Harry Kenworthy opened this session by telling the audience what it would not be: a session about tools and techniques. That framing is the point. Most Lean efforts fail, in his experience, not because the tools don't work but because organizations treat Lean as a set of tools to deploy rather than as an organizational capability to build. The tools are the easy, visible, common-sense part. They are also, as Harry put it, the minor part.
The session walked through four things: why Lean fails too often, why organizational knowledge undermines Lean, when and how leadership should become engaged, and the proven path that the Des Moines, Iowa public school system followed. The school system is the case study, but Harry was clear that the school context is incidental. The conditions a large public school system faces — high accountability, limited budgets, competing priorities, complex human systems, the constant turnover risk of elected leadership — expose weaknesses that exist in plenty of private-sector transformations too. Lean leadership is Lean leadership whether the setting is manufacturing, healthcare, government, or a school district.
Mark Graban, hosting, made the same point from his own experience. He had spoken with school superintendents, including one from the district he grew up in, and found the conversations familiar. The adoption problem doesn't change much from industry to industry. What changes is whether the organization recognizes it as an adoption problem in the first place.
Harry Kenworthy is Principal and Manager of the Quality and Productivity Improvement Center (QPIC, LLC), a consulting organization he founded in 1984 and has run full time since 2004. He worked with Dr. W. Edwards Deming in 1983-85 on a series of two-day seminars across the United States, sponsored by MIT. He has spoken at more than 90 conferences on quality, productivity, Lean, and Six Sigma, and has been published in Quality Progress, Purchasing, and Government Finance Review.
Harry was one of the first practitioners to apply Lean in the government sector, beginning in the mid-1990s. He was a founder of the Connecticut Quality Council and chaired the Manufacturing Council of the Connecticut Business and Industry Association (CBIA). He is a Six Sigma Master Black Belt and the author of "Lean Government NOW!" — a book he describes not as a collection of stories but as a step-by-step guideline, the training manual QPIC uses with clients. This was a returning-presenter session for KaiNexus; Harry had previously presented "Ten Commandments for Lean in Government (and Beyond)." More on his work is at leangovcenter.com.
Harry started with the misread that sets up most failures. When an organization first encounters Lean, the reaction is reassuring. 5S looks straightforward. Single-piece flow makes sense. A lot of it reads as common sense. The conclusion people draw is: we've got this, this will be easy, no need to worry about it.
That confidence is where the trouble starts. In the government sector, Harry sees it concretely in the RFPs and RFQs that organizations issue. Someone inside the organization has seen what other organizations were doing, read a little, attended a couple of sessions, and on that limited knowledge writes a request for proposal. The RFP almost always focuses on tools — so many Kaizen events, so many people trained, so many green belts, so many black belts, a one-hour overview for executives. Harry's reaction when he reads one: all they're focused on is tools. The organization will get some gains. But the tools are the minor part, and the RFP has already designed the effort to under-deliver.
The single biggest reason Lean fails, in Harry's experience, is the absence of management support, leadership, and alignment. Top people delegate Lean. They don't lead it. They avoid developing their own Lean expertise, and they avoid building any organizational discipline around it. The consequence is straightforward: they don't learn, because they're not doing. Harry's recent example — a green belt at a 1,400-person organization in Canada had contacted him, one of the few practitioners in her organization, asking how to get traction with upper management so they would get on board and understand the roles they needed to play. That question, asked from the bottom of an organization, is itself the symptom.
The backbone of Harry's failure analysis is John Kotter's work on change management at Harvard. Lean adoption is a major organizational change, so Kotter's eight elements for sustaining change apply directly. QPIC adapted them for Lean and added detail underneath each one. Run as a checklist of what goes wrong, the eight are:
Is there a sense of urgency — a real reason the organization should be doing this at all?
Is senior management involved? In more than half the implementations QPIC sees, senior leadership doesn't even attend training. They've delegated it.
Is there a guiding coalition — a steering committee, champions in place, an upper-management team that owns this beyond their individual silos, accountability for results?
Is there a vision and strategy? People can articulate where the organization is going, what it will look like, and how each person is involved.
