theonline.mba
← Back to ENT 402

ENT 402 · Unit 2 of 6

Experiment Design and Learning Loops

Product-Market Fit and Startup Experimentation

Start unit · 4 lessons →

Learning objectives

  • Design experiments with clear success and kill metrics
  • Apply "Experiment Design and Learning Loops" to a real venture decision
  • Contribute to your Cohort analysis dashboard deliverable

Unit overview

#LessonCore idea
1Understanding Experiment Design and Learning LoopsCore frameworks for this unit
2How Experiment Design and Learning Loops Works in PracticeCore frameworks for this unit
3Evaluating Trade-offs in Experiment Design and Learning LoopsCore frameworks for this unit
4Experiment Design and Learning Loops: Case Analysis and RecommendationsCore frameworks for this unit

Complete all four lessons, then finish unit assessments on this page.

Unit assessment

Complete each section below. Score 80%+ on the quiz to finish this unit's assessment.

50% applied project30% case work20% knowledge checks

Exercises

Apply what you learned in this unit with structured practice.

ExerciseApplied practice: Experiment Design and Learning Loops45 min
Complete a focused practice exercise on **Experiment Design and Learning Loops**. 1. Choose a real company, product, or situation you know. 2. Apply one core framework from this unit to analyze it. 3. Write your analysis in 300–500 words with a clear recommendation. 4. Cite at least one credible source.

Deliverable

300–500 word analysis document saved to your portfolio under ENT 402.

Rubric

  • Framework applied correctly (not just named)
  • Specific evidence from a real example
  • Clear recommendation with tradeoffs acknowledged
  • Professional writing with source citation
ExerciseDrill: Experiment Design and Learning Loops30 min
Work through the practice problems in the unit lesson without looking at notes. Then check your work against the lesson and write a short reflection: - What you got right - One mistake you caught - One concept to review before the next unit

Deliverable

Problem solutions + 150-word reflection in your ENT 402 workbook.

Rubric

  • Attempted all practice items before checking answers
  • Honest reflection on errors
  • Identifies a specific review action

Case analysis

Analyze a case using frameworks from this unit.

CaseCase analysis: Experiment Design and Learning Loops60 min
Analyze a real business case through the lens of **Experiment Design and Learning Loops**. Choose a public company event, HBR-style case, or documented decision. **Deliverable structure:** 1. Situation summary (150 words) 2. Analysis using this unit's frameworks (400 words) 3. Recommendation (150 words) 4. Risks and what would change your mind

Deliverable

2-page case write-up in your portfolio.

Rubric

  • Case facts are accurate and sourced
  • Analysis uses unit frameworks explicitly
  • Recommendation is justified with tradeoffs
  • Risks are specific, not generic

Knowledge quiz

Check your understanding before marking the unit complete.

1. RelayOps designs a four-week pilot with one primary metric (median emergency dispatch time) and pre-registered success and kill thresholds. What is this experiment's core purpose?

2. Maya runs a concierge week where she manually routes five emergency jobs behind a Google Form intake before software ships. Which learning-loop principle does this apply?

3. RelayOps tracks three pilot sites staggered over four weeks each. Week-two median dispatch time is 11 minutes versus a 12-minute baseline. What is the best interpretation?

4. Jordan wants to change both onboarding copy and routing logic in the same two-week sprint. Why does the experiment design lesson advise against this?

5. RelayOps defines success as median dispatch under 5 minutes on 20+ live emergency jobs with 80% daily active dispatchers. Pilot A hits 4.8 minutes but only 55% daily active. What is the correct call?

6. A learning loop retro lists: hypothesis, metric, result, decision, next test. RelayOps's result is 'median 6.2 minutes, DAU 72%' after four weeks. What decision type fits a pre-committed kill rule of no improvement?

7. RelayOps spends two weeks per learning cycle. Over a 12-week quarter with one cycle at a time, how many full experiments can complete if each needs two weeks of measurement after a one-week setup?

8. An investor asks RelayOps to run paid LinkedIn ads during the pilot to 'accelerate learning.' Why is this usually a poor experiment trade-off pre-adoption?