Skip to main content

Compute the DiD Read Before You Believe It

Calculate a DiD estimate and state the parallel-trends assumption it relies on.

causal-inference5 min read

Canada gets a lifecycle email. Renewal rises from 72% to 78%. The US does not get the email and rises from 70% to 73%. Estimate incremental lift. Difference-in-differences: treated change minus comparison change, interpreted through parallel trends. The common trap is reporting the treated before-after change as lift. That ignores market-wide renewal movement. Step 1 Canada change = 78% - 72% = +6 points This is the treated before-after movement, not yet causal lift. Step 2 US change = 73% - 70% = +3 points This estimates common time movement if the US is a credible comparison. Step 3 DiD…

Read the full lesson

Sign up free: one personalized lesson every day, matched to your role and goals.

← Back to library

Contact us