I came up inside enterprise operations. Amazon, International Paper, Spirit Halloween, Maersk, Levi Strauss. Every one of those organizations had a formal Lean Six Sigma practice. Black Belts, Green Belts, kaizen events on the calendar quarterly, DMAIC projects in PowerPoint, control charts on the wall.
At that scale, with that volume, with dedicated continuous-improvement headcount, the methodology earns its cost. I have seen it produce real, measurable results.
I do not install LSS that way at $10M to $100M operators. Most of what makes formal LSS work at scale becomes overhead at LMM scale, and most of what makes LSS valuable can be applied without ever using the words "Lean" or "Six Sigma" in front of the team.
This post is what to keep, what to drop, and how we use the principles inside an Axis Method engagement without naming them.
What LSS actually is.
Two distinct improvement traditions, fused.
Lean comes from the Toyota Production System. Its core insight is that most of the time spent producing anything is waste, time the product spends sitting in queues, being moved, being checked, being reworked. The Toyota response was to eliminate waste and improve flow. Value-stream mapping, pull production, standard work, visual management, continuous small improvement (kaizen).
Six Sigma originated at Motorola in the 1980s and matured at GE in the 1990s. Its core insight is that variation in a process produces defects, and defects are expensive. The Six Sigma response was statistical: measure variation, identify causes, reduce them, control the process going forward. DMAIC (Define, Measure, Analyze, Improve, Control) is the project structure; control charts, capability indices, and root-cause tools are the toolkit.
Combined as Lean Six Sigma, the methodology gives you both: eliminate waste in the value stream, then reduce variation in what's left. At Motorola or GE in 2002, this produced billions in documented savings. At a $25M building-products distributor in 2026, applied as ceremony, it produces a deck.
What to drop at LMM scale.
Belt certifications.
Black Belts and Green Belts make sense when you have a continuous-improvement function with full-time headcount. At LMM scale, the cost of certification (time, money, opportunity cost of the line manager you certified) usually exceeds the value of the credential. Hire for operational instinct, not for belts.
Multi-month DMAIC projects with steering committees.
The DMAIC structure is sound as a mental model. As a project structure with a charter, a steering committee, monthly tollgate reviews, and a closeout deck, it is overhead. LMM teams move faster than the ceremony allows. Apply DMAIC privately as a check on whether you've thought through a problem; do not stage it as a deliverable.
Heavy statistical tooling.
Control charts, capability indices, hypothesis testing on shop-floor data. These work at scale because the volume produces enough samples to detect signal from noise. Most LMM processes do not have that volume. A run chart you actually look at every day beats a Statistical Process Control chart that nobody opens. Simplicity beats sophistication when adoption is the bottleneck.
Kaizen events as standalone deliverables.
The five-day cross-functional improvement event becomes a deck and a list of action items that gets filed. The principle behind kaizen (continuous small improvement, daily, by the people doing the work) is the actual value. The event format does not transfer well to LMM teams that don't have the slack for a five-day pause.
The methodology as a label.
Saying "we're going to apply Lean Six Sigma" inside a 30-person team invites every objection the methodology has accumulated over 25 years of consulting overuse. The eyes glaze. The team braces for ceremony. Apply the principles silently. Call them whatever the team already calls work.
What to keep at LMM scale.
Value-stream mapping.
This is the single highest-leverage tool from the entire LSS canon at LMM scale. Walk the actual process, end to end, on paper or whiteboard. Mark which steps add customer value and which don't. Most LMM operators have never seen their primary process drawn. Drawing it surfaces 30% of the waste before any improvement work begins. The Stabilize phase of the Axis Method is, functionally, a value-stream mapping exercise.
Standard work documentation.
The Toyota practice of writing down the current best-known way to perform a recurring task. Not aspirational SOPs. Current SOPs. Used. Updated when better ways are found. The Document phase of the Axis Method produces this for every recurring workflow in the business.
Variation reduction at the high-cost hand-offs.
Apply Six Sigma's variation discipline only to the few processes where consistency matters most: anything customer-facing, anything compliance-related, anything where defects compound (manufacturing tolerances, fulfillment accuracy). Do not try to reduce variation everywhere. The juice is in three or four places.
Visual management.
Make the operating state visible in the workspace. Daily metrics on a wall, kanban-style work-in-process tracking, andon-style alerts when something exceeds threshold. The cost is near zero; the effect on cadence and accountability is large.
Pull, not push.
Work flows when downstream is ready, not when upstream finishes. Eliminates the queue waste that compounds at hand-offs. Most LMM operators run pure push systems by default and do not realize how much waste it creates.
Continuous, daily, small improvement.
The kaizen principle, separated from the kaizen event. A daily standup that ends with "what's one thing we can improve about how we work this week" produces more compound improvement over six months than a quarterly improvement project ever does.
How OAG installs it inside the Axis Method.
I never use the phrase "Lean Six Sigma" with a client. I do install most of the principles, mapped to the five stages of the engagement.
| Axis stage | LSS principle applied | What it looks like in the engagement |
|---|---|---|
| Diagnose | Value-stream mapping | End-to-end process walk, drawn on paper, value vs. non-value steps marked |
| Stabilize | Variation reduction at hand-offs | One high-cost hand-off fixed, run chart in place to track |
| Document | Standard work | Permanent SOPs for every recurring workflow, written by the people who do the work |
| Hand-off | Visual management + pull | Daily metrics visible to the team, work-in-process tracking installed |
| Compound | Continuous improvement | Daily improvement question added to the standup; quarterly review of accumulated changes |
None of this is named. The team experiences it as "how we run operations now." The principles do their work without the methodology overhead.
The Toyota distinction that matters most.
If you read one thing about LSS at LMM scale, read about Toyota's distinction between continuous improvement and project-based improvement. Toyota's actual genius was not the toolkit. It was the cadence. Tiny improvements, every day, by the people doing the work, accumulated over decades. The toolkit existed to support the cadence.
LMM operators who try to install LSS through quarterly improvement projects are inverting Toyota's pattern. The projects deliver some improvement but the cadence never gets installed, so the moment the project ends, the team reverts. LMM operators who install the daily-improvement cadence first, even with no formal toolkit, end up further ahead than peers who ran four kaizen events and put a deck in front of the board.
If you want the principles installed inside a documented operating system, that is what the engagement is for. If you want to do it yourself, the references in the glossary entry point you at the parts that are worth reading.
Keep reading.
- Replacing the SaaS Stack with $75/mo of Cloudflare, Most small operators run on $500–$3,000/month of SaaS that compounds with every new hire. …
- What We Cut in the First 90 Days, A typical $10M–$50M operator carries 30–40% operational waste. The first 90 days of an Axi…
- How to Read an Operational Waste Audit, Three numbers come out of the Diagnose phase of every fractional COO engagement. Most oper…
- How to Actually Move a Team from Storming to Performing, Most LMM teams stop at Norming and the leadership thinks that is success. It is not. This …
Talk through it.
If any of this is applicable to where you are, book a scoping call. No pitch deck.