The Lean Startup is about a scientific approach to managing startups/entrepreneurs.

It is recommended that anyone working in a startup stop whatever they are doing, and spent the next 24 hours to complete the book.

I will be summarizing some great concepts that I have learnt.

Introduction: Why We Fail

In traditional management, when you fail, there are only 2 possible explanations:

  1. Fail to plan

  2. Fail to execute

But what if you fail because your hypothesis is wrong in the first place?

An execution of a perfect plan, based on a flawed idea, would only fail.

The book focus on testing hypothesis and fine-tuning it, in a scientific way.

Build-Measure-Learn Cycle

There is usually an hypothesis to an idea, a feature, a marketing plan, a business model, or anything that the customer touches.

We always assume how customer will react.

Hence the need of this cycle/iteration:

  • Build MVP (a Minimum Viable Product)
  • Measure data
  • Learn the goods and bads

![](/images/build-measure-learn-cycle.png %}

The process is actually reverse; You start with planning what you want to learn, what you need to measure, then build it small and ASAP - the eventual MVP.

Our job is to find a synthesis between our vision and what the customers would accept. Iterate and find the balance.

MVP = The Fears

We always argue about building MVP.

There is usually constant debate among team members on 2 accounts:

  1. Fear of low quality/inferior product

  2. Fear of idea being stolen

But, if we do not know who the customer is, we do not know what quality is!

If only it is so easy to get good ideas stolen! The challenge of startup is getting your product noticed by anyone, let alone a competitor.

Even if the idea gets stolen, the only way for either your competitor or you to win is - to learn faster than anyone else. For that, you have a headstart.

Measure - Innovative Accounting

Traditional numbers are “Vanity Metrics”. eg. total number of users, total active users

It’s not measuring the Growth Engine.

Startup is about Growth, and the Engine that drives it. Therefore you must and should measure that, and identify it.

Techniques: Cohorts, split test, A/B testing

Make Data part of the product.

3 A’s of the Data:

  1. Actionable Metric = clear cause and effect
  2. Accessible = Everyone can access and understand in layman terms
  3. Auditable = Talking to customer


There is no way to remove human element - vision, intuition, judgement.

Some people are better because they have talent or experience. But it is inevitable for a pivot. It is not a big deal anyway, if you understood why you need a pivot in the first place.

Have regular Pivot Or Persevere meeting. Especially if there is telltale signs.

Types of pivot:

  1. Zoom-in Pivot
  2. Zoom-out Pivot
  3. Customer Segment Pivot
  4. Customer Needs Pivot
  5. Platform Pivot
  6. Business Architecture Pivot
  7. Value Capture Pivot
  8. Engine of Growth Pivot
  9. Channel Pivot
  10. Technology Pivot

Pivot itself is a hypothesis, though a strategic one.

Pivot is a structural change designed to test a fundamental hypothesis about a product, business model, and growth engine.

Startup’s Runway = Number of Pivots it can still make

Growth Engine

Sustainable Growth = New customers come from the actions of past customers

3 types of growth engine:

  1. Sticky Engine = Customer retention

  2. Viral Engine = Spread socially

  3. Paid Engine = Self funded by advertising

There are obviously other engines.

Focus on one engine at a time (to build-measure-learn).

Five Whys

Ask “Why” five times to understand the root of the problem.

Solve a problem in stages, with incremental improvements.

But there is curse of the 5 blames.. Remember, it’s not about blaming. It’s about improving the system process.

Two simplified rules:

  1. Be tolerant on the first mistake

  2. Never allow the same mistake to be made twice

Waste not

Building products that customers doesn’t want is a big waste.

In this 21st century, we now have supreme productive capacity and technology. We should change our question from

Can it be built?


Should it be built?

We would achieve much greater heights, if our economy is not wasteful.

Other Thoughts

Traditionally, definition of productivity is by functional excellence. Example, a programmer is expected to code all day.

That’s why programmers hate meetings. But to achieve validated learning, an individual must cross-function.

Have small, cross-functional team.

Optimization Vs Learning: Some of the features are merely optimization. It improves the product, but marginally. It is demoralizing when that happens.

In traditional management, when senior management don’t see results, they assume the team did not work hard enough, or not good enough. This is when personnel changes and musical chair is being played.

Organisation has Muscle Memory. It is not easy to change the habits of a company. You are up against a system and individuals’ changes. It’s tough to implement the methods, but it’s for the good.

What should you do?

After reading the book, I ask myself what should I do?

Hope you ask too, and come to the same conclusion as me - to validate the teachings of The Lean Startup.

Think of a minimum change you can introduced, change it, then measure and learn.

PS: If you enjoy, buy the book with my bookdepository affiliate link (:




Back to Home