Is A/B Testing Really Working for You?

If you visit FT, you’ll find that its subscription charges vary. At some point of time, it may be $4.99 a week. While the very next time, the same subscription might cost you $5.39, $5.75, $5.79, or $6.25. These variations are no glitches rather deliberate attempts by online marketers to gather valuable insight about their audience. Based on those insights, the best price is decided.

Over the decade, A/B testing has emerged as a principle method of website testing. Deriving its root from the days’ of postal mails, A/B testing has since emerged as the standard way of monitoring online services and consequently improving them. A/B testing is essentially a focus-group real time testing. During the course of testing, as with the FT example, a fraction of users is diverted to a slightly different version of a given web page and their behaviour is compared against the mass of users on the standard site. If the lesser version proved superior, the original version is phased out, gradually.

Today, A/B testing, in addition to Google (credited with first A/B testing ever),, Netflix, and eBay are all A/B addicts, constantly testing potential site changes on live (often unsuspecting) users.

Basically, A/B testing is jargon for a randomized experiment with two variants: A and B. After the test, either of A or B, based on the test results, is phased out as I mentioned earlier. The web trends, user’s behaviour all underwent several changes day to day. Users’ behaviour from the time of football world cup is totally different to the one during presidential elections. Thus, A/B testing is an ongoing process.

A has a better conversion than B. Thus, the tester sticks with A.

    1. Test The basis of e-mail

The elements that form the basis of an email are Subject, Headline, and CTA, layouts, and snappies.

Subject Line: A subject line alone decides whether the viewer will click over or not. If your viewers are reluctant to open your email newsletters, time to do research on those subject lines. Those lines are simply not attractive anymore and there is something needs to be done. Check different combinations of style and phrase in your subject lines and look for what is working.

Headline: After the subject, the success of your email campaign depends upon the headline. Try to grab eyeballs with something new, spontaneous, innovative, yet funny. A catchy headline almost guarantees a conversion.

See the effects? Stop the “six ways; seven ways” non-sense and add style to your writing.

CTA, layouts and snappies: A click on the “call to action” button simply marks a conversion. Run various sets of tests on “call to action”. Use various combinations of content, placements, colours for each test set. Not just the “call to action”, but also the accompanying content should tempt a user to click on the “call to action” button.

    1. Target returning visitors

There is thousands of website, which receives sudden upsurge of visitors during a phase, then collapsing all of a sudden, never to recover. This is definitely not a recipe for a successful website. Successful websites, on the other hand, should have a rich set of revisiting audience, visiting them on a regular basis, actively interacting with the community. In your tests, learn how many visitors are actually revisiting. Learn their behaviour pattern. What is making them comeback will give you a fair idea what factors make a user return.

    1. Set your own rules

“Xyz survey says that by keeping abc as the head line is the best way to increase conversions,” many people fall for this trap, make sure you are not one of the many. Believing in such craps is a sheer waste of time. If you’re short on time, implement changes that are unique yet effective, in a long run. Stop being a ship when you can be a tiger and mark your own territories. Make your own rules and, rule the world of online marketing.

    1. Make it happen

“Somewhere I read, A/B tests are simple waste of a money and time, is it so?” When 5 people out of 1000 fail to capture results, the phenomenon is called exception, not failure. If after so much of running and planning, you decided not to run those tests. Then you’re either a fool or coward. Either way, online marketing is not the place for you, give up. Nevertheless, any unrun campaign will only add to your, “why am I not getting conversions?” vows.  Not to mention if you don’t run the tests, you are not going to find anyways.

Therefore, run the tests and make sure you come up with a result. To attain great conversions, run A/B tests for each element, take action, and switch accordingly.

Original Published at:

(03) 9553 2824