Has the change been communicated? Harry is a strong advocate for elevator speeches, intranet sites, postings, and communication through huddles — with the introductory communication delivered by the top leadership team, not the Lean office. In a large agency rollout, the top leader might record a short introductory video, but each department head then presents the actual material to their own people. Organizations almost always believe they communicate enough. They almost always don't.
Is there broad-based action — and are projects being selected well? "Go forth and see what you can find" does not work. QPIC wants the top management team to bring projects themselves, develop charters, and have the steering committee rank them on a project selection grid so the highest-leverage work goes first.
Are there short-term wins? Nothing sells success like visible wins, with financial impact attached and recognition ceremonies to consolidate them.
Are gains consolidated to produce more change — and are there internal experts, rather than total dependence on outside consultants?
Harry's framing of this list is almost cheerful: the good news is that all you have to do is not do these eight things. Do the opposite, and you're in far better shape.
Harry put up a true north model — tools and techniques on one side (supporting a challenge, Kaizen, learning to see, continuous improvement), respect for people and culture on the other. He credited the framing to Jeff Liker's view of true north, drawn from the Toyota Production System, and noted that QPIC has reframed it as the "thinking people system" because that name captures what it is actually about. The respect-for-people and culture side is, by a wide margin, the most important.
The culture point came with a hard caveat about sustainability. A state client had recently told QPIC they wanted training built into the organization so that no matter what happened — even if the governor wasn't re-elected — the work would survive, embedded in the culture. Harry's response was honest. That's the hope. But never underestimate a new manager's ability to come in and set a completely different set of tunes to march to. In political arenas especially, a change of party can mean the new leadership simply declares the prior work no good and moves on. Maintaining and sustaining the culture is one of the hardest challenges any organization faces, and no amount of embedding makes it immune to a determined leader who wants to undo it.
QPIC came to the Des Moines Public Schools through the Government Finance Officers Association — Thomas Harper, the district's chief financial officer, was active in GFOA, and QPIC was recommended. The district is sizable: roughly 33,000 students, about 5,200 staff, of whom around 2,200 are in classrooms and about 3,000 are in support functions — finance, HR, buildings and grounds, cafeterias, busing. A district that size has no shortage of improvement opportunities. The journey was covered in Governing magazine, in a piece by Harvard's Steve Goldsmith on bringing Lean to education. QPIC worked with Des Moines from April 2015 until the district reached a self-sustaining mode in mid-2017.
Harry described the district's two key leaders in terms that map onto the failure analysis. The superintendent, Thomas Ahart, recorded a video message about Lean for the whole district — visible leadership of the kind QPIC wants to see. The internal executive champion was Thomas Harper, the CFO. He attended the training, he chairs the steering committee, and he has been the guiding, inspirational presence. One full-time Lean coordinator reports to Harper. Everything else is carried by people the district calls stewards.
The path itself moves through phases.
The demonstration phase (April to October 2015) opened with a two-day boot camp. Top management was required to attend — not optional, not a cameo. A separate boot camp ran for the internal candidates who became the lean stewards; in that one, each person had to arrive with a real project, and over the two days developed a project charter and identified which of the eight wastes were present in it. Those charters became the basis for the steering committee to quantify and decide which projects went first. Boot camps came with 30-day action items — an A3, some 5S activity, setting up daily dynamic data collection — so people were doing things, not just sitting in training. The phase included an in-depth simulation moving from batch processing to single-piece flow, and produced four selected Kaizen events. The whole point of the demonstration phase was for the district to see for itself whether this was worth continuing.
The tools phase came once the district decided to move forward. Harry called the name a bit of a misnomer — it's a deeper dive into tools, but the more important work is structural. About 20 people, fairly senior managers and supervisors, were selected and trained as stewards through a more intensive boot camp. The steering committee was formally established and trained on its own standard work, its charter, and a naming convention for the journey. Des Moines chose to call its effort "continuous improvement" rather than "Lean." Harry advocates avoiding the word "Lean" where possible — it carries the connotation of cutting to the bone, trimming fat, eliminating jobs, which is exactly the wrong message.
