Stephanie Hill opened her session with a clean admission about her own learning curve. Twenty years of continuous improvement work — half of it in manufacturing, half of it in office and transactional environments — taught her that the move from one to the other is not the move most CI practitioners assume it is.
When she first attempted the shift, she described her approach as drag-and-drop. Take the Lean Six Sigma tools that work in manufacturing. Pick them up. Drop them into the office. Expect people to run with them the same way the manufacturing teams did. As you might guess, she said dryly, there were quite a few fumbles along the way.
The fumbles, over time, sorted themselves into a pattern. The same kinds of failures kept recurring. The same kinds of conversations kept producing the same kinds of dead ends. Eventually she could see the structure underneath the failures, and the structure resolved into three specific differences between manufacturing environments and office environments that account for most of what goes wrong when CI practitioners assume the tools transfer cleanly.
The three differences aren't surface features. They're structural conditions about how people relate to their work, how the work moves through time, and what kinds of waste are actually consuming organizational effort. Each one demands a different facilitation approach. Each one breaks the standard manufacturing playbook in predictable ways. And each one, once you can see it, makes the failures stop looking like failures and start looking like the natural consequence of applying the wrong tool to the wrong environment.
The session was Hill's working summary of what those three differences are and what to do about each of them.
Stephanie Hill is the owner of Light Bulb Moment Consulting in Des Moines, Iowa. She has over 20 years of experience applying quality and continuous improvement across manufacturing, retail, insurance, and healthcare industries. She holds certifications in Strategic HR Leadership and Scaled Agile, a Master Black Belt in Lean Six Sigma, a master's degree in Public Health, and a bachelor's degree in Chemistry. Her focus is on helping small businesses and nonprofits improve their effectiveness and efficiency through the combination of people, processes, and performance.
The first structural difference Hill identified is how closely people feel personally tied to the work they do.
In manufacturing, the relationship to the work tends to be intermediated. Hill's mental image is hands — work units moving across people's hands. The work is tangible. You can touch it. It moves. Change is frequent. Experimenting is common. People might change shifts. They might change what they're producing. There's a distinct difference between the person and the item. The work passes through; the worker remains.
Hill acknowledged the exceptions. Skilled labor in manufacturing often carries a stronger sense of identification with the craft. Leadership in manufacturing often identifies strongly with the operation. But for the bulk of frontline manufacturing work, the relationship is one where the work and the worker can be psychologically separated, which makes it easier for the worker to be objective about whether the process is working well.
In the office, this separation usually doesn't exist. The work is often not tangible. It involves creative thinking, relationship building, judgment, communication — none of which is easy to point at and say "this is the thing I produced today." The person may have spent years studying for the role. They may have spent more years practicing it. Change is infrequent, and when it does happen it usually feels like the personal responsibility of the individual or their leader rather than a routine adjustment to a flowing system.
The result is that the work and the worker are tightly bound. Telling someone in the office that a process isn't working is operationally different from telling someone on a manufacturing line that a process isn't working. The first lands as criticism of the person. The second lands as information about the process.
Hill made this concrete with two scenarios that she ran past the audience using a live polling tool.
Scenario one: imagine you just made a mistake while completing a mortgage form. The mistake has cost a potential buyer their dream home. Your manager comes to you and asks: "Why did this happen?" In one word, describe how you would feel.
The audience answered in real time. Lousy. Stressed. Guilty. Shame. Disappointment. Frustrated. Terrible. Regretful. Blame system fail.
Scenario two: a customer called customer experience and reported they had cut themselves on a defective product. The representative replaced the product and paid two thousand dollars in medical bills. Meanwhile, the shift supervisor gathered the crew that assembled the product that day and asked: "Why did this happen?" Describe the approach the team would take.
Notice the difference in the question. In the first scenario Hill asked about feelings. In the second she asked about the approach. The asymmetry wasn't accidental. It mirrored what actually happens in the two environments. In the office setting we naturally look at the individual — individual accountability is the default frame. Individual accountability is useful for decision making but produces predictable defensiveness when applied to problem investigation. In manufacturing, the same situation gets approached as collective ownership. A defect happens. You get the crew. You question the process, the tools, the materials. You don't single out a person.
