Follow Along

RSS Feed Join Us on Twitter On Facebook

Get Engaged


Featured Article

Using Your Site Statistics: Getting a Bird’s Eye ViewUsing  Your Site Statistics: Getting a Bird’s Eye ViewHow well do you understand your website design? Ask yourself....Are the design, usability, and informational factors top...

Latest Comments

Latest Articles

How to Improve Conversions  Through A/B TestingHow to Improve Conversions...
What is A/B testing? A/B testing is a method of testing that allows one...
Dialing in Your Call to ActionDialing in Your Call to...
Throughout the course of any marketing campaign, it is necessary to count conversions as...
Evaluating Digital Marketing ROIEvaluating Digital Marketing ROI
Imagine investing $15,000 on a print-based ad campaign. Your company would want proof the...

Our Sponsors


Latest Search Videos

Join Us

How to Improve Conversions Through A/B Testing
Written by Brian Harnish
Monday, 06 May 2013 05:12

What is A/B testing? A/B testing is a method of testing that allows one to see what certain design changes on their website do to their conversion numbers. In order to see what changes are effective, A/B testing is necessary so that blind changes that lead to negative results do not occur. So, what are the benefits of A/B testing?

link soup or link soup

A/B testing allows you to see, with quite certain results, how changes on your website will impact your conversions. What one particular audience likes on one website is not going to be something that another audience likes. It is therefore necessary to pursue A/B tests that gauge what your audience prefers and how they are going to respond to different on page items.

A/B testing gives you a clear path to higher conversions. Conversions are no accident. Using A/B testing can tell you what your audience is most likely to click on. If they hate red buttons but prefer green ones, this will show up in an A/B test. If they like animated ads but hate still ads, this will show up in the A/B test as well.

Have Clear, Defined Goals

The most important part of an A/B test is to make sure that your goals are defined properly early on. This gives you a clear picture to start with that will, in the end, provide the most effective modifications that will benefit you in the future.

Without clearly defined goals, you will not be able to see the benefits of the exact actions that your audience takes on certain items on the website.

Say, for example, that you wanted to have a goal of higher conversions. We all want to see that, right? That kind of goal is too vague. You want to include specifics in your goal, like: I want to see an additional 25 conversions every week as a result of changing the button from green to red.

A specific goal you wish to achieve marks the starting point (although by no means is the full impact) of your A/B testing.

Always Use a Large Audience Sample

It’s not always possible to use a large audience sample size. If you use a sample size of only 50 visitors and find that they all respond to your recent changes positively, you may run into the situation where 499 other visitors dislike the new changes.

Therefore, it’s important to use a large enough audience size.  Evan Miller discusses the pitfalls of not calculating a proper audience sample size in his article, How Not to Run An AB Test.

One Step at a Time...

take a/b testing one step at a timeThis is a pitfall that you want to avoid at all costs. It is important to not test too many elements on a page. If you do, then you will find that it is even more difficult to figure out what change you made that had that positive impact.

To mitigate this issue, make sure that you only change one element. For example, if you want to see what happens if you change a particular piece of text (say, in the graphic of your call to action), then you will want to perform an A/B test that includes the original, and 1 variation of that original.

This will keep results clear and allow you to see, with great accuracy, the changes that your audience prefers and what they want to see happening on your website.

Analyze With a Fine-Toothed Comb

After the A/B test is complete, you will want to go through your analytics with a fine-toothed comb to find out what happened on your pages. Look at all of the important metrics of SEO analytics, not just a few.

Take a look at:

  • bounce rate
  • past conversions
  • how many new conversions took place
  • the amount of time visitors spent on the page as a result of that headline change
  • whether or not they converted on that page or went to another page as a result of that change

By looking at these additional analytics as well as the A/B test results, you can find out the exact impact that your change will have.

Make It So

Of course, if you find that your changes have a positive impact as a result of the A/B test, you will want to make those changes. Follow your results closely, and see how they impact your business.

It’s possible you may want to test these changes again. Perhaps different modifications will yield even better results. You never know for sure unless you make these changes and test to see how they are received by your audience. It’s also possible for visitor tastes to change seasonally, depending on your industry. This is why it’s important to keep testing, tracking your conversions, and making sure that your current changes still maintain their effectiveness.

Having clearly defined goals will make or break your A/B testing experiment. If you make changes randomly without testing to make sure that they are actually positive changes, you are shooting yourself in the foot and could end up hurting your site in the long run.

The mistake of not doing A/B testing has the potential to cost you a lot more than spending the time to make sure that these changes are positive changes. Through A/B testing of changes with a proper plan in place, it’s quite possible that the changes you make will provide you with dividends in conversions of website traffic.

Brian Harnish -

Building websites his Brian Harnish's passion. He's been building websites since he taught himself HTML and took a college Photoshop class to create the graphics in the summer of 1998 when he built his first real website. He thoroughly enjoys reading almost anything and everything he can get his hands on regarding SEO, search marketing, as well as web design & development.

Also hook up via


More articles by this author

SMX West Recap: Key Takeaways and Experiences SMX West Recap: Key Takeaways and Experiences
As a follow up to my conference survival guide...
An SEOs Guide to a Painless Website MigrationAn SEO's Guide to a Painless Website Migration
It can be a daunting and scary experience to migrate...
Last Updated on Monday, 06 May 2013 05:20

Add comment

Security code

Getting Around the Site

Home - all the latest on SNC
SEO - our collection of SEO articles
Technical SEO - for the geeks
Latest News - latest news in search
Analytics - measure up and convert
RSS Rack - feeds from around the industry
Search - looking for something specific?
Authors - Author Login
SEO Training - Our sister site
Contact Us - get in touch with SNC

What's New?

All content and images copyright Search News Central 2014
SNC is a Verve Developments production, the Forensic SEO Specialists- where Gypsies roam.