You're reviewing results from a growth experiment on a product surface and need to decide how to interpret the estimated lift, not just whether it passed a significance threshold. The team wants a clear read on uncertainty around the treatment effect.
How would you think about confidence intervals when evaluating a growth test?