The word "why" is what does the damage. Why feels personal. Why triggers defense. The same word that produces useful diagnostic information in a manufacturing setting produces emotional shutdown in an office setting, because the person on the receiving end isn't separating themselves from the work the way the manufacturing worker can.
The second structural difference is about time.
In manufacturing, work units typically move through processes in a short window. Cycle time is fast. A part enters the line, gets transformed, exits the line. Even in batched environments, the cycle of work for any individual unit is usually measurable in hours or days.
In the office, work units often take weeks or a month to move from start to finish. A loan application enters the system. It passes through underwriting, verification, approvals, document preparation, closing coordination. The full cycle from initiation to completion stretches across many handoffs and many days. The cycle time is long.
Hill acknowledged the exceptions in both directions. Manufacturing has long-cycle environments — aircraft assembly takes much longer than a process step. Office work has short-cycle environments — a customer service center processes calls in minutes. The generalization isn't universal. It is, however, directionally correct often enough that the consequences are worth taking seriously.
The frequency difference produces three downstream effects on CI practice.
Data collection becomes harder. Long cycle times mean fewer completed work units to observe, fewer data points to analyze, longer waits between baseline measurement and post-improvement measurement. The same baseline that takes a manufacturing team a week to establish might take an office team a quarter. Data still matters — Hill noted she was preaching to the choir on this point — but the path to getting it is structurally slower.
Going to the gemba becomes harder. When work happens at a slower cadence, or unpredictably, watching the process unfold becomes difficult. You can't reliably schedule observation time around a process that runs once a week and then sits idle. The CI team that wants to go observe the work has to either wait for the work to happen or build observation into how the work is naturally structured.
Momentum becomes harder to sustain. A kaizen event in manufacturing might compress the full cycle — diagnose, design, implement, reflect — into a single week. The team experiences the result of their work inside the same event where they did the work. An equivalent office improvement might take months to fully implement. The team's motivation has to stretch across that span. Participants may shift focus to other priorities that arise. Leaders may pull attention to competing initiatives. Even once an improvement is implemented, the team may rush past the reflection step to deal with whatever has accumulated in their inbox during the implementation period.
The Q&A on this point produced a useful audience-level data point. Hill polled the group on how they were currently going to the gemba during COVID. The top response was virtual tools. In-person was second. A few had stopped going entirely. A few had never done it in the first place. The virtual-first reality of office observation during the pandemic, Hill noted, was actually well-suited to office work — much of which had moved to environments where the only practical observation method was virtual anyway.
The third structural difference is which wastes are doing the damage.
Hill walked through the standard eight wastes briefly — defects, overproduction, waiting, non-utilized talent, transportation, inventory, motion, extra processing — and noted that audiences typically identify inventory as the dominant waste in manufacturing. That answer is correct for manufacturing. It is mostly irrelevant in the office, because the same dynamic that produces inventory in manufacturing — physical work units accumulating between process steps — doesn't happen in most office work in a way that maps cleanly.
What does happen in the office is a different set of waste patterns that the standard eight-waste taxonomy doesn't quite capture. Hill listed the ones she sees most often.
Miscommunication during handoffs. Information gets lost, dropped, or distorted as it moves between people. The classic illustration is the telephone game — whisper a message at one end, it comes out different at the other end. Hill has run this exercise with audiences and has never seen the original message survive intact. In actual office work the cost is real money and real customer pain, but the structural mechanism is the same.
Work not balanced. Some people have too much. Other people don't have enough. The misallocation produces both burnout and underutilization in the same organization simultaneously.
Unclear roles and responsibilities. This produces duplication of work as multiple people attempt the same task, and it produces decision-making bottlenecks as decisions wait for an unclear owner to surface.
