What is lean? Lean methodology and the five lean principles

What is lean methodology?

Lean methodology is a set of principles and methods for improving work by removing waste. Waste is any activity that uses time or resources without adding value for the customer.

The Lean Enterprise Institute describes lean as creating needed value with fewer resources and less waste. It pairs that with constant experimentation toward perfect value with zero waste.

In practice, lean asks three questions of any process. What does the customer value? Which steps create that value? What gets in the way of those steps flowing smoothly?

Who invented lean?

No single person invented lean. It grew out of the Toyota Production System, which Toyota built up over several decades and describes as resting on two pillars: jidoka and just-in-time. The word lean came much later, from researchers who studied Toyota.

1. 1924Sakichi Toyoda

   His Type G automatic loom stopped by itself when a thread broke. That idea became jidoka: build quality in by stopping at the first sign of a problem. The jidoka page shows how it works today.

2. 1930sKiichiro Toyoda

   When Toyota started making cars, he set out just-in-time: make only what is needed, when it is needed, in the amount needed.

3. 1950s to 1970sTaiichi Ohno

   A Toyota production engineer, he turned these ideas into a working system on the shop floor, backed by Eiji Toyoda, who later led the company. His book Toyota Production System describes the result in his own words.

4. 1988John Krafcik

   A researcher on MIT's International Motor Vehicle Program, he used the term lean production in a Sloan Management Review article, "Triumph of the Lean Production System".

5. 1990The Machine That Changed the World

   James Womack, Daniel Jones and Daniel Roos compared lean production with mass production and brought the term to a wide audience.

6. 1996Lean Thinking

   Womack and Jones took lean beyond car making and set out the five lean principles below.

At Toyota Named and written up outside Toyota

The five lean principles

In Lean Thinking, Womack and Jones summed up lean in five principles. They are the usual starting point for any lean effort, and each one builds on the one before.

1. 1Value

   Value is defined by the end customer, not by the company. For each product or service family, it is what the customer would willingly pay for: the right thing, at the right time, at a price that makes sense to them. Womack and Jones start here because every later step depends on it. A team that misjudges value can build a very efficient process for making the wrong thing.

   In practice, this means talking to customers and watching how they use what you deliver. A clinic might learn that patients care most about how long they wait for test results. A software team might find that a report is always exported to a spreadsheet and never read on screen. A gemba walk takes managers to the place where the work happens, so they see this for themselves.

   Once value is clear, anything the customer would not pay for becomes a candidate for removal. The eight wastes give names to the usual suspects, such as waiting, rework and overproduction.

   One way to make value concrete is to write it from the customer's side. A bookkeeping firm might describe itself as doing monthly bookkeeping. From the client's side, the same service reads: I get accurate numbers by the 5th of each month, so I can decide what to spend. The second version says what to measure: on time, accurate and useful for a decision.

   Ask: What is the customer actually paying for?

2. 2Value stream

   The value stream is every action needed to bring one product or service from order, or raw material, to the customer. Most teams only see their own part of it. Laying out the whole stream shows the handoffs, queues and approvals between departments, where no single team owns the delay.

   Womack and Jones sort the steps into three groups. Some steps create value. Some create no value but are needed for now, because of current equipment, rules or technology. The rest create no value and can be dropped straight away. The middle group is the target of later improvement work.

   A value stream map is the standard tool for this. It records each step with its process time and the waiting time before it, so the gap between work and waiting becomes visible. The map often shows that an item spends far more time waiting than being worked on. Value stream mapping software helps keep the map up to date.

   Ask: Which steps would the customer notice if they disappeared?

3. 3Flow

   With waste taken out of the value stream, the next job is to make the remaining steps run without stopping. Traditional operations group work by department and by batch. Parts wait until a full batch is ready before they move on, and documents sit in an inbox until someone gets to them. Flow means a unit of work moves to the next step as soon as it is ready.

   Batches feel efficient for each step but slow down the whole. Every item in a batch waits for the others, and a defect found at the end may affect the entire batch. Smaller batches, and ideally one-piece flow, shorten lead time and bring problems to light sooner.

   Several methods support flow. Takt time sets the pace to match customer demand. SMED cuts changeover time so that small batches pay off. Limiting work in progress stops queues from growing, as the guide on why limiting WIP makes teams faster explains, and the Little's Law calculator shows how WIP, throughput and lead time relate.

   Ask: Where does work wait, and why?

4. 4Pull

   Pull means nothing is made upstream until someone downstream needs it. The opposite, push, builds to a forecast or a schedule and hopes demand matches. When it doesn't, stock piles up in one place while shortages appear in another.

   In a pull system, the downstream step sends a signal when it uses something, and the upstream step replaces only that. In a supermarket, the signal is the gap on the shelf. In a factory, it is often a kanban card that travels back to the supplying process. In office and software work, a team pulls the next task only when it has room for it, which is how a kanban board with WIP limits works.

   Pull depends on flow. If lead times are long, the signal takes too long to answer and the system falls back on stock, which is why Womack and Jones place pull after flow. The pull system page covers the main designs, and the kanban calculator works out how many cards a loop needs.

   Ask: Is anything made before someone asks for it?

5. 5Perfection

   The last principle closes the loop. Once value is defined, the stream mapped, flow created and pull in place, the team starts again. Each round removes waste, and with it gone, smaller problems that were hidden before come into view. Womack and Jones describe perfection as a direction to keep moving in, not a state anyone reaches.

