Here’s a stat for you need to munch on.
“Poor A/B testing methodologies are costing online retailers up to $13bn a year in lost revenue” (Qubit)
And though this stat is specific to online retailers, A/B testing woes are generalizable to all marketers.
Whether you run an e-commerce store, a digital marketing agency or an insurance business, everybody is looking to increase conversions without going over budget and A/B testing is the perfect way to both stay on a budget and increase your conversions.
A/B testing is not an unheard of practice; every marketer out there knows what A/B testing is – most marketers don’t fail at A/B testing because they don’t know what testing is, but they fail at testing because they don’t follow the right procedures for it.
This is going to change today.
Outlined in this post are exact steps you need to take when you start A/B testing your website no matter what industry or niche you belong to. Armed with this outline not only will you be able to better understand how A/B testing works but you’ll also get ideas of what to test on your website.
Let’s start by giving a brief description of what A/B testing is.
Here’s how Econsultancy defines A/B testing:
“A/B testing involves comparing a change to a website against the original with consumers on a live website. The sales impact of the change can be measured by comparison to the performance of the original.”
Essentially A/B testing involves you creating a variation of your original website and sending traffic to both variations at the same time, and after some time you find out which variation your visitors responded to the most.
Sounds simple enough right?
Well, yes and no.
Though the process of A/B testing is simple if you have the right tool by your side, don’t rush off to test your websites just yet, check if you have everything on the outline all settled.
You need to make sure you have everything from this outline all squared before you start creating headline variations.
It’s smart to do customer research before you start your business. By all means create buyer personas and find out the preferences of your potential clients before you create your website. However, when you’re A/B testing, you need to forget everything about your target customers, so that you don’t make any assumptions about them based on your research.
Assumptions limit your creativity and your A/B tests success depends on your creativity.
You don’t start a race without knowing where the finish line is, right? Similarly, you shouldn’t start off with your A/B test without knowing when to stop. You need to tick off an end point before you begin the testing process.
It’s easier to do something that’s already been done, however, CRO results are not repeatable because every website and the visitor clicking on it is different.
Just because Sally increased her click-through rate with a red CTA button doesn’t mean that a scarlet button is the right choice for you. Learn from your mistakes, experiment what works on your website instead of blindly following best practices.
Don’t get yourself all hot and bothered by testing too many elements at a time – it’s called ”A/B testing” people, not A-Z testing. Testing one element at a time ensures efficacy of your results, because if you tested your CTA button color and copy at the same time how will you find out the real cause of your increase in conversions.
Stick to one element at a time and you’ll do just fine.
Statistical significance seems like a big word, but it embodies a very simple concept. Statistical significance relates to the ability of your test results being generalizable. For example, if you’re A/B testing your headline and only 7 visitors click onto your website in your set time frame for the test, this test doesn’t have statistical significance because 7 visitors is not a good population size. Keep your population size sizable to achieve accuracy for your A/B tests.
Don’t start A/B testing your website just because everyone else is doing it. Wait for your website to start getting considerable traffic and then begin testing because otherwise you won’t get accurate results.
If you are getting below 2, 000 unique visitors a month you shouldn’t be testing just yet.
I know there are testing case studies online that show whopping increases in conversions, but you need to manage your expectations. Don’t expect to get a huge increase in your conversion rate from all your tests.
Remember you need to learn to make your own mistakes, however here is some A/B testing case study inspiration.
Your website headline needs to be clear and inviting. It should be able to entice your visitors into going through the rest of the website and then taking action at the end. Here’s a headline formula you should use when crafting headlines.
Monthly1k, an AppSumo course to help start a web business, conducted a headline A/B test.
The winning variation was variation A increasing sales of the course by 6.49%. The headline is clearer than variation B as it puts the reward of the course in a dollar value.
The lead capture form on your website needs to be designed properly in easy to fill out fields and shouldn’t ask your visitors for information that is not relevant to your offer.
Just Rewards tested the design of their lead capture form; these were the two variations created.
Your CTA button needs to be contrasting and have clear personable copy on it, for more information on how to create click-worthy buttons go here.
Amsterdam Printing conducted a button test where only the button copy was tested, these were the two variations created.
Variation A was the winning variation with 45% lift in revenue. Why did visitors respond to the “Personalize Now” button more? Perhaps because it shows urgency which variation B copy does not.
A/B testing is not hard if you follow the right steps. Don’t be afraid to experiment with your page elements to see which one is the winner.
Still unsure about what to test on your website? Contact us and we’ll help you out.