The stewards' role is the structural heart of the model, and Harry was emphatic about one thing: the stewards do not own the Lean process. They support it. They meet with department heads roughly monthly, understand what projects are underway — Kaizen events, business process improvements, "just do its," benchmarking — and help spread the work. Ownership sits with the department heads and managers. In Des Moines that means five "chiefs": curriculum, schools, finance, HR, and facilities/operations. Harry named the most common adoption mistake plainly. Organizations train the stewards and coaches but not the senior managers, then expect those trained people to go out and run projects. That is a recipe for thin results and badly frustrated coaches.
Harry laid out QPIC's preferred organizational model for Lean in government, which he called hub-and-spoke.
At the top sits the chief executive, who needs a genuine working understanding of Lean — someone who has sat through a boot camp and done hands-on work, not made a one-hour cameo appearance. When the chief executive has actually done the work, the lights go on; they grasp what is possible. When they haven't, the opportunity is significantly diminished. Below the chief executive is the leadership core team, including the executive champion — in Des Moines, CFO Thomas Harper — who leads the Lean steering committee, which meets monthly. Below that is the steering committee itself, and a Lean coordinator.
The coordinator detail is striking. One full-time person supports the Lean effort across a 5,200-employee organization. Her office is near the superintendent's, giving clear access to top management. She is the hub. The spokes are the stewards — trained internally to do the coaching across departments. The model works only because ownership sits with the leadership and management layer. They are the ones driving it, and the ones the executive champion and superintendent are constantly asking: what are you doing, what are your plans, what's on your project grid, what about training? Those questions are what drive the behavior. Out at the ends of the spokes are team leaders and improvement teams. Over time, Kaizen events start as the dominant activity and steadily become a smaller share of the work.
The Des Moines coordinator, Emma Knapp, earned her Six Sigma Black Belt. But Harry was specific that her main job is not training delivery — it is supporting the stewards, giving them more tools, more training, more competency, so that they can support the management layer. When an attendee asked whether one person in the improvement department is really enough, Harry's answer was that one person is unusual, and it works only because the model multiplies capability rather than concentrating it. He compared it, half-apologetically, to a Ponzi scheme — capability spreading outward through the stewards rather than depending on the central office to do everything. The purpose of a Lean office is to hold deep expertise and develop it in others so it multiplies, not to be the place that owns and drives the effort. If the Lean office is the only thing driving the work, the work stops the moment that office stops pushing.
A small but firm piece of advice ran through the session: never call this a program. To most people, a program is something that comes and goes — it has a short, defined duration. The word itself signals impermanence. Organizations should give the effort an initiative name and build an elevator speech around it. Harry's examples: the City of Phoenix called its effort "Advance Phoenix," with "advance team members" as internal coaches; Denton, Texas called its work "Denton's Lean Journey"; a motor vehicle department called its effort "Evolution"; Denver had a "Peak Performance Academy."
The Des Moines elevator speech answered four questions, deliberately. What are we doing — establishing an approach to evaluate district processes, identify improvement opportunities, and drive efficiency, effectiveness, and better student outcomes. Why are we doing it — the district is continually asked to do more with less, and needs to ensure the sustainability of its priorities. What do you expect of me — participation, collaboration, commitment, sharing ideas. What's in it for me — empowerment to improve your own processes and a real ability to make work better, so employees can maximize their contribution to student achievement. That speech, a short video, and a 45-minute training rollup form the opening communication. The steering committee charter and the project rating grid follow.
When Harry turned to culture, his summary was simple: teach everybody how to solve problems. The mechanics are familiar — the Plan-Do-Check-Act cycle, A3 thinking, the 5 Whys to peel back the onion to root causes. But Harry stressed two things that he believes get shortchanged.
The first is the planning phase. People rush past it, grabbing for solutions before they have clarified what the problem actually is. He reached for the Einstein line — given an hour to work on a problem, spend 55 minutes defining it and five minutes solving it. The same discipline applies to every PDCA cycle and every A3.
