Lean Startup

Agile

Business Model

Strategy

Build, Measure, Learn. Eric Ries' method to validate your product hypotheses with an MVP before investing months of development.

Description

The Lean Startup is a methodology created by Eric Ries that enables product development while minimizing waste through rapid Build-Measure-Learn cycles, MVPs (Minimum Viable Products), and validated learning based on real customer feedback. Formalized in his book The Lean Startup (2011), its goal is to transform the radical uncertainty of startups into a manageable scientific process. Eric Ries forged this methodology while co-founding IMVU in 2004, after spending five months building a product nobody wanted. His mentor Steve Blank, creator of the Customer Development method, had already shown him that startups fail because they execute a plan instead of testing hypotheses. The Lean Startup fuses this idea with Toyota's Lean Manufacturing and Agile development to create a rapid learning loop. The principle works like a GPS that recalculates the route in real time: you formulate a hypothesis ("users want X"), you build the smallest product capable of testing it (the MVP), you measure real user reactions, and you learn whether to persevere or pivot. Dropbox applied this logic by creating a simple demo video before writing a single line of code, growing from 100,000 to 4 million users in 15 months. General Electric trained over 40,000 employees in the Lean Startup methodology through its FastWorks program, developing a gas turbine 40% cheaper and 2 years faster than through its traditional process. The Lean Startup method is not limited to startups: any product developed under uncertainty (new market, new segment, new technology) benefits from this approach where validated learning replaces the static business plan.

Objectives

Identify problems
Structure development
Foster innovation
Reduce product risks

Used by

-Dropbox (MVP video to validate the value hypothesis before development, growing from 100K to 4M users in 15 months)
-General Electric (FastWorks program training 40,000 employees, developing a turbine 40% cheaper and 2 years faster)
-Airbnb (successive MVP iterations, from sharing air mattresses to a global rental platform)

Advantages

Reduces development waste. You only build what user data justifies, not what your gut suggests.
Accelerates time-to-market. An MVP can be tested in days, not months. Dropbox went from 100K to 4M users in 15 months thanks to this approach.
Transforms failure into learning. A pivot is not a failure, it is data that brings you closer to Product-Market Fit.
Applicable beyond startups. General Electric trained 40,000 employees in Lean Startup (FastWorks program), proving the method works in large organizations.

Limitations

Can push toward short-termism. The obsession with MVP and rapid iteration can prevent thinking about a long-term vision or breakthrough innovations that require time.
Less suited to regulated markets. In healthcare, finance or aerospace, you cannot "test in production" with a rough MVP. Adapt the method to the regulatory context.
Assumes rapid access to users. If your sales cycle is 6 months (B2B enterprise), the Build-Measure-Learn loop turns slowly. Use proxies: interviews, prototypes, letters of intent.
The pivot culture can become an excuse. Pivoting too often without truly testing a hypothesis thoroughly is a common anti-pattern. Each pivot must be justified by data, not impatience.

How to apply Lean Startup

1
Formulate your riskiest hypotheses
Identify the two fundamental hypotheses of your project: the value hypothesis (do users find your product useful?) and the growth hypothesis (how will you acquire new users?). Formulate them in a falsifiable way: "We believe that [segment] will pay [price] for [solution] because [reason]." Always start by testing the riskiest hypothesis. Output: 2-3 critical hypotheses ranked by risk level.
2
Build a Minimum Viable Product (MVP)
The MVP is not a degraded version of your final product. It is the smallest artifact capable of testing your hypothesis. Examples: a landing page with a "buy" button (Dropbox used a video), a clickable Figma prototype, a manually operated service (Wizard of Oz). The MVP should teach you something, not impress. Rule: if you can build it in less than a week, it is a good MVP. Output: an MVP deployed and exposed to real users.
3
Measure with actionable metrics
Forget vanity metrics (visitor count, downloads). Focus on actionable metrics: activation rate, D7 retention, payment conversion, NPS. Use Eric Ries's Innovation Accounting framework: define a baseline (where you are), a goal (where you want to go), and an experimentation plan to close the gap. Output: a dashboard with 3-5 actionable metrics and their targets.
4
Learn and decide: pivot or persevere
Analyze your MVP data. If metrics are improving toward your targets, persevere and iterate. If they stagnate despite your optimizations, pivot. A pivot is not a failure; it is a strategic change based on evidence: customer segment pivot, channel pivot, revenue model pivot, technology pivot. Instagram pivoted from Burbn (a check-in app) to photo sharing after observing which feature users actually used. Output: a documented decision (pivot or persevere) with the supporting data.
5
Accelerate the Build-Measure-Learn loop
A startup's competitive advantage is not money or size; it is learning speed. Reduce the time of each Build-Measure-Learn cycle. If a cycle takes 3 months, find ways to do it in 3 weeks. Use continuous deployment, A/B tests, and early-adopter user cohorts. Each iteration must produce a measurable validated learning. Output: a shortened Build-Measure-Learn cycle with accelerated test frequency.
6
Apply the Five Whys to recurring problems
When a problem occurs (production bug, user complaint, failed experiment), ask "Why?" five times to trace back to the root cause. A technique inherited from the Toyota Production System. First why: symptom. Fifth why: systemic cause. Invest proportionally at each level: quick fix for the symptom, structural change for the root cause. Output: a causal analysis and action plan by level.
7
Define growth metrics by engine
Identify your dominant growth engine among the three defined by Eric Ries: viral engine (users invite other users), paid engine (acquisition cost is lower than customer value), retention engine (users stay for a long time). Each engine has its key metric. Focus your efforts on one engine at a time. Output: growth engine identified with pilot metric.
8
Iterate toward Product-Market Fit
The Lean Startup is a continuous process, not a one-time exercise. Each Build-Measure-Learn loop brings you closer to Product-Market Fit: the moment your product meets the market need so well that growth becomes organic. Measure your progress with the Sean Ellis Test ("How would you feel if you could no longer use this product?"). If more than 40% answer "very disappointed," you are approaching fit. Output: Product-Market Fit score measured and iteration plan.