   This is the role of kaizen, or continuous improvement. Improvement becomes part of normal work, done in small steps by the people closest to the process. PDCA (plan, do, check, act) gives each change a structure: try it on a small scale, check the result, then adopt it or try something else.

   Adopted changes go into standard work, the written current method. It is the baseline the next round is measured against, so a team can tell whether a change made things better. Kaizen software can track each idea from suggestion to result.

   Ask: What problem did the last improvement expose?

What counts as waste?

Taiichi Ohno grouped the waste he found in mass production into seven types: overproduction, waiting, conveyance, processing, inventory, motion and correction.

Many lean practitioners now add an eighth: talent that the organization does not use. The eight wastes page gives an example of each, from offices as well as factories.

Outside the factory the same wastes take other forms. Some illustrative examples from offices, services and startups:

Core lean concepts

A few ideas come up again and again in lean. Knowing them makes the methods easier to follow. The glossary defines each term in more detail.

Value-added and non-value-added work

A step adds value only if it changes the product or service in a way the customer cares about. Lean aims to cut the rest.

Muda, mura and muri

Muda is waste, mura is unevenness and muri is overburden. Uneven work and overloaded people tend to create waste, so lean tackles all three.

Just-in-time and jidoka

The two pillars of the Toyota Production System: make only what is needed, and stop to fix problems at the source.

Gemba

The gemba is the place where the work happens. Lean leaders go there to understand a problem, rather than relying on reports.

Standard work and kaizen

A standard records the current agreed way to do a job. Kaizen improves it, and the improved method becomes the new standard.

Respect for people

Lean depends on the people doing the work to find and solve problems. Toyota names respect for people as one of the two pillars of the Toyota Way, next to continuous improvement.

The core lean methods

Lean has a large toolkit, but a handful of methods do most of the work. Each one links to a page with steps, a worked example and common mistakes.

The full library is on the lean tools page.

Types of lean

The same principles apply wherever work flows through steps. Each field has adapted lean to its own kind of work, and the names below are the ones you will meet most often.

Lean, Six Sigma and agile

The three are often mixed up. Lean works on flow and waste, Six Sigma works on variation and defects with statistics, and agile works on short feedback loops in software and product work. Many organizations combine lean and Six Sigma as Lean Six Sigma.

Agile shares lean's focus on small batches, and kanban came to software straight from lean manufacturing.

The guide to lean and Six Sigma explains the differences in more detail.

How to tell if lean is working

Measure the flow of work, not how busy people are. A handful of measures covers most of it, and a team board needs only three or four.

Flow efficiency compares the time spent working on an item with its total lead time, and the flow efficiency calculator works it out. Choose measures the team can see every day, and drop any measure that rewards hiding problems or rushing work.

Why lean efforts stall

The principles are simple to state, and the methods are well documented. Lean efforts still stall, and the reasons tend to repeat. Knowing them in advance is cheaper than finding them out.

1. 1The tools arrive without the problem

   A plant runs 5S on every area or a team installs a kanban board because lean says so, not because a customer problem calls for it. Once the first round is done, nobody can say what the effort was for, and it fades. Start from a problem the customer feels, and pick the tool that fits it.

2. 2Improvement sits outside normal work

   A kaizen event or a training wave produces a burst of change, then everyone goes back to the day job. Without a daily routine that surfaces problems and assigns owners, the gains slip back. This is why daily management and standard work matter as much as the event itself.

3. 3Leaders delegate it

   Lean asks managers to go to the gemba, ask questions and coach, which takes their time every week. When it is handed to a project office or a consultant, people read it as a program that will end. The lean management guide covers what leaders do differently in a lean system.

4. 4Efficiency is measured in the wrong place

   Local measures, such as machine use or tickets closed per person, reward big batches and busy people. Lean improves the flow of the whole value stream, so a step may look less efficient while lead time for the customer falls. If the measures do not change, the old behavior comes back.

5. 5Cutting waste turns into cutting people

   If the first visible result of lean is job losses, the people closest to the work stop pointing out waste. A common safeguard is an early, public commitment that improvement will not cost anyone their job, with freed-up people moved to new work instead.

How to start with lean

Start with one process that matters to a customer and that the team can change. A month is enough for a first full cycle. Here is one way to plan it.

1. Days 1 to 5

   Pick and define

   Choose the process, and write down where it starts and ends, who works in it and what the customer gets from it. Keep the scope small enough to walk through in an hour.

2. Days 6 to 10

   See the work

   Follow real requests through the process and draw it on paper first, as the guide on value stream mapping on paper suggests. Mark waiting, rework and handoffs, and ask the people doing the work what slows them down.

3. Days 11 to 15

   Choose one target

   Pick the biggest waste you can see, and say what better looks like in numbers, such as a shorter lead time or fewer returns.

4. Days 16 to 23

   Test one change

   Run a small PDCA experiment, such as a checklist, a WIP limit or one approval fewer. Measure before and after.

5. Days 24 to 30

   Hold the gain and pick the next target

   If the change worked, write it into standard work. If it didn't, record what the team learned and adjust the test.

After the first month, make problems visible every day. A short daily huddle at a board, with clear owners for each issue, keeps improvement going after the first project.

Add software only when paper limits you. The lean manufacturing and lean management guides cover what to use, and when.

Frequently asked questions