Skip to content
Log in

Product Analytics: From Data to Decisions

Turn product data into decisions you can defend. Define metrics that don't lie, read funnels, retention and cohorts, design and read A/B tests, work out LTV and payback, and forecast growth, all on realistic data from a bill-splitting app.

Free32 lessons~12 h 9 min of learning

Start the course

  1. Module 1

    What exactly are we measuring?

    Events, properties and users; totals vs unique users and other counting traps; metric definitions two people can agree on; a tracking plan for a new feature.

    1. 1.1Events, properties and usersThe language of product data: what an event records and how event and user properties differ.~21 min
    2. 1.2Totals, unique users and counting trapsWhy the same week of data gives different numbers of 'active users', and how to count them right.~22 min
    3. 1.3A metric is its definitionWrite a metric so that two people get the same number: events, exclusions, window and time zone.~25 min
    4. 1.4A tracking plan for a new featureBefore payment reminders ship, plan the data: the questions, the events and properties that answer them, where each is logged, the metric definitions and the checks before launch.~26 min
    Module examComplete 4 more lessons to unlock

  2. Module 2

    Which numbers should we steer by?

    North star and the metric tree; volume metrics, quality metrics and ratios that mislead; DAU, WAU, MAU and stickiness at the product's natural rhythm.

    1. 2.1The north star and the metric treeOne metric for the value users get, and the inputs teams can actually move.~20 min
    2. 2.2Volume, quality and ratios that lieTell how much from how well, and catch a per-group ratio that moved because the mix of groups changed.~24 min
    3. 2.3DAU, WAU, MAU and stickinessCount active users at the rhythm people actually use the product.~23 min
    4. 2.4Case: a north star for HalvesPick and defend a north star metric for Halves, with six months of numbers behind it.~28 min
    Module examComplete 4 more lessons to unlock

  3. Module 3

    Where do new users get stuck?

    Funnels: order, conversion window and who counts; reading a funnel without fooling yourself; finding the activation moment.

    1. 3.1Funnels: order, window and who countsHow a funnel is computed and why the same steps can give different conversion.~21 min
    2. 3.2Reading a funnel without fooling yourselfRight censoring, date ranges and other ways a funnel misleads.~22 min
    3. 3.3Activation: the moment it clicksDefine the action that separates users who stay from users who leave, and know what that comparison can't prove.~21 min
    4. 3.4Case: fix the onboarding funnelFind where new Halves users drop off and what to fix first.~28 min
    Module examComplete 4 more lessons to unlock

  4. Module 4

    Do people come back?

    Day-N, rolling and bracket retention; reading and averaging a cohort table; the retention plateau as a product-market-fit signal.

    1. 4.1Retention: day N, rolling or bracketThree ways to measure retention and which fits a product people use weekly or monthly.~19 min
    2. 4.2Reading a cohort tableRead rows, columns and diagonals, and average cohorts without the classic mistakes.~19 min
    3. 4.3The plateau: retention and product-market fitWhat a flattening retention curve says about the product, and what it doesn't.~19 min
    4. 4.4Case: couples, flats and tripsIs Halves retention healthy for each kind of group? Read a year of cohort data by segment and write the verdict.~28 min
    Module examComplete 4 more lessons to unlock

  5. Module 5

    Did the change work?

    Hypothesis, primary metric and a decision rule up front; test size and duration; who is in the test; reading a finished test honestly.

    1. 5.1Hypothesis, metric and decision ruleWrite the test down before it starts.~19 min
    2. 5.2How big and how longSample size and duration for a product test.~19 min
    3. 5.3Who is in the testTriggered enrollment, dilution and randomizing whole groups.~19 min
    4. 5.4Case: reading a finished testEffect, interval, guardrails and the decision.~29 min
    Module examComplete 4 more lessons to unlock

  6. Module 6

    What can we learn without a test?

    Fair version comparisons; traffic mix and seasonality; correlation, selection and 'users who do X retain better'.

    1. 6.1Comparing versions fairlyThe rules that make a before/after comparison worth reading.~19 min
    2. 6.2Traffic mix and seasonsCompare like with like when the users themselves changed.~22 min
    3. 6.3Users who do X retain betterCorrelation, selection bias and what a feature's users can tell you.~19 min
    4. 6.4Case: the version or the traffic?Did the new Halves version help, or did the users change?~26 min
    Module examComplete 4 more lessons to unlock

  7. Module 7

    Are we making money on it?

    ARPU, conversion to paid and trial-to-paid; the LTV curve by cohort; CAC, payback and ROI by cohort and channel; pricing and paywall decisions.

    1. 7.1Revenue metricsARPU, ARPPU, conversion to paid and trial-to-paid, and how a change in the paywall trigger mix fakes a better paywall.~24 min
    2. 7.2The LTV curveCumulative gross profit per cohort user by month, how to read it, and when a young cohort can be extrapolated.~20 min
    3. 7.3CAC, payback and ROIJudge a paid channel by the cohort it buys: CAC, payback period and cohort ROI at a stated age, and why a month's profit divided by that month's spend flatters ads.~22 min
    4. 7.4Case: price or paywall?Read a price test and a paywall before-and-after honestly, then recommend what Halves Plus should do next.~28 min
    Module examComplete 4 more lessons to unlock

  8. Module 8

    How will we grow?

    Forecasting active users from cohorts; measuring virality honestly; channels, capacity and modern attribution; investigating an unexplained spike.

    1. 8.1Forecasting active usersNew users × retention: build a monthly active users forecast from cohorts, with a separate curve for summer cohorts, and price a cut in paid acquisition in users.~25 min
    2. 8.2Invites and viralityThe viral coefficient k, cycle time, why k below 1 still pays, and how to measure invites without flattering yourself.~21 min
    3. 8.3Channels and attributionChannel capacity and marginal CAC, a break-even cost that values the invitees a paid user brings, attribution through the stores' privacy frameworks, and incrementality from a geo holdout.~23 min
    4. 8.4Capstone: the unexplained spikeSign-ups more than doubled over a weekend. List the possible causes, rule them out one by one with the data at hand, and write the memo.~28 min
    Module examComplete 4 more lessons to unlock