Skip to content
Log in

Statistics for Product Managers and Marketers

Stop guessing from dashboards. Learn to read metrics honestly, run A/B tests you can trust and explain the numbers to your team, all on realistic product and marketing data.

Free24 lessons~12 h 27 min of learning

Start the course

  1. Module 1

    Describing data

    Typical values, spread, outliers and charts: how to summarise any product metric honestly.

    1. 1.1Typical value: mean, median, modeMean, median, mode and trimmed mean on real-looking CatChow orders: what each one answers, why one bulk order can fool the average, and how not to average averages.~36 min
    2. 1.2Spread: range, SD and quartilesTwo couriers with the same average delivery time, and why only one keeps the promise: range, variance, standard deviation, quartiles, IQR and the coefficient of variation.~38 min
    3. 1.3Outliers and skewed dataWhy product metrics have long tails, how to find outliers with the IQR rule, and what to do with them: fix errors, segment different customers, and never delete whales without thinking.~33 min
    4. 1.4Charts: choosing the right oneHistograms, box plots, bar, pie, scatter and line charts: which question each one answers, how bin width and axes change the story, and the classic ways charts mislead.~40 min
    Module examComplete 4 more lessons to unlock

  2. Module 2

    Comparing two groups

    From samples to significance: when a difference between two groups is real and when it's noise.

    1. 2.1Samples, populations and the normal distributionWhy we study samples to learn about all users, and why the bell curve shows up everywhere.~32 min
    2. 2.2The t-testComparing the means of two groups and deciding whether the gap is bigger than chance.~31 min
    3. 2.3Mann–Whitney: when data isn't normalA rank-based test for skewed metrics like revenue and session length.~33 min
    4. 2.4p-value and significanceWhat a p-value really says, what it doesn't, and statistical versus practical significance.~28 min
    Module examComplete 4 more lessons to unlock

  3. Module 3

    A/B testing in practice

    Confidence intervals, power, conversion tests and the pitfalls that break real experiments.

    1. 3.1Confidence intervalsReporting a range of plausible values instead of a single number.~30 min
    2. 3.2Power and sample sizeHow many users an experiment needs to detect the effect you care about.~26 min
    3. 3.3Conversion rates and proportion testsComparing conversion rates between variants and reading uplift correctly.~31 min
    4. 3.4Pitfalls: peeking, multiple testing, novelty effectThe most common ways real A/B tests go wrong, and how to avoid them.~27 min
    Module examComplete 4 more lessons to unlock

  4. Module 4

    Many groups and before/after

    ANOVA, interactions, paired tests and repeated measures for comparing more than two groups or the same users over time.

    1. 4.1ANOVA: comparing many groupsTesting three or more variants at once without inflating false positives.~30 min
    2. 4.2Multifactor ANOVA and interactionsWhen two factors act together, e.g. a promo that works on mobile but not on desktop.~29 min
    3. 4.3Paired tests: before and afterComparing the same users before and after a change.~29 min
    4. 4.4Repeated measuresTracking the same users across several points in time.~28 min
    Module examComplete 4 more lessons to unlock

  5. Module 5

    Relationships

    Correlation, causation and regression: how metrics move together and how to predict one from others.

    1. 5.1Correlation is not causationMeasuring how two metrics move together, and why that alone proves nothing about cause.~35 min
    2. 5.2Linear regressionPredicting one metric from another with a straight line, and checking the fit.~30 min
    3. 5.3Multiple regressionSeveral drivers at once: separating the effect of price, channel and season.~26 min
    4. 5.4Logistic regression and classificationPredicting yes/no outcomes such as churn or conversion, and measuring how good the predictions are.~32 min
    Module examComplete 4 more lessons to unlock

  6. Module 6

    Modelling and segmentation

    Forecasts, clustering users into segments, finding survey drivers, and choosing the right test.

    1. 6.1Models and forecastsWhat a model is, how to forecast a metric, and how to tell whether the forecast is any good.~29 min
    2. 6.2Clustering for user segmentationLetting the data suggest user segments instead of guessing them.~31 min
    3. 6.3Factor analysis: surveys and NPS driversFinding the few hidden themes behind many survey questions.~31 min
    4. 6.4Course recap: choosing the right testA decision guide that ties the whole course together.~32 min
    Module examComplete 4 more lessons to unlock