The lean startup
The Lean Startup by Eric Ries, Entrepreneur & Silicon Valley Pioneer
Core Thesis
Most startups fail — not because they build something bad, but because they build something nobody wants. Traditional business planning is fiction dressed as strategy. Eric Ries argues that startups need a scientific approach to innovation: test assumptions rapidly, learn from real customers, and adapt before running out of time and money.
A startup is not a smaller version of a big company. It is a human institution designed to create new products under conditions of extreme uncertainty.
The Three Core Engines
1. Build → Measure → Learn (The Feedback Loop)
The heartbeat of the Lean Startup. Instead of spending years perfecting a product in isolation, you cycle through this loop as fast as possible:
- Build the simplest possible version of your idea
- Measure how real customers actually respond
- Learn whether your core assumptions are true
Speed through this loop is your competitive advantage — not the size of your budget or team.
2. The MVP — Minimum Viable Product
The most misunderstood concept in startup culture. An MVP is not a cheap, half-finished product. It is the smallest experiment that tests your most critical assumption.
Examples:
- Dropbox launched a 3-minute demo video before writing a line of code — signups exploded overnight
- Zappos founder bought shoes from local stores and resold them online to test if people would buy shoes on the internet — before building any inventory system
The MVP answers: “Should we build this at all?” before you invest years finding out.
3. Validated Learning
The true measure of startup progress is not lines of code, features shipped, or money raised. It is validated learning — concrete evidence that you understand your customers better than you did yesterday.
Ries warns against vanity metrics — numbers that feel good but don’t drive decisions:
- Total page views
- Registered users
- Press mentions
Replace them with actionable metrics:
- Retention rates
- Revenue per customer
- Conversion at each funnel stage
The Pivot — One of the Book’s Most Powerful Ideas
When your experiments show your assumptions are wrong, you face a choice: persevere or pivot.
A pivot is not failure — it is a structured course correction while keeping one foot in what you’ve already learned.
Types of pivots:
| Pivot Type | What Changes |
|---|---|
| Zoom-in | A single feature becomes the whole product |
| Zoom-out | Your product becomes one feature of something bigger |
| Customer Segment | Same product, different target customer |
| Platform | From app to platform or vice versa |
| Business Architecture | High margin/low volume ↔ Low margin/high volume |
| Value Capture | How you monetise changes entirely |
The hardest part of pivoting is knowing when. Too early and you abandon something that just needed time. Too late and you’ve burned all your runway defending a dead idea.
The Five Whys — Finding Root Causes
Borrowed from Toyota’s manufacturing system. When something goes wrong, ask “Why?” five times in succession to reach the true root cause — which is almost always a human or process problem, not a technical one.
Example:
- The website crashed → Why?
- A server was overloaded → Why?
- We didn’t anticipate the traffic spike → Why?
- We had no monitoring system → Why?
- No one was assigned to infrastructure → Root cause: organisational gap
Fix the root cause, not just the symptom.
Innovation Accounting
Traditional accounting doesn’t work for startups because there’s no history to measure against. Ries introduces innovation accounting — a framework for proving genuine progress:
Three milestones:
- Establish a baseline — where are you actually starting from?
- Tune the engine — run experiments to improve each metric
- Pivot or persevere — make the call with evidence, not hope
This replaces gut-feel decisions with a disciplined, data-driven process.
The Two Types of Growth
Ries identifies three engines of growth, each with a different logic:
| Engine | How It Works | Key Metric |
|---|---|---|
| Sticky | Retain existing customers | Churn rate vs. acquisition rate |
| Viral | Customers recruit other customers | Viral coefficient (>1 = exponential growth) |
| Paid | Buy customers through advertising | Lifetime value vs. acquisition cost |
The mistake most startups make: running all three at once, measuring none properly.
Large Companies & The Lean Startup
The book’s final act is about sustaining innovation inside established organisations. Big companies face the opposite problem from startups — too much process, too little experimentation.
Ries recommends:
- Creating protected internal startup teams shielded from corporate metrics
- Allowing “sandbox” experiments with real customers but limited scope
- Treating innovation as a portfolio — not every bet needs to win
Key Takeaways
| Old Thinking | Lean Thinking |
|---|---|
| Plan everything upfront | Treat the plan as a set of assumptions to test |
| Launch when it’s perfect | Launch the smallest thing that teaches you something |
| Success = shipping features | Success = validated learning |
| Failure is shameful | Failure is data — pivot early and often |
| Scale first | Learn first, then scale |
The Central Message
Stop building things nobody asked for. The goal of a startup is not to execute a plan — it is to find a plan that works. Get out of the building, test your assumptions with real people, measure what actually matters, and be willing to change direction when the evidence demands it.
The Lean Startup isn’t just a business book — it’s a framework for making decisions under uncertainty, applicable to product teams, solo founders, corporate innovators, and anyone trying to bring a new idea into the world.