Warning: file_get_contents(): php_network_getaddresses: getaddrinfo failed: Name or service not known in /home/crdoc/public_html/conversion-blog/wp-content/plugins/digg-digg/include/dd-class.php on line 1745
Warning: file_get_contents(http://feeds.delicious.com/v2/json/urlinfo/data?url=http%3A%2F%2Fwww.conversiondoctor.com%2Fconversion-blog%2Fusability-split-test-results-link-appearance-matters-more-than-you-think): failed to open stream: php_network_getaddresses: getaddrinfo failed: Name or service not known in /home/crdoc/public_html/conversion-blog/wp-content/plugins/digg-digg/include/dd-class.php on line 1745
As I mentioned in an earlier blog post, most Usability experts don’t conduct actual split testing.
Usability testing is important, but if you want to optimize your landing page for a particular ACTION, then split testing or multivariate testing is necessary.
Over the past few years I’ve put MANY usability “best practices” to the test in an effort to ensure that the popular usability dogma is indeed accurate.
Over the next few months I’ll be sharing the results of some of these tests here on my blog (but many of my findings are simply too good or controversial to share with the “general public” so those will still be reserved for my paying clients/audiences.)
Today I want to share some data on link appearance that actually DOES support conventional usability wisdom.
(Although as you will see, many if not most web designers routinely ignore this wisdom.)
For several years I have counseled my clients to make every effort to keep most of the links on their websites blue and underlined.
The reasoning for this is simple… Blue, underlined links are what most people (particularly those new to the web) are accustomed to seeing. Everyone with even a few minutes of online experience automatically knows that if text is blue and underlined, it’s a link. (Or at least it SHOULD be a link.)
But until a year ago, I never actually split tested this to make sure that this nugget of “common wisdom” was actually true.
I decided to conduct a scientific test to see which links actually receive the highest click through rates.
I tested 4 different link formats:
Not underlined, not blue. (But matching the color scheme of the website and other link colors.)
Underlined, not blue.
Blue, not underlined.
Before sharing the test result data, I think it’s important that I share a bit of information about the website, niche and audience with which the test was conducted.
1. The action being tested/optimized for was the download of a free report.
2. The website tested did not use a squeeze page or forced opt-in page, it simply displayed a download like that said: “Click Here to Download Your FREE Report!”
3. No other variables were being tested at the same time. Only the link appearance.
4. The niche was NOT an internet marketing niche. The niche was a “hobbyist” niche targeting an audience composed of 80% women who were non-tech savvy (the technographics of your audience are important to understand when testing.)
5. The only traffic source for testing was Google PPC traffic.
Why do I share all of this data prior to showing you the test results?
Because, test results do not happen in a vacuum. As much as I appreciate other marketers and testers sharing their data and results, often the data is suspect or meaningless without at least some of the above information.
Different audiences respond to test variable differently. Just because one audience in one market responds positively to a particular variable, it doesn’t mean that those results are automatically “valid” for another site, in a different niche. (Or even for the SAME site, using a different traffic source!)
Anytime you see anyone sharing test results, thank them, but take their recommendations with a large grain of salt until you can validate their findings with your OWN testing.
For any test results to have real value and meaning, you must understand some of the “background” information about that test.
Okay… So now that is out of the way here are the test results:
Not underlined, not blue. ““ 26.9% CTR
Underlined, not blue. ““ 32.5% CTR
Blue, not underlined. ““ 31.8% CTR
Blue, underlined. ““ 36.3% CTR
Even though this particular test was conducted with an audience consisting of 80% non tech savvy women, I’ve seen similar numbers and behavior in many different niches and demographics.
As you can see, the results of this test were about what you would expect and they confirm the “rule” of sticking with blue underlined links wherever possible.
However, there was one unexpected (and eye-opening) finding.
Notice how “underlined NOT blue”, slightly out performed “Blue not underlined”. This says to me that underlining is an even more important usability factor than using blue links.
I did not expect that…
But of course the old standby combination of blue AND underline is still the best.
Bottom line… If your site has been designed by one of those web designers that love using CSS to remove the underlining from your links, simply to make things look “clean and neat”, switch them back to “blue and underlined” and you should see a nice jump in the click through rates on your links and deeper site penetration.
P.S. ““ Keep an eye out for another post in a few days where I’ll share some additional (and profitable data on “link behavior”…)
About the AuthorEric Graham is a serial entrepreneur, author, speaker, copywriter and consultant. Enter your name and email address below to get notified when new response boosting tips, tested conversion strategies, updates, articles and videos are posted.
You'll also get immediate access to a powerful 90 minute Conversion Boosting video as a free gift for signing up (and a lot more!)
And while you're here, go ahead and follow Eric on Twitter!
Claim Your FREE Conversion
Fill out the form below and you'll immediately receive (via email) full instructions on how to watch this powerful recording!
- What in the Heck is “Secondary Link Feedback”?! (And how can it help you sell more?)
- Submit Button Usability Split Test Results: Size DOES Matter!
- Can an Affiliate Link Invalidate Split Test Results?
- Link to Free “Conversion Rate Optimization” Webinar Recording (only for a limited time…)
- Checkout page split test results… (Product descriptions matter!)