Testing Sticky Ads

As advertisers began to measure and pay based on the viewability of their ads, we had the challenge of raising the ad viewability rates sharply and quickly to meet the standard threshold of 85%.

Viewability Methods Employed

Ad viewability is defined as at least 50% of the ad unit being in the user's frame of view for a minimum of 1 second. (30% of the ad unit if it is a larger sized ad creative.) There are a number of proven techniques we tested in pursuit of our goal.

Sticky Ads

Sticky ads remain in view as the pages scroll. While they vastly improve viewability, they may result in a negative user experience, hence the need for extensive testing before implementation.

Lazy Loading

Lazy loading ensures the ads only load once the ad slot is in the user's frame of view. This does mean far fewer ad impressions, but guarantees quality ones. We implemented this on all BTF units.

Reducing Latency

One of the biggest reasons for ads to not meet viewability metrics is slow loading times. We made adjustments to the ad call orders, pass-backs, and stripped out extra code to reduce any latency.

Design Solution

Balancing the advertisers' viewability requirements with the need to provide a frictionless user experience required extensive testing. We ran multiple A/B tests with a variety of designs. Measuring time on page, pages per visit and bounce rate, we were able to pinpoint the most successful variation to satisfy all of our requirements.

Unobtrusive but effective:
Increase from 50% to >90% viewability

As users scroll, the top ad also begins to scroll, but at a slower rate than the rest of the content. This allows the user to have the experience of being able to control the ad, while locking in the viewability. After the user has scrolled down once, the ad no longer sticks, providing ad-free subsequent scrolls.

View this ad unit in action here.

Thanks for stopping by!

See Next Case Study: Bopsidy

Julie Baum

UX Designer, San Francisco

ROTTEN TOMATOES