Failure to optimize manual versus electronic processes. The default move in many organizations is to automate. Hill's caution: automating a bad process makes the bad process faster and more expensive, not better. Simplification has to precede automation, or you've just engineered a more efficient way to produce waste.
Excessive inspections and approvals. Particularly when they aren't aligned with the actual risk involved. The inspections and approvals consume time the customer is waiting through. Many of them exist for reasons that have outlived their original justification.
Unread documents and emails. Hill shared an example from a previous company where she received 1,500 unnecessary emails in her first month. When she calculated the cost across all the recipients of those emails, it came to roughly $40,000 per year that the company was losing on people processing emails they shouldn't have been receiving in the first place.
Work done in series rather than parallel. Not always avoidable, but often. Parallelizing what can be parallelized produces faster outcomes for the customer.
Work too complicated or too simple. Hill referenced the Yerkes-Dodson inverted-U from 1908 — there's a sweet spot for human performance where complexity is high enough to engage but not so high as to overwhelm. Office work that lands outside this sweet spot in either direction produces predictable performance problems.
Performing all work within the critical process. This one Hill clarified during Q&A. The point isn't over-processing. The point is timing. If work that doesn't have to happen during the customer-facing process is happening during the customer-facing process, the customer is waiting through activity that could have been done in advance. Her example: calling a company to change an address, and being kept on the phone for 20 minutes while the representative does work that could have been done after the call ended. The single-minute exchange of dies (SMED) framework from manufacturing — separating internal from external setup — applies directly. What can be done outside the critical window should be moved outside it.
Hill grouped these wastes into three categories that capture most of what goes wrong in office work: communication, alignment, and decision-making.
Poor communication is the volleyball problem. Two players go for the ball without coordination. Either both go and collide, or both back off and the ball drops. In an office, this produces duplicated work or missed work. Customer needs get overlooked. Without open and honest communication, resentment accumulates and turnover follows.
Poor alignment has two forms. Vertical misalignment is staff who don't understand the mission, vision, values, or high-level measures of the company. Horizontal misalignment is teams working in silos, competing for resources, or pulling against each other rather than together. Both produce predictable inefficiency.
Poor decision-making has multiple causes. Organizations that have historically punished wrong decisions produce employees unwilling to make decisions at all. Roles and responsibilities that don't include explicit decision-making authority produce confusion about who can decide what. Cultures that value hierarchy push every decision upward, which both delays the work and stunts the development of mid-level and frontline staff who never get the practice of deciding.
Hill's adjusted approach for the association problem rests on three practices.
Pre-interviews. Before any improvement event, Hill spends time individually with the people who will be in the room. She wants to gauge their buy-in. She wants to learn their prior experience with improvement work, including any past experiences that may have soured them on it. She documents barriers people share informally. And she builds the relationship and the trust that will be needed in the room.
The reason this matters: talking about your work in front of other people is intimidating. It creates vulnerability. If your association with the work is strong — as it almost always is in office settings — then putting your process up for examination feels like putting yourself up for examination. The pre-work is what makes the in-room work possible.
Roles versus names in mapping. When mapping a process with a team, Hill uses roles ("loan processor," "underwriter") rather than individual names. The reason is simple. If a process step is labeled "inefficient" and your name is sitting next to it, you experience that as a personal accusation. If the same step is labeled "inefficient" next to a role, the work can be discussed without the personal sting.
Two additional benefits. People are more willing to suggest changes to their own work when their name isn't attached. And if only one person from a multi-person team can attend the event, the role-based labeling means the work being discussed represents everyone who does that work, not just the one person who happened to be in the room.
Root cause approach beyond the Five Whys. Hill has found the Five Whys less effective in office work than in manufacturing for two reasons. Office problems are typically a blend of contributing factors rather than a single linear causal chain — ask five different people why something happened and you'll get five different answers, all of them at least partially correct. And the word "why" itself triggers defensiveness in the office context, exactly as the mortgage form scenario demonstrated earlier in the session.
