You work on a consumer-facing product team that launched a marketing campaign and now needs to know whether it actually moved the business. A clean A/B test was not set up, so the observed results may be confounded by seasonality, channel mix, or other concurrent changes.
How would you evaluate whether the campaign is working if you can’t run a clean A/B test? Explain how you would separate signal from noise, what evidence you would trust, and what could make you wrong.
