← Back to Journey Map Phase 06 · Validation & Experimentation
Phase 06 · Validation & Experimentation
A/B Testing
Run controlled experiments with two variants to determine which performs better
Statistics / direct mail tradition, popularised digitally by Google · 1900s / 2000s ★ Must Know

A/B testing is a controlled experiment where two variants — A (control) and B (treatment) — are shown to randomly assigned user groups. Statistical analysis reveals which variant performs better with confidence.


When you have sufficient traffic to achieve statistical significance, and want to optimise a specific metric in an existing flow.


  1. Define the hypothesis: 'Changing X will improve metric Y by at least Z%'
  2. Define the primary metric and the minimum detectable effect
  3. Calculate required sample size for statistical significance
  4. Randomly assign users to Control (A) and Treatment (B)
  5. Run until required sample size is reached — never stop early
  6. Analyse: is the difference statistically significant? (p < 0.05) Then decide.

🎵 Spotify

Hypothesis: showing a context label under each recommended track will increase track completion rate by 10%. Control: recommendation without label. Treatment: recommendation with context label. Run for 2 weeks across 2M users. Result: +14% completion rate in Treatment, statistically significant. Ship globally.

📊 Trade Surveillance

Please contact the author for more information on these examples at linkedin.com/in/kshitijrege



Statistics / direct mail tradition, popularised digitally by Google 1900s / 2000s


← Return to Product Journey Map