Skip to main content
In contrast to one-time purchases, alterations in a recurring revenue model can influence an app’s financial health for extended periods, ranging from weeks to years. Even minor modifications can accumulate over time, resulting in notable shifts in revenue. Therefore, when experimenting with Qonversion, you might consider halting the exposure of new users to the A/B test without ending the experiment altogether. We advise using the Pause button if this situation applies to you. By doing so, Qonversion:
  • Continues to send variant-related data to your app for users already involved in the test.
  • Sets the traffic allocation percentage to zero, ensuring the A/B test impacts no new users.
Keep in mind: devices that fetch a config while the experiment is paused are evaluated and recorded as excluded. When you resume by raising the traffic level back, those users are not re-admitted — only users the experiment has never evaluated join the test. See Changing experiment traffic mid-flight.
What’s Next