Her alternative is a 6M brainstorming approach — human, machine, mother nature, measurement, method, materials. Each person silently writes one contributing factor under each M before any discussion happens. This produces more openness, more creativity, and substantially less defensiveness than the same investigation conducted through repeated "why" questions.
During Q&A, Mark surfaced a related move that experienced facilitators sometimes use as a softer alternative to "why" — phrasing like "what factors might have led to that" or "what allowed that to occur." The phrasing is slightly less loaded. Whether it's enough depends on the underlying trust in the organization. In environments where psychological safety is genuinely low, even the gentler phrasing may not be enough; the facilitator has to read the room and choose accordingly.
The adjustments here address each of the three downstream effects.
On data collection. Don't wait for perfect data. If you know anecdotally that there's a problem worth working on, start the work and build the data collection plan into it. Define a data collection plan early so you know what you're capturing, where you're capturing it, and how you'll know whether the improvement worked. Use the virtual tools that have become widely available — live polling, idea boards with anonymous voting, shared documents — to capture inputs that don't require everyone to be in the same room at the same time.
On going to the gemba. Hill's adjustments work around the predictability problem. Have team members record themselves on video doing the process, and share the recordings during the event. Have people notify the facilitator when the work is happening so observation can be scheduled around the natural cadence rather than against it. Distribute tick sheets or time-study tables so the work being observed includes its own data collection. Benchmark against organizations doing similar work — videos online have made this easier than it used to be. During the event itself, bring multiple people who do the process so the mapping captures genuine variation rather than one person's version of the standard.
On sustaining momentum. Hill's approach is to build a structured monitoring plan into the implementation. A visual progress mechanism. Action items with owners. Data on whether the situation is actually improving. Pre-defined responses for items that go off track. The team meets at least weekly initially, then less frequently as the implementation matures. Each meeting reviews status, data, barriers, and the team's ongoing motivation. The facilitator gauges whether expectations are being met or whether water-cooler frustration is building unspoken. The business owner is prepared to address concerns as they surface.
A practical principle Hill emphasized: set the expectation of imperfection at the start. PDCA is cyclic for a reason. Learning from the process and adapting it isn't failure — it's the work. Teams that expect to get it right the first time get demoralized when they don't. Teams that expect to iterate stay engaged through the iteration.
The adjustments here align with the three categories of office waste — communication, alignment, and decision-making.
Look at company strategy deployment. How is strategy actually moving through the organization? Is it moving at all? At what cadence? Without visible strategy deployment, vertical alignment is structurally impossible.
Partner for open culture. This usually takes more than the CI function alone. Work with HR. Work with internal communications. The shift to a more open and honest culture requires buy-in across the influential parts of the organization, not just within the improvement team. And be a visible role model for vulnerability and learning from failure. The leaders who model what they're asking of others get different results than the leaders who only ask.
A mix of improvements across levels. Some improvements happen at the individual level. Some happen at the team level. Some happen at the organizational level. Working all three creates vertical and horizontal alignment that none alone can produce.
Document roles and responsibilities, including decision-making authority. Make explicit who decides what, at what level. The clarity removes a significant source of friction in office work that no amount of methodology training can substitute for.
Map to highlight office waste. Hill uses a mapping approach she learned from Steve Dickinson at PQS that visually flags the waste patterns common to office work. Yellow for intersection or determination steps — the decision points that the standard "happy path" SOP often skips over but that produce the bulk of process variation. Green for handoffs — the points where information is most likely to be lost or distorted. Pink for waits — annotated with rough time ranges, validated later, but included from the start because the cumulative weight of wait times across a process produces an aha moment for teams that previously thought the work was moving faster than it was.
The work Hill described is facilitation work. It depends on building trust, asking the right questions in the right way, structuring conversations so people can engage without defensiveness, and helping teams sustain attention across long implementation cycles. No platform performs this work on a facilitator's behalf. The pre-interviews have to happen between humans. The mapping conversations have to be facilitated. The decisions about whether to keep going have to be made by leaders willing to look at the data and engage with what it shows.
