Building Progressive Web Apps

Appendix B. Full-Page Interstitials or: How I Learned to Hate the Door Slam

In what some might call a desperate effort to get more app installs, many sites have resorted to using full-page interstitial ads to hide away their entire site behind an ad for their mobile app.

A lot of research has been done showing how much users hate the door slam. I won’t even waste your time with a link to this research. You probably already know the answer instinctively. If not, you are welcome to visit the “I Don’t Want Your F***ing App” Tumblr page.

But let’s ask a different question. Are full-page interstitial ads effective?

In 2015, Google decided to run an experiment to answer this very question. When Google released the results of its experiments with full-page interstitials, the answer was pretty clear:¹

• Only 9% of users presented with a full-page interstitial clicked the “Get App” button (remember, this is just the first step in the installation funnel).

• 69% of users immediately abandoned the page as soon as the interstitial opened. These users neither went to the app store nor continued to the website that was just a click away.

After seeing these numbers, Google decided to run an experiment and see how replacing the interstitial with a small, unobtrusive app banner would affect actual product usage. The results were surprising:

• 1-day active users on the mobile website increased by 17%.

• Native app installs went down just 2%.

Based on this, and other experiments, Google decided ...