Power Sizing for Rippling Onboarding Test

Business Context

Rippling wants to test a new onboarding prompt in the Rippling app switcher that encourages newly invited admins to complete payroll setup. Before launching, the Product Growth team wants to size the experiment so it has enough power to detect a meaningful lift without running unnecessarily long.

Problem Statement

Use the baseline conversion rate and target effect size to determine the required sample size per variant for a two-arm A/B test on a binary conversion metric. Then estimate how long the test will need to run given expected traffic, and assess whether the observed post-launch result would be statistically significant.

Given Data

Assume a two-sided test and independent Bernoulli outcomes.

Requirements

1. State the null and alternative hypotheses for the experiment.
2. Compute the required sample size per group to detect the target lift from 18.0% to 19.5% at 80% power and 5% significance.
3. Estimate the number of days needed to reach that sample size with the given traffic split.
4. Using the observed post-launch data, run a two-proportion z-test.
5. Compute a 95% confidence interval for the observed lift.
6. Conclude whether Rippling should treat the result as launch-ready evidence.

Assumptions

• Randomization is at the eligible admin level.
• No interference between units and no major traffic mix shifts during the test.
• Normal approximation is appropriate because expected successes and failures are large in both groups.