What infrastructure does in this context is reduce the structural friction that makes office improvement work fail to sustain. The long cycle times Hill identified as one of the three key differences are a particular problem for tracking systems built around manufacturing-style cadences. Improvements that take months to fully implement need a place to live for those months — a place where their status is visible to leadership, their action items are tracked through completion, and the data on whether the improvement is working can be captured alongside the implementation itself.
The momentum challenge Hill flagged — teams losing focus across long implementation windows — has a structural component that infrastructure can help with. When improvement work is visible to the leadership team continuously rather than only at the start and end, leaders can engage at the moments when engagement matters rather than only at the kickoff and the final report. The "I forgot we were even working on that" failure mode, which is endemic in office improvement work, becomes structurally less likely when the work is surfaced in regular leadership rhythms rather than buried in a slide deck that only gets revisited quarterly.
The categories of office waste Hill identified — communication, alignment, decision-making — are also categories where infrastructure can do real work. Improvements that surface in one part of the organization need to be visible in other parts where the same patterns are operating, or the wheel gets reinvented in parallel by teams who never knew their problem had already been solved next door. The volleyball problem Hill described — two players reaching for the same ball, or both backing off — has an organizational analog that visibility infrastructure can help address, by making it clear who is working on what and where the gaps are.
Hill mentioned in Q&A her own experience using KaiNexus at a previous company, where the team used it in manufacturing in a relatively simple kiosk-style configuration and in the office for project management, ideas, and a broader set of improvement work. Her observation was that the platform was flexible enough to fit both environments, and that the team supporting it was responsive to challenges as they came up. The flexibility matters in light of the three differences she identified. The same tracking and visibility capabilities that work for manufacturing have to bend to fit the longer cycles, more individualized work patterns, and different waste categories of office work. Infrastructure that requires office improvement to behave like manufacturing improvement won't get used, which is the same dead end the drag-and-drop methodology produces.
None of this changes what Hill was teaching. The facilitation is the work. The pre-interviews are the work. The decision to use roles instead of names is the work. The discipline of running 6M instead of Five Whys when the room can't handle Five Whys is the work. What infrastructure does is keep the work visible across the long implementation cycles where office improvement actually has to live.
What are the three key differences between Lean in manufacturing and Lean in the office? First, association to work — manufacturing workers can usually separate themselves from the work passing through their hands, while office workers identify closely with work that often took years of education and practice to learn. Second, frequency of work units — manufacturing typically has fast cycle times where a work unit moves through a process quickly, while office work often takes weeks or a month to fully complete a single cycle. Third, significance of waste — the waste patterns that matter in manufacturing (especially inventory) are different from the waste patterns that matter in the office (especially communication, alignment, and decision-making failures).
Why doesn't drag-and-drop Lean work in the office? Because the tools developed for manufacturing assume conditions that don't hold in office work. Standard Five Whys analysis assumes the question "why" can be asked without triggering defensiveness — which is reasonable in manufacturing where the worker can separate from the work, and unreasonable in the office where they can't. Standard kaizen-event compression assumes the full cycle of improvement can be contained within a week — which works for fast-cycle-time processes and breaks down when implementation takes months. Standard waste taxonomy emphasizes inventory — which is correct for manufacturing and largely irrelevant for the office, where communication failures, misalignment, and decision bottlenecks are doing most of the damage.
What's the practical implication of "association to work" being stronger in the office? That asking someone to look critically at their work feels, to them, like being asked to look critically at themselves. The same question that produces useful diagnostic information in a manufacturing setting can produce emotional shutdown in an office setting. Facilitators have to adjust how they ask questions, how they label process steps, and how they create psychological safety in the room. The work isn't to make office workers behave like manufacturing workers. It's to use facilitation techniques that respect the structural reality of how office workers relate to their work.
Why use roles instead of names when mapping office processes? Because labels stick. If a process step is labeled "inefficient" next to a person's name, the person experiences that as a personal accusation. If the same step is labeled next to a role, the work can be discussed without the personal sting. People are also more willing to propose changes to their own work when their name isn't attached. And if only one person from a multi-person team can attend the event, role-based labeling means the work being mapped represents everyone who does that work, not just the one person in the room.
