless, effective
consumer testing is critical for establishing what works and what just looks pretty.
First, make sure you implement both A/B and multivariate testing on the site. The difference
between the two? A/B testing requires dividing up your site traffic evenly between two
versions of the same page, and then evaluating which one works better in terms of
conversions or clicks. Multivariate evaluation, in contrast, tests for the effectiveness of
multiple elements (e.g. a call to action button, a headline) within the same web page. Make
sure you do both kinds of tests for your website to get the best data on what works well,
Second, remember it’s the little things. While most of your competitors will test for the big
things (basic layout, colors, headlines)on their websites, go the extra mile and test for the
little things too. You’d be amazed how differently customers react to the simplest changes
in a “contact us” form, for example. Don’t be afraid to try different versions and test for
Unlike a house makeover, precise data on a website’s effectiveness is available, and you don’t
even have to call a realtor. Get the right information to ensure your new website will really
work for you.
Posted on September 30, 2013 in Uncategorized by cmblogger
When it comes to designing the right website to get the customers you want, don’t take chances on gut feelings or worse, the personal preferences of your web designer. With A/B testing, you don’t have to. Test your site’s performance to optimize for success.
To evaluate your site effectively, however, you must identify the key goals that you’re trying to achieve. Without this information, you won’t be able to determine what works, what doesn’t, and how to make the changes necessary in getting you where you need to go.
First, who do you want as leads? Who are the site visitors you’re hoping to turn into customers? Remember, the more general your answer, the less likely you are to create a website that meets the specific wants of a particular demographic. Generic website designs designed to address everyone’s needs usually end up satisfying none.
Second, what do most of your consumers want when they get to your website-and what is the “essential” experience you would like them to have? For example, your company may sell more purple widgets off your site than anything else, and you may want them to buy blue widgets instead. Both these facts are critical pieces of information as you test.
Understanding the answers to these questions will allow you to effectively evaluate site content and refine for even more conversions. Evaluating what makes a good website isn’t just about looking at graphics or content. Base decisions on goals and hard data to ensure your success.
Posted on September 13, 2013 in Web Design by cmblogger
Everyone agrees that A/B testing is useful when it comes to determining what variable works and what doesn’t for a website design. In fact, one company recently demonstrated that this kind of evaluation can lead to an increase in sales of more than 35%. However, when you are getting ready to create your test, make sure you understand how and when to utilize this helpful tool. The wrong kind of testing can lead to faulty information and a design that spells disaster for your customers.
The A/B test is used exclusively to evaluate two different versions of one element of your web page. For example, you would use this test to evaluate whether you received more calls from a phone number printed in yellow at the top of your homepage, or if one printed in black was more successful.
However, when it comes to testing for several different issues, multivariate testing may be a better choice. With multivariate testing of your phone number color, font and placement, for example, you’ll create 8 different versions of the same page, assuming you’re choosing from only two options for each variable. Equal numbers of site visitors should be directed to visit each page in order to get the data you need to make a decision.
Consult your web designer to make sure your testing will answer the questions you’re asking. Armed with the right information, you’ll have a great website, with design choices based on hard data instead of guesswork.