Although descriptive statistics and correlations may help data scientists understand certain phenomena, causal inference is often key to business and product decision-making. However, finding strong causal relationships typically requires more focus on experimentation set-up. In this blogpost, Antoine Rebecq describes three methodologies that Shopify uses for causal inference including A/B tests, quasi-experiments (where the treatment and control group are divided by a natural, but non-random criteria), and counterfactuals (where the user creates a model to estimate what might have happened in the absence of the treatment). He discusses the importance of robustness checks and recommends using DAGitty to analyze causal diagrams.