Skip to content
Leantensify Learn

The course, in order

30 lessons, about 6.7 hours of work, covering Foundation, White Belt, Yellow Belt, Green Belt, Black Belt and Master Black Belt. Every one ends in a tool you operate on real data. All free, and you do not need an account to read.

Start at the beginning

  1. Foundation

    Signal or noise: reading a number that movesplot a run chart from time-ordered data and describe what it shows. · distinguish common-cause from special-cause variation given a run chart. · explain why reacting to every fluctuation makes a stable process worse.14 min · run-chart, funnel-experiment
  2. The average that describes nobodycalculate a mean, a median and a range from a small set of measurements by hand. · explain why the mean alone can mislead, using a case where the median tells a different story.13 min · summary-statistics
  3. Work is a process, and somebody is waiting at the end of itdescribe a piece of work as a process with inputs, steps and outputs. · identify who the customer of a process is and state what they actually want from it.12 min · sipoc-canvas
  4. The difference between what hurts and what caused itstate the difference between a symptom and a root cause for a problem in my own work.11 min · five-whys-chain
  5. Two people, one measurement, different answersclassify a measurement as continuous, count, ordinal or categorical data. · write an operational definition precise enough that two people measure the same thing.14 min · operational-definition
  6. The best worker, and why they are notexplain why ranking people on the output of a stable process measures luck rather than performance.13 min · red-bead

  7. White Belt

    The eight wastes: what you are actually looking atname the eight wastes and give an example of each from a process I know.13 min · waste-spotter
  8. The 4% problem: how much of your process is actually worth paying forclassify a process step as value-add, business-necessary non-value-add, or waste. · calculate the value-add ratio of a process from its step times.12 min · waste-walk
  9. DMAIC: what each phase produces and why the order holdsdescribe what each phase of DMAIC produces and why the order matters.12 min · problem-statement-builder
  10. Writing a problem statement that does not contain its own answerwrite a problem statement that contains no cause, no solution and no blame. · tell a problem statement apart from a goal statement and write both for one issue.12 min · problem-statement-builder
  11. The hidden factory: why every step passes and the process failscalculate defects per unit and first time yield from inspection data.13 min · rolled-throughput-yield
  12. Mapping the gaps, not the boxesdraw a basic process map and mark where work waits.12 min · process-map
  13. Go and see: what a walk gives you that a report cannotexplain what visual management is for and give an example that would work in my area. · prepare for a Gemba walk by listing what I will observe and what I will ask.12 min · gemba-planner
  14. Scoping a project you can actually finishcomplete a SIPOC for a process I work in. · describe my own role on an improvement team and what a Green Belt will expect from me.13 min · sipoc-canvas
  15. PDCA: the Check is the part everybody skipsrun a PDCA cycle on a small change and state what I would measure at each step. · collect data against a written plan without changing how the work is done.14 min · run-chart
  16. 5S: why the tidy area keeps going backconduct a 5S audit of a workspace and score it against each of the five stages.12 min · five-s-audit
  17. Pareto: finding the few categories that carry the problembuild a Pareto chart from defect counts and identify the vital few categories.12 min · pareto-chart
  18. 5 Whys: knowing when you have reached a cause you can act onrun a 5 Whys chain and recognise when it has reached a cause I can act on.11 min · five-whys-chain
  19. Fishbone: widening the search before you narrow itpopulate a fishbone diagram with a team and group causes under the right bones.13 min · fishbone

  20. Yellow Belt

    Control limits are not specification limitschoose the right control chart for a given kind of data and say why. · read a variables control chart pair and say whether the limits can be trusted. · explain why a specification limit must never be drawn on a control chart.15 min · control-chart
  21. Capable, or just in control?calculate Cp and Cpk from a set of measurements and state the conditions that make them meaningful. · explain the difference between Cpk and Ppk and say which one answers a customer's question.15 min · process-capability
  22. Measuring the gauge before you measure the processjudge whether a measurement system is contributing more variation than the process it measures.15 min · gage-rr

  23. Green Belt

    Is this difference real? And is it big enough to care about?choose the right comparison test for how the data was collected, and say why. · state what a p-value does and does not tell me about a difference. · report an effect size and a confidence interval beside a p-value and say which decision each one supports.15 min · hypothesis-test
  24. Two things move together. Now what?fit a line to two measurements and state what its slope means in the units of the process. · explain why a fitted relationship is not evidence of cause, and name what would be.15 min · regression
  25. What the fitted line is hidingread a residual plot and say whether a straight line was the right shape for the data. · identify a point that decides the fit on its own, and say what to do about it before drawing any conclusion.15 min · regression
  26. Four suppliers, one answer, and the question it does not answersay what a significant F test has established and what it has not. · explain why running every pairwise comparison inflates the error rate, and quantify it. · read a set of corrected pairwise comparisons and state which groups a study could actually separate.15 min · anova
  27. Change everything at once, carefullyset up a two-level factorial design and say what each run is and why every combination is needed. · explain why changing one factor at a time misses interactions, using a concrete pair of settings. · read an interaction and explain why the main effects involved cannot be recommended separately. · decide which effects in an unreplicated experiment stand out from noise, and say what kind of judgement that is.15 min · doe

  28. Black Belt

    Half the runs, and knowing exactly what you gave upexplain what aliasing is and why a fractional design estimates sums of effects rather than single effects. · read a design's resolution and say which orders of effect it leaves confounded with each other. · choose a fractional design for a given factor count and run budget, and state what it will not be able to tell me. · name the follow-up experiment that would separate a confounded pair, and say why the first study cannot.15 min · fractional
  29. The gradient is zero here. That is not the same as best.use centre points to decide whether a region needs a curved model at all. · classify a stationary point as a maximum, a minimum or a saddle, and say what each means for the setting to recommend. · say why a predicted optimum outside the experimental region is a direction rather than a setting. · compute a path of steepest ascent from a first-order model and say when it is the right move.15 min · rsm

  30. Master Black Belt

    The study you can afford, and what it can actually seestate the four quantities a study design fixes — the difference worth detecting, the sample size, the significance level and the power — and solve for any one of them given the other three. · compute the sample size a comparison needs from the smallest difference worth acting on, and say why that difference is a business decision rather than a statistical one. · say what a study whose sample is already fixed by circumstance can and cannot detect, and decide on that basis whether it is worth running at all. · explain why a non-significant result from an underpowered study is uninformative rather than evidence that there is no effect. · show that power computed from the effect already observed adds nothing to the p-value, and name the question to ask instead. · distinguish the sample size needed to ESTIMATE a quantity to a stated precision from the size needed to DETECT a difference of a stated size.15 min · sample-size

Test yourself

The mock exams are assembled from a stated blueprint and report which competencies you did not demonstrate. Sitting one before you start is a reasonable way to find out which of these lessons you actually need.

Beyond this

Green Belt onwards is being written. It is listed as unwritten rather than hinted at, because a course that implies more than it contains is the fastest way to lose the people it is trying to help.