Digital Marketing, Attribution, and Growth Experiments

Turn off your best-performing channel tomorrow. How many sales would you actually lose? Most marketers can't answer that — and it's the only question that matters. We spend against dashboards that credit a sale to whichever ad happened to be nearby, then call the correlation "ROAS." This course is about the harder, more useful truth underneath: which of your spending actually *causes* customers, and which is just taking credit for people who were coming anyway. You'll build the whole acquisition system. Start with the economics that decide everything — CAC, LTV, payback, the funnel and the loop — and a North Star worth steering by. Instrument it honestly: events, UTM tagging, conversion tracking, and the privacy reality nobody warns you about, where tracking is partial, platforms grade their own homework, and even the "cookies are dying" story turned out to be wrong. Work the channels — paid search, paid social, SEO, email, referral — and learn to read where each one actually breaks instead of judging it on one blended number. Then the heart of it. Every attribution model — last-click, multi-touch, the fancy data-driven one — is just a rule for splitting credit, and switching models reshuffles the winners without anything changing in the real world. So you'll learn what genuinely proves cause: incrementality tests, geo holdouts, and A/B experiments done right — significance, power, and the pitfalls (peeking, Simpson's paradox) that make most "wins" evaporate. No statistics background, no code. Just the difference between a number that looks good and a number you can bet a budget on.