Summary

The Lean Startup methodology starts with a MVP, tests assumptions, and continuously improves through the build-measure-learn cycle. Smoke test ideas to see if they generate buzz before committing. Stay flexible and pivot when necessary, using feedback from customers to make data-driven decisions using innovation accounting. Embracing failure is a natural part of the startup process.

“Reading is good, action is better”

Key Takeaways

• Innovation: scientific discovery, repurposing existing tech, new business model, new location / customer segment
• Key to success is more ability to accelerate through build, measure, learn loop than being first to market
• Smoke test: see if an idea generates buzz - demo video, preorders, email list etc.
• Build-measure-learn feedback loop: invest in the ability for rapid change
  • Validated learning: empirically demonstrating business prospects
  • MVP to get user research feedback
  • eg. WLA weekly questionnaire
• Don’t worry about scalability with MVP
  • Concierge: customer over-obsession beyond unit cost sensibility
  • Wizard of Oz: fake it til you make it
• Innovation accounting: figuring out cost benefit for a particular innovation
• Types of pivots:
  • Zoom in / out, customer segment, customer need, platform, business architecture, value capture, engine of growth (sticky, viral, paid), channel, technology

Notes

1: Wisdom

1: Start

• Applying science to startups, innovation
• Have an unchanging true north
• Customer acquisition, improving operations, steering product by pivoting
• Build measure learn feedback loop

2: Define

• Startup: Institution designed to create new products/services under conditions of extreme uncertainty
• Innovation: scientific discovery, repurposing existing tech, new business model, new location / customer segment
• Invest in the ability for rapid change

3: Learn

• Validated learning: empirically demonstrating business prospects
• Reduce friction at all costs
• MVP to get user research feedback

4: Experiment

• “If you can not fail, you can not win”
• MVP tests >> surveys, interviews
• Identity risks and assumptions before test

2: Steer

5: Leap

• Analogue, antilogue: analogous products that prove or disprove hypothesis
  • Eg. iPod and Sony Walkman
• Start learning, know your customer
• Determine what to to build / test first

6: Test

• Don’t worry about scalability with MVP
• Concierge: customer over-obsession beyond unit cost sensibility
• Wizard of Oz: fake it til you make it
• Commit to iteration: need to be ready for MVPs to fail

7: Measure

• Innovation accounting: figuring out cost benefit for a particular innovation
• Smoke test: see if an idea generates buzz - demo video, preorders, email list etc.
• Cohort analysis:
• Agile development: sprints, stories
• Have clear validation step for every feature
• Carefully consider metric choice
• Make validation reports accessible

8: Pivot or persevere

• Pivoting takes courage: pre-schedule meetings to evaluate status
• Types of pivots:
  • Zoom in / out: niching up / down
  • Customer segment:
  • Customer need: related problem
  • Platform: application ↔ platform
  • Business architecture: high margin low volume or vice versa
  • Value capture: monetization
  • Engine of growth: viral, sticky, paid
  • Channel: sales process / channels
  • Technology:
• Later stage pivots: Geoffrey Moore
  • Chasm:
  • Tornado:
  • Bowling alley:

3: Accelerate

9: Batch

• Study: Christmas cards: easier to do one at a time rather than assembly line
• Decrease iteration, feedback time
• Small batches: release routinely

10: Grow

• Customers drive sustainable growth:
  • Word of mouth: referrals, vitality
  • Side effect of usage: visibility
  • Funded advertising: LTV / CAC > 1
  • Repeated purchase / use: upsell
• Initiative objective: one of:
  • Finding new customers
  • Serving existing customers better
  • Improving overall quality
  • Driving down costs
• Engines of growth:
  • Sticky: track churn rate
  • Viral:
  • Paid:
• Focus on one engine at a time

11: Adapt

• Avoid going too quick, sacrificing quality
• Five whys: go deeper than fixing surface issue. Ask why five times and travel up chain until failure of process is found

12: Innovate

• Three components:
  • Scarce but secure resources
  • Autonomy to develop business
  • Personal stake in the outcome

13: Epilogue

• Management: human systems engineering
• “Reading is good, action is better”