Data Analysis, Insights, and Reporting
This domain covers turning data into decisions. Know descriptive vs. diagnostic vs. predictive vs. prescriptive analytics; most marketing reporting is descriptive, but insight means explaining why and recommending what next. Understand segmentation of data, cohort analysis, trend vs. seasonality, and the difference between a metric moving and a metric meaningfully moving.
Know basic statistics: mean/median/mode, standard deviation, significance testing, confidence intervals, and the danger of small samples. Grasp A/B testing logic-hypothesis, control, sample size, statistical significance, and practical significance. Understand visualization best practices: choose chart type by data relationship (bar for comparison, line for trend, scatter for correlation).
Traps: presenting data without insight (a dashboard is not a recommendation). Confusing statistical significance with business importance-a tiny, real effect may not be worth acting on. Overfitting or cherry-picking time windows to tell a favorable story. Another trap: comparing averages across groups with very different distributions, or ignoring outliers.
Worked rule of thumb: Structure every insight as situation → finding → so-what → recommended action. If a test shows a 2% lift with wide confidence intervals and a small sample, the honest conclusion is "inconclusive-extend the test," not "winner." Report uncertainty alongside results.