While there are several factors that go into launching a successful retargeting or display advertising campaign, none is simultaneously so critical and so widely overlooked as banner design.
Yesterday, we were lucky enough to participate in a webinar with the folks at Clever Zebo as a part of their magnificent web series, the Path to Landing Page Genius. Though typically focused on landing pages, also a critical part of any retargeting campaign, yesterday’s webinar focused on a topic near and dear to ReTargeter: banner design and best practices.
While examining the results of several banner ad A/B tests, three principal takeaways emerged, and they may not be what you think.
1. Avoid distractions
One of the most common mistakes in banner design is over-inclusiveness. Often, advertisers will attempt to jam pack as much information as possible into a single ad in a misguided attempt to get the most out of each impression. Prima facie, this may appear to be a good tactic, but attempts to include too much information often end up conveying no meaningful information at all.
With any banner ad, you have only a moment to grab the attention of your viewer and make an impression. Additionally, banner ads offer extremely limited real estate. Packing information in tiny, unreadable text will neither effectively relay that information nor make a lasting impression. Including too many images or overly complex graphics will have the same negative effect.
Due to limited space and the small window of opportunity to catch the viewer, your goal should be to make a lasting impression. Eye-catching backgrounds should be paired with short taglines that explain your product or value proposition as effectively as possible. There can only be one call-to-action, and it should be simple, declarative, and bold.
2. Use photos only when appropriate
It is often assumed that ads featuring a photo of a person will always perform better than ads without a photo. Sometimes this is the case, but just as often, photos can backfire.
The mere inclusion of a person’s photo far from guarantees a high-performing ad. Haphazard inclusion of seemingly arbitrary photos will do nothing to explain your product and is another of the most common mistakes made in banner design. It is often assumed that, because you have such a brief window to secure attention, ads without photos will be overlooked, but a well-designed ad with a brightly colored background or simple artwork can also do the trick. In fact, bright colors and vibrant artwork are often more eye-catching than a photo.
However, when appropriate, photos can help the viewer to feel more of a connection to your product or the benefits it provides. This is best accomplished if you can show someone using your product or service. Given the limited space available within a banner, and depending on your product or service, this use case may not be a viable option. For example, at ReTargeter, there is not a good way to show our desired audience, marketers, using our online advertising products. You may also be able to use photos to express a benefit of your product or service, such as convenience or peace of mind. If you can effectively demonstrate your benefits using a photo, then it can be a highly effective tactic.
3. Keep your tests simple
When it comes to A/B testing, you might want to get as much information as possible out of each test, but including too many variables will create noise and lead to inconclusive results.
Within a single test, only change one aspect of the banner at a time. For example, within a single test you could compare two different taglines while keeping everything else (background, call-to-action, etc.) consistent. Conversely, in another test you might change the background image while keeping the taglines and call-to-action consistent. Attempting to compare both within the same test will not tell you much about why the winning version won and will not give you much to build on next time you create new banners.
If you have a high volume of impressions, and are thus gathering quite a bit of data, you can run multivariate tests to compare multiple variations at once. However, multivariate testing is significantly more complicated than A/B testing and requires more work on your part.
The bottom line: keep you banners simple, beautiful, and readable, don’t include any elements that are not strictly necessary, and don’t over-complicate your testing. In short, keep it simple.