5 Ways to Get the Most From A/B Testing
You’re starting out with your brand into the World Wide Web, but how do you know what will work best and lead to more interaction with your viewers?The answer is… Experiments! Behold the glory of A/B testing!
Established websites and newcomers alike can benefit from the results that A/B testing can provide. Knowing what will work best with your viewers can be hard to gauge, but with a little help from the science of A/B testing you can use factual information to figure it out!
To be honest it’s not really science, but it just sounds more exciting than when you use words like ‘variant’ and ‘control’!
What is A/B testing?
In simple terms, A/B testing is setting up two versions of an element, eg a web page, with just one variation, and assessing which one gets the best results. In more technical language, A/B testing is a randomized experiment that compares your control site to a newer ‘treatment’ site. A/B testing is a form of statistical hypothesis testing when the web designer says “Hey, I think my viewers might be more inclined to read my blog/buy my product if my site were blue!” The designer creates a new blue site and gives it a try to see which site has better results.
I told you that science speak is exciting!
By using software such as Visual Website Optimizer, Kissmetrics or Google Analytics (free!), this testing allows your random viewers to be directed to Site A or Site B to evaluate the effectiveness of each.
What can A/B testing do for you?
A/B testing allows webpage designers to create metric representations of their projects as it is much easier to make decisions when there are numbers involved.
For example:
Site A has a 4% conversion rate, or 4% of viewers are participating in your call to action.
Site B has a 1.2% conversion rate, or only 1.2% of viewers are participating in your call to action.
With numbers describing the outcome of each site it is easy to choose the site that reflects better numbers while bringing you closer to your goal.
What attributes should be A/B tested?
According to Colten VanTussenbrook, blogger extraordinaire from Resell.biz, here are some commonly tested website elements:
- Color (could be an overall theme, or even just the color of the call to action button)
- Position on the page containing a call to action
- Different fonts, size, placement, etc
- Amount of text and wording
- Images
- Headings
Tips for making your A/B testing as efficient and effective as possible:
- Be bold.
Changing something as simple as a color or word choice can influence your readers to take action. Be willing to get creative and experiment with what works best for you and your audience.
- Invest yourself.
You might have to invest a little time, energy and cash to make the most of your A/B testing, but if well planned and executed the results will be worth it.
- Be patient.
Don’t make a decision based on a small accumulation of visitors, wait it out for definitive results. When considering a statistical experiment the more the merrier!
You want a site to have as much traffic as possible before investing time and money into A/B testing. You will have better results with more viewers.
- Try, try and try again.
Didn’t get the results that you expected? That’s okay! Make a new plan and try again!
Have more tips to add to our list? Send us a tweet @VPS
(which also boosts our conversion rate for our own A/B testing!)