You are planning an A/B test on a conversion funnel. The baseline conversion rate is 8.0%, and you want to detect an absolute lift of 0.8 percentage points to 8.8% with 80% power at a 5% two-sided significance level. Traffic is split 50/50 between control and treatment, and the experiment receives 40,000 eligible users per day in total.
How long should the experiment run to have enough sample size to detect the target lift with the requested statistical significance and power?