MODULE 03 · Junior · 90 minutes

The discount lifted conversion — and brought in less money

Conversion moved from 5.8% to 6.4%. We work out whether that is a win or noise, and why the answer the business actually needs sits in revenue rather than in the p-value.

Open this module in the simulator All 23 lessons

The first two modules are free, no card. The rest are $20/month.

🏢 The situation

After your EDA report a debate broke out. Marketing: "Discounts increased conversion, we're awesome!" Max: "Or maybe conversion would have grown anyway — it's the season."

Lena:

"A classic. A battle of opinions is settled only by an experiment. Marketing already ran a test: 20,000 users were randomly split in half, group A saw a 15% discount, group B didn't. Conversion A = 6.4%, B = 5.8%. Marketing is shouting 'victory'. Your job is to check whether the difference is significant or just noise. And explain the result to Max in plain words."

🎯 Your task

  1. Understand basic statistics: distributions, mean vs median, variance.
  2. Learn how an A/B test works and what statistical significance is.
  3. Give the business an answer: does the discount work — and at what cost.

📚 Theory

Distributions — the language of data

The normal distribution is a symmetric "bell" (human height, measurement errors). Skewed distributions have a long tail (incomes, order values: many small ones, few huge ones).

Practical consequence: for skewed data the mean lies. An average order of $45 with a median of $12 means the "typical" order is $12, and the mean is pulled up by whales. Report the median and quantiles.

Hypothesis testing in plain words

⚠️ Frequent mistakes that get you flagged in review:

  1. The p-value is not "the probability that H₀ is true".
  2. "Not significant" ≠ "no effect". Maybe there's just not enough data.
  3. Statistical significance ≠ practical significance. A 0.01% conversion lift can be significant with millions of users — and useless for the business.

A/B test: the minimal checklist

  1. Randomization: users are assigned randomly (not "A — Moscow, B — St. Petersburg"!).
  2. The decision metric is chosen before launch (e.g. profit per user, not just conversion).
  3. Sample size is computed in advance (power analysis); the test is not stopped the moment it "became significant" — that's peeking and it breaks the statistics.
  4. For conversions (success proportions) use a z-test for proportions or chi-squared; for means — a t-test.
A/B test: is the difference in means signal or noise? B: 5.8% A (discount): 6.4% difference +0.6 pp the distributions overlap heavily → p-value = 0.07: indistinguishable from noise
The more the group distributions overlap relative to the difference in means, the higher the p-value: the observed gap is compatible with chance.

What to remember

  • Median and quantiles for skewed data; the mean is easily misleading.
  • p-value < 0.05 → reject "chance"; but significance ≠ business value.
  • Fix the metric and sample size before the test; peeking is forbidden.
  • The final argument is always money / the target business metric.

Next in this module: Practice

Open the module →

Nearby lessons

02 First task: exploratory data analysis 04 Predicting delivery cost

The whole program — 23 lessons