9 min read
Confidence interval in A/B tests: what it really tells you
What a 95% confidence interval tells you in an A/B test, how to read one that crosses zero, and why it beats a bare p-value for ship decisions.
Blog
RSS feedPractical articles on statistics, A/B testing and data-driven decisions for product managers and marketers.
9 min read
Now Next Later roadmap: how the format works, with an example and answers to "but when?"
How a Now Next Later roadmap works: what each column means, why confidence falls with distance, the disclaimer and how to answer "but when?".
9 min read
Outcome-based roadmap: what it is and how to build one, with a one-page example
What an outcome-based roadmap is, how it differs from a feature roadmap, its five components and a one-page example you can adapt.
8 min read
RICE prioritization: the formula, a worked example and the mistakes that break it
RICE prioritization explained: Reach × Impact × Confidence ÷ Effort, what each scale value means, a worked example with six ideas and common traps.
9 min read
A/B test sample size: how many users you need
How to calculate A/B test sample size from baseline, MDE, power and significance, with a worked example and the mistakes that waste weeks.
8 min read
Correlation vs causation in product analytics
Correlation vs causation for product teams: confounders, reverse causality, selection bias, and how to test causality with experiments.
7 min read
Mean vs median: when the average lies about your users
Mean vs median for product metrics: why averages mislead on revenue, session length and other skewed data, and how to choose the right one.
8 min read
P-value explained for product managers
P-value explained without jargon: what it tells you in an A/B test, what it doesn't, the most common misreadings and how to report results.