The second is something Harry said he rarely sees on PDCA or A3 templates at all: a small red box at the bottom that says ensure there is a good measurement system. You may be measuring your process with a measurement system that itself is broken. In manufacturing, Harry estimated, the process measurement system needs work about 85% of the time. Without operational definitions — shared understanding of what is being measured, the same baseline for everyone — a team can run improvement cycle after improvement cycle while the real problem is the measurement system. When an attendee asked whether ensuring a good measurement system is a Six Sigma topic, Harry confirmed it sits in the Six Sigma realm, and gave a government example. A social worker assessing whether someone in their home can feed themselves — an adult daily living skill — is making a judgmental measurement. Ask several social workers to rate the same person on a one-to-six scale and the ratings scatter all over the place. Closing that variation takes real training toward consistent measurement.
The picture of how it all fits: you make an improvement through PDCA, an A3, a "just do it," a BPI, or benchmarking; that raises the standard of work; and then you have to lock it in and hold the gains. Holding gains requires full employee engagement. You cannot simply announce the new standard and hand people new standard work — the people doing the work have to understand it, be engaged in it, and own it themselves, or the process slides backward. Harry named establishing and maintaining rigorous standard work as the single biggest challenge QPIC sees in organizations.
He closed the culture section with the eight wastes — DOWNTIME, with the N for non-utilized people and extra processing as the eighth-waste pairing — and a benchmark: QPIC's goal with a client is always at least a 50% improvement, and they rarely fall short of it, sometimes reaching 90 to 95% because there is that much waste in the process. He also defended benchmarking, which some in the Lean community dismiss. In government and schools, benchmarking against similar districts and entities is a genuine opportunity to learn. It doesn't mean the work is done — it establishes the baseline you improve from, a higher plateau to start at.
QPIC's core leadership development process is built on the conviction that you cannot train your way to better leadership. It starts with an anonymous 360-degree survey, compiled into graphs and data and shared with the leader. The leader drafts an action plan for what they think they should work on; a QPIC consultant adds what the coach observed; the plan is finalized around the areas that will shift the person toward Lean leadership thinking. Then come six monthly coaching sessions — roughly 90 minutes each, face to face, observing the actual workplace and discussing progress. A second 360-degree survey follows six months later. QPIC has seen employee engagement factors improve by 50% or more through this process.
Harry tied it to Kata — practicing a routine, the discipline Mike Rother is most associated with — and to a line he said he likes a lot: the ability to take a new behavior and make it an ingrained habit can only come through deliberate practice and coaching. Nobody learns golf by watching it on television or reading books. You cannot sit someone in a classroom, teach them to be a great leader, and send them back to their normal job. The new habit sticks only when knowledge is followed by practice and a coach who helps the person improve.
The most candid exchange came at the end, when an attendee asked how QPIC handles a new client who wants to dictate what to teach them. Harry called it a real challenge — and Mark Graban said he could identify with it too.
It traces straight back to Harry's opening slide: "don't worry, we've got this, here's what you need to teach us." If the people at the leadership end are genuinely trained, practicing Lean sensei, fine — defer to them. But if the person dictating the curriculum has seen a few videos, skimmed some training material, attended a couple of Kaizen events, it's a hard road. Harry named the underlying condition "bright person's disease" — the leader so confident in their own intelligence and credentials that they won't really listen to an outside helper. You can demonstrate some successes and hope they come around. Sometimes it's simply impossible. Many leaders got where they are by knowing the answers and directing people, and that same disposition is what blocks the humility the work requires.
Mark connected it to a book he has recommended on past webinars — "The Toyota Way to Lean Leadership" — whose first chapter is built on leading with humility. His dry observation: he has never seen an RFP ask for humility training. The RFQ comes in counting Kaizen events, green belts, black belts, and a one-hour executive overview, and then the consultant hopes that once inside, in the final stages of setting things up, they can work some magic, bring in references, and shift the leader's view. When a leader is adamant about running things their own way, that is a very difficult challenge — and one of the most honest notes the session ended on.
Harry's session is a strategy talk. The proven path, the hub-and-spoke model, the failure modes — none of it is software, and Harry was not selling any. But the model he described has a structural dependency that infrastructure speaks to directly.
