A lean maturity assessment scores where a site sits on a lean maturity model, usually across culture, flow, and management system, so you can measure the gap between today's practice and an ideal state and build a prioritized roadmap. It grades habits and behavior, not how many tools you own.
The rest of this guide explains why maturity is scored by behavior rather than tool count, what the dimensions and levels look like, how to run an assessment without it turning into a self-congratulation exercise, and how to turn the score into an improvement plan. If the underlying philosophy is new, start with lean manufacturing; this is how you measure your progress through it.
What is a lean maturity assessment?
A lean maturity assessment is a structured evaluation that rates how deeply lean thinking is embedded in a site, on a scale from reactive firefighting to a self-sustaining improvement culture. It is a mirror, not an audit. An audit checks whether you follow a standard; a maturity assessment asks a harder question, has lean become how people actually work here, or is it a set of activities that happen when someone from corporate visits?
The distinction matters because tool counts lie. A plant can have 5S boards, kanban cards, and a value stream map on every wall and still be immature, the boards go stale, the cards get ignored, the map was drawn once and never updated. Maturity measures whether the behavior behind each tool is real and sustained. That is why the Shingo Model is built on principles and behaviors rather than tools: results come from behaviors, behaviors come from systems, and systems come from the principles leaders actually live.
Why score maturity instead of counting tools?
Because tools are easy to install and easy to fake, while behavior is hard to fake and is what actually delivers results. A team can be trained on every lean tool in a quarter and change nothing about how it runs the line the following Monday. Scoring maturity forces you to look past the artifacts to the habits: does the morning huddle actually surface problems, or read yesterday's numbers aloud? Does an andon signal get an immediate response, or a shrug?
The Shingo Institute captures this in three insights: ideal results require ideal behaviors; purpose and systems drive behavior; and principles inform and govern behavior. Read backward, that is the whole case for maturity scoring, if you want durable results, you have to grow the behaviors, and you cannot grow what you do not measure. A maturity model gives behavior a scale.
What are the dimensions of lean maturity?
Most maturity models score three or four dimensions so a site does not get a single blurry grade. A common split is culture, flow, and management system, sometimes with results as a fourth. Scoring them separately keeps a plant honest, plenty of sites have strong tool discipline on the floor and no leadership behavior above it, or the reverse.
| Dimension | What it asks | Signs of low maturity |
|---|---|---|
| Culture & people | Do operators find and solve problems, and are their ideas used? | Improvement comes only from management; non-utilized talent |
| Flow & process | Does value flow, or sit in queues and batches? | Big WIP, long lead times, firefighting expedites |
| Management system | Do leaders coach at the gemba and follow standard work? | Leaders manage by report; standards drift; no daily cadence |
| Results & sustainment | Do gains hold, or evaporate after each event? | Metrics spike then slide back; no standard work updates |
What do the maturity levels look like?
Most models use a five-level scale from reactive to self-sustaining. The exact labels vary, but the arc is consistent: you move from fighting fires, to installing tools, to standardizing them, to leaders coaching the system daily, to a culture that improves itself without being pushed. Here is the arc in plain terms.
| Foundation | Detail | Source |
|---|---|---|
| Behavior over tools | Ideal results require ideal behaviors; principles and systems drive behavior | Shingo Institute, Shingo Model |
| The improvement aim | Perfection is the fifth lean principle: repeat the cycle without end | Lean Enterprise Institute |
| Direction-setting | Maturity gaps feed the annual improvement plan through policy deployment | Hoshin kanri |
How do you run a lean maturity assessment?
Run it as a structured walk and interview against a written rubric, scored by evidence you can see and touch, not by opinion. The output is a score per dimension plus a short list of the highest-value gaps. Here is the sequence that keeps it honest.
- Pick the model and rubric first. Choose or adapt a maturity model with defined dimensions and level descriptions, and write the rubric before you assess anything. Everyone should know what a "level 3 flow" looks like in advance, so the score is repeatable across assessors.
- Assess by evidence, not opinion. Walk the floor. For each dimension, ask for proof: show me the standard work, show me the last five problems this board caught, show me a gain from three months ago that held. If the evidence is not visible, the maturity is not there, whatever the binder says.
- Interview across levels. Talk to operators, supervisors, and site leaders separately. Maturity gaps show up as gaps between what leaders believe is happening and what the floor experiences. When those two stories disagree, the truth is usually on the floor.
- Score each dimension separately. Resist a single overall grade. A site at level 4 on flow and level 2 on management system needs a very different plan than the reverse, and one blended number hides that.
- Find the constraint, not every gap. A full assessment surfaces dozens of gaps. Rank them, and identify the one or two whose improvement would unlock the most, usually a leadership or measurement gap that is holding back everything below it.
- Turn the top gaps into a roadmap. Convert the priority gaps into specific, owned, time-bound improvement targets, and fold them into the site's annual plan through hoshin kanri so they get resourced instead of forgotten.
- Re-assess on a cadence. Maturity is a trend, not a snapshot. Re-score every six to twelve months with the same rubric so you can see whether the behaviors actually moved, which is the only proof the roadmap worked.
How do you turn the score into a roadmap?
Convert the two or three lowest-scoring, highest-leverage gaps into concrete behaviors with owners and dates, then sequence them so the foundational ones come first. Do not try to raise every dimension at once. If the management system scores lowest, fixing leader standard work and the daily coaching cadence usually lifts culture and sustainment along with it, because those behaviors were the constraint holding the others down.
Sequence matters. Installing more tools on a site that scores low on management system just adds more artifacts to go stale, that is how plants get stuck at level 2 for years. The productive order is almost always: stabilize the management system and measurement first, then flow (mapped with a value stream map), then let culture deepen as people see that surfacing problems actually leads to fixes. Tie the roadmap to lean metrics that move so you can prove progress with numbers, not just a higher self-assessed score next quarter.
What do maturity assessments get wrong?
The common failure is grading yourself generously and calling it progress. A maturity assessment is only useful if it can return bad news, and the pressure to score well is enormous, nobody enjoys writing "level 2" next to their own plant. Three habits keep it honest. Use outside eyes or at least a cross-site assessor, because a team cannot see its own blind spots. Demand visible evidence for every score, so the grade rests on what exists, not what is intended. And separate the assessment from anyone's performance review, because the moment a low score threatens a bonus, every score mysteriously improves and the tool stops measuring anything.
The other failure is treating the number as the goal. The score is a diagnostic, not a trophy. A site that honestly scores itself a 2 and climbs to a real 3 has done more than one that declares itself a 4 and changes nothing. This is the same trap that catches tool-first programs and belt-collecting Lean Six Sigma efforts: the credential becomes the objective and the behavior stays put.
What does an honest maturity assessment need?
It needs evidence you can trust, and that is where the data plumbing matters. Half the questions in a maturity walk are really data questions: did that gain hold, how fast do problems get caught, does the daily huddle work from real numbers or yesterday's memory? On paper-based sites, those answers are guesses, and a maturity score built on guesses drifts toward flattery.
This is the layer Harmony provides: digitize the capture operators already do, connect the machines and systems you already run, and make the evidence a maturity assessment needs visible in real time, with no rip-and-replace. When CLS moved production logging off paper, supervisors could see whether a problem recurred within the shift instead of guessing at month-end, which is exactly the evidence a maturity assessor asks for. See how the data layer supports the daily cadence a higher maturity level depends on, alongside the plant KPIs you already track.
Score honestly, by evidence, one dimension at a time. Fix the constraint, not every gap. Re-assess on a cadence. That, not a flattering number, is how a maturity assessment moves a plant up the ladder.