Worried you’ll spend months building something nobody wants? The lean startup methodology says build a small test, measure what real customers do, then learn before you invest more. That loop is the whole idea.
This explainer is part of the shift2020 series on business strategy.
| Item | Value |
|---|---|
| Author of The Lean Startup | Eric Ries |
| Book published | 2011 |
| Core cycle | Build, Measure, Learn |
| Startup defined as | A human institution creating a new product or service under extreme uncertainty |
| Amazon’s purchase of Zappos | $1.2 billion, in 2009 |
| Innovation accounting | Three steps, starting with a baseline |
These figures come from HYPE Innovation’s overview of the method and Oberlo’s chapter summary. Figures can change or be restated, so check the source before you rely on them.
Where the lean startup methodology comes from
The name comes from lean manufacturing. According to Oberlo’s summary, the model takes its name from the Toyota manufacturing revolution led by Taiichi Ohno and Shigeo Shingo, and that summary lists the Toyota principles Ries carried over to new ventures.
- Drawing on the knowledge and creativity of individual employees
- Shrinking batch sizes
- Using just-in-time production and inventory control
- Accelerating cycle times
Eric Ries set out the method in a book published in 2011, as HYPE Innovation notes. It defines a startup broadly as any “human institution designed to create a new product or service under conditions of extreme uncertainty.” So a team inside a large company can count, not only a small firm in a garage.
Lean thinking also changes how you see effort. One summary of the book says lean thinking defines value as providing benefit to the customer, and anything else is waste.
The Build-Measure-Learn loop
The Build-Measure-Learn loop applies the scientific method to uncertainty. As HYPE Innovation describes it, you formulate a hypothesis, create a small product or feature to test it, learn from the results, and adjust.
A practical pass through the loop looks like this:
- Write down one idea you can test, phrased so it can be proved wrong.
- Choose the minimum set of metrics that will show whether it worked.
- Build the smallest thing that lets real customers react.
- Measure what they actually do.
- Decide whether to run another cycle or move to a different idea.
HYPE Innovation says it is important to define a specific idea to test and identify the minimum set of metrics to measure. The same source adds that the hypothesis must be testable, which means collecting enough data to judge the result.
The loop is not limited to new products. The same article says it can test customer service ideas, management review processes, website text and offers, or new features on existing products.
The guiding line is Ries’s own: “We must learn what customers really want, not what they say they want or what we think they should want.”
Minimum viable products and validated learning
A minimum viable product (MVP) is, in Ries’s words quoted by HYPE Innovation, “that version of the product that enables a full turn of the Build-Measure-Learn loop with a minimum amount of effort and the least amount of development time.” Oberlo’s summary adds that it should not be confused with the smallest product you could bring to market. It is simply the quickest, easiest way through the loop.
An MVP does not have to involve code. HYPE Innovation says it can be a slide deck outlining the customer journey or a series of design mockups, as long as you test with actual customers.
Two well-known examples show how small a test can be:
- Zappos: Nick Swinmurn photographed shoes in local stores, posted the pictures online, and bought a pair at full price only after a customer ordered it. Amazon bought Zappos in 2009 for $1.2 billion, according to Oberlo.
- Dropbox: the team used a simple video demonstrating the file-syncing concept as its MVP, as Faster Than Normal’s summary reports.

Validated learning means drawing conclusions from actual customer behaviour. Oberlo contrasts this with survey or interview answers about what customers might hypothetically like.
Beware vanity metrics. HYPE Innovation tells how Ries’s company IMVU showed registrations rising like a hockey stick graph, yet the chart did not show whether users valued the service and hid the marketing costs of acquiring those registrations. See the HYPE Innovation article for the full account.
Innovation accounting is the tool for tracking real progress. The first step is to establish a baseline, for example with a marketing-focused smoke test or an online sign-up form, and even unfavourable first numbers give you something to improve. The next step the source names is tuning the engine.
Pivot or persevere?
After enough cycles you face the hardest call. Faster Than Normal quotes Ries defining a pivot as “a structured course correction designed to test a new fundamental hypothesis about the product, strategy, and engine of growth.” The same summary cites Twitter, which moved from a podcast platform called Odeo to microblogging, and Instagram, which moved from a check-in app called Burbn to photo sharing.
One book summary frames the rhythm this way: products constantly change, strategy changes occasionally, and vision rarely changes. You can read that framing in ashishb.net’s summary.
If you want to see how newcomers challenge established firms, disruptive innovation theory offers a useful companion lens to this method.
Using the method inside larger organisations
The method is not only for tiny ventures. HYPE Innovation says the ideas have proved valuable particularly for large established organizations rather than just startups, and Faster Than Normal reports that large corporations use them for internal innovation projects and government agencies apply the cycles to policy development. You can find both claims in the Faster Than Normal summary and the HYPE Innovation article.
Teams that gather ideas from partners, customers and outsiders can pair experiments with open innovation, since a good test needs good hypotheses to start from.
Know the limits before you adopt it. Watch for these traps:
- Untestable hypotheses: HYPE Innovation says it is crucial that the hypothesis can be clearly tested and validated.
- Analysis paralysis: one summary warns against repeatedly talking to customers and whiteboarding instead of testing, per ashishb.net.
- One-sided reading: Faster Than Normal suggests stress-testing the thesis against cases where the advice fails and keeping a short list of when not to use this lens.
Frequently asked questions
What is the lean startup methodology in one sentence?
Show answer
It is a way of building new products by running quick Build-Measure-Learn experiments with real customers, instead of committing to a long plan. Eric Ries set it out in a book published in 2011, according to HYPE Innovation.
Is an MVP just a cheaper, smaller product?
Show answer
Not quite. Oberlo’s summary says it should not be confused with the smallest product that could be brought to market. It is the quickest and most effortless way through the Build-Measure-Learn loop.
What counts as a vanity metric?
Show answer
HYPE Innovation says likes and total accounts created on Facebook may have their place but can be considered vanity metrics. It argues the average time spent per day, per user, says more about the platform’s true value. Read more in the source article.
Does the method only work for startups?
Show answer
No. Faster Than Normal says it applies to any organization operating under conditions of extreme uncertainty, including large corporations and government agencies.