The hub-and-spoke model concentrates very little staff at the center. One full-time coordinator supports 5,200 employees. A set of stewards, each carrying Lean work alongside a full-time job, coaches department heads roughly monthly. The whole thing works only if the small center can see what is happening at the ends of every spoke without being in the room. That is a visibility problem before it is anything else. When the executive champion asks a chief "what's on your project grid, what are you working on, what about training," that question needs a real answer that the chief, the steward, and the coordinator can all see the same way. Des Moines tracked roughly 163 projects across finance, operations, the office of schools, and the office of academics, broken down by department and project type — Kaizen events, rapid action projects, business process improvements, "just do its," benchmarking. Holding that portfolio in a form everyone can see, with status visible at a glance, is exactly the work a purpose-built improvement platform does. A spreadsheet held by one coordinator does not scale to a hub-and-spoke model; the model assumes the work is visible to the spokes and the center at once.
Harry's emphasis on holding the gains points the same direction. An improvement raises the standard of work, and then the gain has to be locked in — which requires knowing, over time, whether the improved process is holding or sliding back. That is a tracking and monitoring function. Harry named rigorous standard work as the biggest challenge organizations face; sustaining it means the improvement and its standard don't disappear once the Kaizen event ends and attention moves on. Infrastructure that keeps the improvement, its new standard, and its status in view is part of what makes "hold the gains" operationally real rather than aspirational.
His point about short-term wins and consolidating gains connects too. Nothing sells success like visible wins with financial impact attached, and Harry wants those wins recognized and consolidated to produce more change. Wins that are visible across the organization — searchable, with their impact recorded — do more selling than wins buried in one department's files. The Des Moines results page, public on the district's website, is a version of exactly that: the work made visible enough that other departments, and other districts, can see it and learn from it.
None of this changes the strategy Harry laid out. Leadership has to lead. Ownership has to sit with operational managers, not the Lean office. People have to be taught to think, not just to do. What infrastructure does is hold the structure together at scale — give the hub-and-spoke model a shared surface, keep the project portfolio visible, make the gains trackable, and put the wins where the rest of the organization can see them. The model Harry described is a visibility-and-ownership model, and visibility at scale is a tooling problem as much as a leadership one.
Why does Lean fail most often? In Harry Kenworthy's experience, the leading cause is the absence of management support, leadership, and alignment — not a problem with the tools. Top leaders delegate Lean instead of leading it, avoid developing their own Lean expertise, and avoid building organizational discipline around it. Because they aren't doing the work, they don't learn it. The tools (5S, single-piece flow, value stream mapping) are the easy, visible, minor part. Adoption fails on leadership, culture, and intentional design — the parts organizations consistently underestimate.
What are the eight failure modes of Lean deployment? QPIC adapted them from John Kotter's change management work at Harvard: (1) no sense of urgency; (2) senior management not genuinely involved; (3) no guiding coalition — steering committee, champions, accountability; (4) no clear vision and strategy; (5) the change not communicated well, ideally by leadership rather than the Lean office; (6) no broad-based action and weak project selection; (7) no short-term wins; (8) gains not consolidated and no internal experts. Harry's framing: to improve your odds, simply do the opposite of these eight.
What is the "proven path" the Des Moines school system followed? A phased adoption. The demonstration phase (April-October 2015) opened with a mandatory two-day boot camp for top management, a separate project-based boot camp for the future stewards, 30-day action items, and four selected Kaizen events — so the district could see the value for itself. The tools phase followed: about 20 stewards trained through a more intensive boot camp, a steering committee formally established with its own standard work and charter, and a naming convention. QPIC worked with the district from April 2015 until it reached self-sustaining mode in mid-2017.
What is the hub-and-spoke model? QPIC's preferred organizational model for Lean adoption. At the hub is a small central Lean office — in Des Moines, a single full-time coordinator supporting 5,200 employees. The spokes are internally trained stewards (coaches) who support department heads across the organization. Above them sit the chief executive (who must have genuine hands-on Lean experience), the leadership core team, an executive champion, and the steering committee. The critical feature: ownership of the Lean journey sits with operational leadership and management, not with the Lean office. The office develops capability in others so it multiplies; it does not own or drive the effort.
