The Art of A/B Testing - 9 Tests You've Never Tried
A/B testing is as much an art as it is a science. Creating the test, driving traffic, and then measuring the results is the science. But coming up with what to test, well that’s the art.
Depending on the context of your page and what you're offering, different tests make sense under different conditions. You’ll need to use your discretion to determine whether running a certain test is right for you.
If you’re just getting going, I’d recommend checking out my article on 10 Easy A/B Tests to Help Increase Conversions. But if you’re experienced with A/B testing and need a little inspiration on some new types of tests to run, then this articles for you.
Remember to be patient and always allow for a good amount of traffic to flow through your page before determining a winner. While there’s no magic number in this regard, try to give each variation a fighting chance by ensuring you hit at least 1,000 impressions before calling it a day.
A/B Testing Idea #1: Put your CTA below the fold
Conventional wisdom says to keep your CTA above the fold, but that might not always be the best for your page. Depending on what type of product or service you're offering, sometimes you won’t be able to communicate enough information to users above the fold. Especially for complex offers, test putting your CTA below the fold to give your users enough time to digest your content before making a decision to convert.
Content Verve tested this exact hypothesis on one of their landing pages.
What did they find?
They received a 304% increase in conversions for the variation with the CTA below the fold.
A/B Testing Idea #2: Autofocus on the form
Particularly if your landing page has a lot of content, try autofocusing on the form to see how that impacts conversions rates. What this means is that when a user visits your landing page, instead of loading the top of your landing page first, the page will automatically load to the location of the form to make sure users focus on exactly what’s important.
If you're creating your landing page on Wishpond you can simply select “autofocus on form field” in the form field settings to enable this option.
If you’ve coded your website up in HTML, you can find out how to set a HTML form field focus attribute here.
A/B Testing Idea #3: Remove everything below the fold
Finding the right page length is a fine balance. A general rule of thumb is to have shorter pages for simple requests such as an email address, and longer pages for decisions that involve a bigger commitment (for instance a purchase). This may not always be true however so test out different page lengths to determine what’s right for your audience.
By deleting everything except one core content area (in other words a page that doesn’t scroll), you can cut to the chase and find out what makes a difference on your page and what doesn’t.
A/B Testing Idea #4: Add two CTA’s right next to each other
You want to keep your landing page as focused as possible, therefore it only makes sense to have one CTA right?
Well, not always.
Best New Bingo Sites, on online gambling website, ran an A/B test where they compared two CTA’s against a variation with only one. What they found was that the dual CTA’s generated 13% more conversions!
What are some situations you use dual CTA’s next to each other?
Whenever you have a complicated offer that requires users both “learn more” and “sign up” on the next page.
When people submit your form, they need to feel as though they can trust you with the information they’re providing.
Variation 1: 100% Privacy - we will never spam you!
Variation 2: We guarantee 100% privacy. Your information will not be shared.
They found that the second variation generated 19% more conversions than the original sentence that contained the word “spam”.
People who are on the fence about converting are in a sensitive state. Text like the word “spam” can act as a red flag especially when someone is submitting personal information. Rather than thinking about the benefit of completing your form, people start to focus on the downsides (potentially getting spammed) and this reduces the overall number of conversions.
A/B Testing Idea #6: Add more navigational categories to your home page
When optimizing your homepage, sometimes clarity trumps simplicity when it comes to optimizing for your different goals.
Union Mission wanted to encourage more people to volunteer and donate so they A/B tested separating the “Give” label into three subcategories of “Donate”, “Ways to give”, and “Learn”.
Their hypothesis was that if they could be more clear with where they navigated their users, they could increase the total number of click-throughs and conversions.
A 300% increase in click-throughs and a 181% increase in overall donations.
A/B Testing Idea #7: Make two separate CTA’s on different parts of the page
If you have a long landing page with two CTA’s, rather than duplicating the same CTA at the bottom of your page, replace it with another one that says something different.
Secondary CTA phrases phrases can include:
- Learn More
- Get Started Now
- Let’s Go
A/B Testing Idea #8: Change your page based on the weather
We’ve all heard that personalization is the key to marketing. Part of that includes considering the context by which the user is viewing your page, i.e. what the weather is like at the time of their visit.
Use liquid code to A/B test a variation of your page that changes depending on the weather.
This can include offering different products, deals, and call to actions.
A/B Testing Idea #9: Change your form layout
When it comes to optimizing your form, there are other ways to test other than just adding or deleting form fields.
Try re-arranging the layout of your form vertically or horizontally to see how that affects conversions.
The company Arenaturist A/B tested this strategy and found their vertical form variation was able to convert 52% better than their horizontal form.
There you have it - 9 A/B tests to spark your imagination. Have a cool A/B test you’d like to share with the community? Post it in the comments below.