8 Ideas to improve your Banner campaign with A/B Testing

Optimization of your banner campaigns with A/B testing will improve the CTR and lower your Cost per Click (CPC).

A/B Testing

Counting on your first banner design to deliver is a huge mistake. So this is where A/B testing comes in. A quick recap - A/B testing is also known as split testing and refers to actively comparing results, for example, conversion/click through rates (CTR), for different versions of your landings page, direct mail or in this case, banner ads. Although A/B testing provides many advantages, in practice only a hand full of companies make use of it. We think that this is strange. Previously, marketing teams were unable to tap into the power of A/B testing because it required costly engineering and IT resources. Today, however, there are simple and cost-effective tools to create, manage, and optimize your banners.

We have compiled eight tips, which you can use right away to improve your banners with A/B testing. Most of these tips can be implemented today. So let's get started and improve your banners conversion rates this week.

1: Don’t just use the same old boring CTA text. Be specific, personal & convincing

You absolutely have to experiment with your CTA text. Actionable words such as “get” and words that imply little commitment such as “free” can have a significant impact on convincing users to proceed. Always put yourself in the customer’s shoes. Be personal, use the first person; “Start my free trial” instead of “start your free trial”.

Unbounce.com was able to improve their CTR by 90%, simply by changing the possessive determiner form “your” to “my”. Lastly, be very clear in what the benefits are to clicking your CTA. Stay away from meaningless phrases such as “Click here”.

2: Switching Background Color

There are several options for background colors. In addition to testing different shades of colors, test solid fills against gradient fills to find the background color that appeals most to your user. Always make sure that your CTA is contrasting well with your background color.

3: Create CTA Buttons in Green, Blue, Orange, & Red

Forget what you have heard. There is no ideal color for CTA buttons. To find out what works best, you won't get around a good old fashion A/B test. Feel free to test different shades of your CTA button but as a rule of thumb, stay away from low contrast colors like, white, black, and brown.

4: If you have multiple value propositions, test which one sounds most appealing to your customer

Even if you don’t have a second value proposition to offer, you can still modify the one you have to contain more or less information. Keep in mind not to pack your banner with text. Cut down on your message content and only include what is necessary to convey the added value you are offering and nothing more. A/B tests can help you find out what proposition is most valued by your target group.

5: Test product images against images of people

Product images are an excellent way to increase confidence in your product. If you don’t sell a physical product, test images for both genders. A case study published by Rocketfuel.com suggests that when presenting a picture, human faces increased the CTR by 4%. If you use images of people, try to arrange them, so that they are looking towards your CTA. Humans naturally follow the eye gaze of others as shown by KissMetrics. Take advantage of this and draw attention to where you want it.

6: Serif Vs. Sans Serif Fonts

Serif fonts are more decorative and appeal to an older and more serious group. Sans serif fonts, on the other hand, are evidently more casual and easy to read; this makes it an ideal choice for slow readers and the visually impaired. Legibility is of particular importance when choosing a font. Different target groups might individually prefer different fonts. Keep in mind what you are selling and whom you are trying to sell to.

If your product is cool, attractive, or exclusive, try using a serif font. If you have a product that is supposed to be easy and simple, stick to sans serif fonts which are easy to read.

7: Shape, Position and Interaction Effects for CTA Button

You want to choose shapes that resemble buttons that invite interaction. Bad choices for CTA shapes include shapes, which are not commonly associated with Buttons such as triangles. Create buttons with round edges to focus attention on the message.

8: CTA Position & Interaction Effect

Try different positions for your CTA. Keep the hierarchy of information in mind when placing your button. Placing your CTA at the top of your banner will make your CTA appear meaningless without context. A study conducted by VWO resulted in a 232% increase in CTR by just repositioning the CTA button and reducing clutter around it. Interaction or mouse hover effects can help your CTA buttons stand out as clickable elements and improve your click rate. Test with interaction-effect gives you the highest CTR for your banner campaign and quickly adjust according to your findings.

“How long should I run my A/B Test?”

After starting an A/B test you have to ask yourself, how long should I run this test before drawing conclusions? If you don’t run your test for long enough, your results just might be due to chance. Running it for too long on the other hand will make you miss out on valuable conversions. To know when your results are statistically significant, we suggest using tools such as Optimizely's sample size calculator or VWO significance calculator. If you want to read about an A/B test that was carried out from beginning to end, look into this case study regarding an A/B test conducted for Sony Europe.

“ A/B Testing with Bannerwise”

Previously, marketing teams were unable to tap into the power of A/B testing because it required costly engineering and IT resources. Instead of spending hundreds of euros on a single banner set, use Bannerwise to create independently and publish multiple banner sets at a fraction of the cost. Our tool will help you create our own cost-effective A/B test so you can start optimizing today.

Show Comments