This is a guest contribution from Linda Bustos, director of merchant strategy with Workarea. A Demac partner, Workarea is a modern SaaS commerce platform purpose-built for medium to large businesses to achieve their full potential online.
The phenomenon is described as the “result of website usability tests where a majority of the test subjects either consciously or unconsciously ignored information that was presented in banners.”
This isn’t surprising. Our brains are good at filtering anything that competes with the task at hand. Leaderboards and skyscraper units are peripheral graphics we ignore without much thought as we scan for something of substance to click.
But hero banners aren’t typical ad units. They’re in-your-face, front-and-center. Do they really have the same effect on users?
Their prominence and dominance makes them difficult to miss, and they’re far more relevant to the customer’s site experience than third-party ads.
Sadly, it’s their animated behavior that makes them (consciously or unconsciously) perceived as ads by many.
“[Rotating carousels] are similar in size to MPU adverts and when coupled with the fact that carousels often “auto play” they become animated mimicking an advert. These cases result in situations where banner blindness means that the moderately experienced web user tunes out from such content because of the similarity to adverts.” Adam Fellowes
Web usability godfather Jakob Nielsen cited “banner blindness” a top reason one reason a user testing subject could not answer the question “Does Siemens have any special deals on washing machines?” despite “much time scrutinizing” the homepage, and the offer occupying both the largest home page real estate — and appearing again in a second content block.
“Most times carousels act as living and breathing friction on a page, distracting the prospect from completing their next step, they reduce conversion and cause high bounce rates.” — Talia Wolf, founder, GetUplift
3. They often slide too fast.
When banners advance too quickly, it hurts visitors who don’t read as fast, or are not native speakers of your site’s language. The more engaging your image and copy, the longer your customer lingers, irritating those most likely to take action. It can also be difficult to identify which “dot” marker corresponds to the slide you want to return to.
Conversely, the recommended five-second-hold-before-sliding gives users plenty of time scroll around and miss the rest of your banners.
4. They don’t always scale well to mobile.
When responsive sites scale down hero banners for mobile, text can become difficult to read, and images can lose context — sliding or static. Carousels add to the pain, as their controls (typically < and > arrows, or “dots”) can cover text and calls to action.
5. They can hurt SEO.
Optimizing sliders for search engines requires wrapping slider headings with h1 tags. The keywords within these tags tell search engines what the images are about, but too many keywords and h1 tags can devalue keyword relevance. Sliding carousels can also slow down page load speed which can impact your desktop and mobile rankings, as well as bounce rates (which is a ranking factor Google tracks in its algorithm).
Why you should pause your sliders and A/B test static hero banners instead.
Static banners are less distracting, less likely to be perceived as ads and more user-friendly across devices.
But you only get to show one banner per page load. A/B testing helps you identify your top performing banner, and gives you insight into the offers that appeal most to customers.
Why not just let ‘em rotate, and measure click through?
We know visitors don’t grab the popcorn to watch the slide show, and “first-slide-bias” will skew customer behavior. Erik Runyon found that among five sites he tracked, 89% of home page slider clicks were through slide one.
Sure, many conversion ninjas advise against A/B testing during the holidays. Testing during any traffic spike introduces behavioral variables that can produce less reliable winners — meaning your holiday winners may not perform well consistently.
But when your experiment is contextual to the holiday season, and your objective is to glean insights into holiday shopping behavior, it’s the most wonderful time of the year to test. Not only does higher traffic mean quicker test runs, it can also make the highest revenue impact.
Tips for testing home page heroes
1. Start with the right hypothesis
When A/B testing slides, consider testing offer-focused, merchandising-focused and guided-selling focused calls to action against each other. When graphic design complements the key message, but doesn’t compete with it, you can more reliably attribute click through to the content of the slide versus the imagery.
Offer-focused slide – no to minimal imagery
Merchandising oriented slide – imagery supports call to action
Guided selling oriented slide – imagery complements but doesn’t compete with copy
Alternatively, test different merchandising approaches. REI’s banners feature most-pinned, top rated and co-op exclusives. Testing these against each other can tell REI which approach is most appealing and persuasive. Learnings can be applied to future campaigns, both on-site and through social and email campaigns.
Exclusive products merchandising
Pro tip: take inspiration from your email tests. Many online retailers run successful email split testing and neglect to apply the same process to their home page banners. Get into the habit in 2018!
2. Avoid micro-merchandising
Banners occupy a lot of home page real estate. “Micro-merchandising” (single product features) is risky. The larger your catalog, the lower the likelihood you’ve selected the most appealing and highest performing product to dedicate the main hero to.
Instead, take advantage of general, mass-appeal offers like sitewide discounts, shipping offers, gift finders, sale categories and top-rated lists.
3. Consider the impact of imagery
Banner images and calls to action are very different variables that impact attention, interest and action. It’s difficult to attribute a winning banner to the offer or the design. For this reason, consider using the same image to represent different offers, or keep imagery similar in style.
4. Measure the right metrics
Don’t just measure click through and conversion, but also revenue. Promoting sale items versus new or top rated gifts (for example) could have the same conversion, rate but very different revenue contribution.
5. Know what to do with your results
During the holidays, it’s often necessary to change banners frequently, from Black Friday offers, to Cyber Monday, Cyber Week and beyond. If you’re planning to make rapid changes, limit your testing only to periods where the duration of the banner rotation’s traffic is enough to drive a statistically significant result.
You may not have time to apply your winner for long during your peak season. Never fear, you should still test. Focus on the insights you receive from the test for future banners, campaigns and holiday merchandising. At the very least, pausing your sliding banners will give you a usability boost for holiday shoppers.