Why does the word "why" trigger defensiveness in office settings? Because asking someone in the office why something happened often lands as asking why they personally made a mistake, given the strong association between office workers and their work. The mortgage form scenario in the session demonstrated this — asking someone "why did this happen" after a costly mistake produced words like guilty, shame, stressed, regretful. The same word in a manufacturing setting, asked of a crew that produced a defect, produces a structural conversation about process and tools. The difference isn't about the people. It's about the work environment they're operating in.
What is the 6M alternative to the Five Whys? A brainstorming approach using six categories: human, machine, mother nature, measurement, method, and materials. Each person silently writes one contributing factor under each category before discussion begins. This surfaces more contributing factors than a linear causal chain captures and substantially reduces the defensiveness that "why" questioning produces. The approach assumes office problems are typically a blend of contributing causes rather than a single root cause — an assumption Hill has found to be correct more often than not in office work.
What are the most common forms of office waste? Miscommunication during handoffs (the telephone-game pattern). Work not balanced across team members. Unclear roles and responsibilities. Failure to optimize manual versus electronic processes (automating bad processes). Excessive inspections and approvals that aren't aligned with actual risk. Unread documents and emails. Work done in series when it could be done in parallel. Work that is too complicated or too simple for the person doing it. Performing work within the critical customer-facing process that could have been done outside that window.
How should office improvement teams handle long cycle times? Don't wait for perfect data — start the work and build data collection into the implementation. Use virtual tools to capture observations that can't be scheduled live. Have team members record themselves doing the process so observation can be asynchronous. Have multiple people who do the work participate in mapping so genuine variation is captured. Build a structured monitoring plan with visible progress, action items, data on improvement, and pre-defined responses for items that go off track. Meet weekly initially, then less frequently as the implementation matures. Set the expectation of imperfection at the start so iteration is treated as the work rather than as failure.
What is the volleyball analogy for office communication? When a ball comes onto your side of the court, the standard practice is for one player to call "got it" so other players back off. Without that explicit communication, two players might collide reaching for the same ball, or both might back off and let the ball drop. In the office, the equivalent is two people duplicating work because neither knew the other was doing it, or work falling through the cracks because neither party realized it was theirs. The fix is explicit communication about ownership, the same way a volleyball team uses explicit communication about who has the ball.
How does Stephanie Hill handle building emotional safety in events? Pre-interviews before any event to gauge buy-in, learn prior experience, document barriers, and build trust before people enter the room. In the event itself, virtual tools (anonymous voting on idea boards, live polling) let less-vocal participants contribute without having to speak up. Sticky notes — physical or digital — let people write down ideas individually before any group discussion. Using roles instead of names in mapping removes personal exposure from process critique. Facilitator's job, in Hill's words, is to make sure people feel emotionally safe and open and honest while they are in the room.
Why does automating a bad process cost the company money? Because automation amplifies whatever the underlying process produces. A bad process produces defects, waste, and rework slowly when run manually. Automated, it produces defects, waste, and rework faster — at the cost of the automation investment on top. Simplification has to precede automation. The path is: streamline first, then automate the streamlined version. The reverse order ends up automating waste.
What is the SMED principle and how does it apply to office work? SMED (Single Minute Exchange of Dies) is a manufacturing technique for separating internal setup (which must happen while the machine is stopped) from external setup (which can happen while the machine is still running). The office equivalent is separating work that must happen during the customer-facing process from work that can happen before or after it. Hill's example was a customer service call where 20 minutes of the customer's time was consumed by work the representative could have done after the call ended. The principle: anything that doesn't have to happen during the customer's window shouldn't happen during the customer's window.
![[Watch Now] How to Leverage Lean for Long-Term Success](https://no-cache.hubspot.com/cta/default/326641/eb5ac7b8-b040-48b7-810e-1589561ffff9.png)
.svg)
Copyright © 2026
Privacy Policy