Illinois Policy Institute

How site flow interruptor offers affect conversion on mobile devices

Experiment ID: #2722

Illinois Policy Institute

Experiment Summary

Timeframe: 06/06/2017 - 06/16/2017

On the mobile experience, the slide-out offer produced a 91.6% lift in conversion over the exit intent popup. This is primarily caused by the way the two site flow interruptors display on mobile devices—the slideout actually functions more like an exit intent, taking up the entire screen. Since exit intent popups require a cursor, that experience is replaced with a very small offer bar at the bottom of the screen which is easier to ignore. While the exit intent popup produced a lift on desktop, this test result shows us that we cannot assume that the same tactic will work for different experiences.

Research Question

Which site flow interruptor will maximize conversion on mobile devices?


C: Exit Intent
T1: Slide Out


 Treatment NameConv. RateRelative DifferenceConfidence
C: Exit Intent 0.11%
T1: Slide Out 0.21%91.6% 99.5%

This experiment has a required sample size of 12,090 in order to be valid. Since the experiment had a total sample size of 49,053, and the level of confidence is above 95% the experiment results are valid.

Flux Metrics Affected

The Flux Metrics analyze the three primary metrics that affect revenue (traffic, conversion rate, and average gift). This experiment produced the following results:

    0% increase in traffic
× 91.6% increase in conversion rate
× 0% increase in average gift

Key Learnings

Illinois Policy Institute wanted to promote their new eBook, Madigan’s Rules to their website visitors. However, they wanted to match the right promotional tactic to the correct audience segment. They had previously tested slide-out offers and exit intent popups, but knew that the two tactics presented differently on mobile and desktop.

They did a split test between the two offers and closely monitored traffic and conversion from desktop visitors in contrast to traffic and conversion from mobile visitors to match the ideal “site flow interruptor” to to the optimal experience.

Experiment Documented by NextAfter

Question about experiment #2722

If you have any questions about this experiment or would like additional details not discussed above, please feel free to contact them directly.