Why shouldn't the central Lean team "own" Lean? Because when improvement is owned by a central team rather than by operational leaders, change stalls at the edges of the organization. If the Lean office is the only thing driving the work, the work stops the moment that office stops pushing. Harry's framing: the stewards support the Lean process; the department heads and chiefs own it. The Lean office's purpose is to hold deep expertise and multiply it through others — not to be the engine.
Why does Harry advise against calling it a "Lean program"? Two reasons. First, the word "program" signals impermanence — programs come and go and have short, defined durations. Organizations should give the effort a lasting initiative name and an elevator speech instead (examples: "Advance Phoenix," "Denton's Lean Journey," "Evolution," "Peak Performance Academy"). Second, Harry advises avoiding the word "Lean" itself where possible, because it carries the connotation of cutting to the bone, trimming fat, and eliminating jobs — the opposite of the respect-for-people message. Des Moines called its effort "continuous improvement."
Why must senior leaders go through the full training, not just an overview? Because hands-on experience is what makes the concept real. When a chief executive sits through a two-day boot camp and does actual hands-on Lean work, "the lights go on" — they connect the dots on how Lean could be impactful. A one-hour cameo overview does not produce that understanding, and the opportunity is significantly diminished. QPIC has senior leaders attend an intensive two-day workshop, with preparation work, arriving with real projects they want to work on.
What kind of executive champion does Lean adoption need? Someone at a high enough level to carry significant clout and respect, so the organization takes the effort seriously. In Des Moines, the executive champion was the chief financial officer, one of the district's top six leaders, reporting to the superintendent. The champion attends training, chairs the steering committee, and guides the effort. This works best when the top leader (superintendent, agency head, city manager) is also clearly telling their team that the organization is going on this journey and that strong commitment is expected.
What does "teach everyone to solve problems" mean in practice? Building problem-solving capability across the organization using PDCA, A3 thinking, and the 5 Whys. Harry stressed two underemphasized parts. First, the planning phase — people rush to solutions before clarifying the actual problem (he cited the Einstein line about spending 55 of 60 minutes defining the problem). Second, ensuring a good measurement system — a step rarely shown on PDCA or A3 templates. You can run improvement cycle after cycle while the real problem is a broken measurement system or the absence of shared operational definitions.
What is "dynamic data collection"? Simple tools used in the work area so that the employees doing the work capture data the same day, rather than relying on backward-looking monthly reports. It can be as basic as a picture or a printed screenshot on which workers tick off where an error or rework occurs. The patterns are never uniformly distributed — concentrations show up, which point toward root cause analysis and improvement. Beyond the data itself, the practice engages people: when workers see what's happening in real time, they start generating improvement ideas themselves.
How do you hold the gains after an improvement? By raising the standard of work and locking it in — which requires full employee engagement, not an announcement. The people doing the new work have to understand it, be engaged in it, and own it themselves, or the process slides backward. Harry named establishing and maintaining rigorous standard work as the single biggest challenge QPIC sees in organizations, and noted there are specific methodologies for engaging people in the new standard so it actually holds.
Why does QPIC emphasize leadership development through coaching rather than training? Because you cannot train your way to better leadership. A new behavior becomes an ingrained habit only through deliberate practice and coaching — nobody learns golf from television or books. QPIC's process pairs an anonymous 360-degree survey with an action plan and six monthly face-to-face coaching sessions of about 90 minutes each, observing the actual workplace, followed by a second 360 six months later. The approach has produced employee engagement improvements of 50% or more.
How should a consultant handle a client who wants to dictate the curriculum? Harry called it a genuine and sometimes unsolvable challenge. If the people setting direction are trained, practicing Lean experts, defer to them. If they've only seen a few videos or attended a couple of Kaizen events, it's a hard road — especially when a leader has "bright person's disease," too confident in their own credentials to listen to outside help. You can demonstrate early successes and provide references and hope the leader's view shifts. Sometimes it doesn't. The honest answer is that a leader adamant about running things their own way is very difficult to move.
.svg)
Copyright © 2026
Privacy Policy