Process Improvement Fails When It Skips the People Who Do the Work
Six months after a major operational overhaul, a familiar scene plays out in boardrooms everywhere. The workflow diagrams on the screen look pristine. The projected savings look impressive. Down on the facility floor, order fulfillment times are slipping, defect rates are climbing, and supervisors are spending half their shifts managing informal workarounds their own people quietly rebuilt. The map was elegant fiction. It never had to survive contact with the floor, because the floor was never in the room where it got drawn.
The cause is rarely a lack of analytical skill or good intent. It is a structural flaw in how leadership approaches process improvement, designing new systems in a conference room without the people who execute the work every day.
Where Process Improvement Actually Starts
Most process improvement projects begin with a kickoff meeting that includes a sponsor, a few directors, an outside consultant if the budget allows for one, and almost nobody who runs the process day to day. That is the first design flaw, and it happens before a single workflow gets mapped.
Top-down redesigns treat a business process like a predictable mechanical system. Leaders study high-level data, identify the bottleneck, and mandate a new workflow from above. That view has value, but it misses the friction that determines whether a process actually works. The people running the machinery, handling the calls, or processing the inventory carry information no dashboard captures. They know which vendor shipment routinely arrives mislabeled. They know the software screen that needs four extra clicks to clear a routine error. They know the exception that happens so often it should not be called an exception anymore, and the informal channel that keeps materials moving when the official one stalls.
Taiichi Ohno built an entire manufacturing philosophy on this exact problem, decades before "process improvement" became a category on a consulting proposal. His approach at Toyota treated the factory floor as the primary source of design intelligence, not a destination for instructions produced somewhere else. He argued for standards built by the people executing them rather than imposed from a level above, and made the case directly in his own writing on the subject:
"Standards should not be forced down from above but rather set by the production workers themselves." — Taiichi Ohno, Toyota Motor Corporation, from Toyota Production System: Beyond Large-Scale Production (1988)
That line was written about a factory floor. It applies just as directly to a claims department, a supply chain function, or a client onboarding process at a professional services firm. A standard that nobody who works the process helped set is a standard that will quietly erode the moment leadership stops watching.
What Gets Lost When Design Skips the Floor
When leadership excludes frontline staff from the design phase, the damage shows up in three distinct places. Theoretical efficiency replaces practical execution, since a workflow built on ideal parameters looks elegant on paper and fails the first time reality varies from the plan. Resistance turns passive rather than vocal, as employees comply outwardly while quietly reverting to the old habits the moment supervision looks away. And the feedback loop disappears entirely, because a process handed down as a finished mandate gives nobody a reason to flag what is not working until the workaround is already built.
Ohno's own instinct ran the opposite direction. He placed far more weight on direct observation of the floor than on the reports and dashboards produced about it, on the theory that data describes what happened while a walk through the actual work reveals what is happening now. That bias toward ground truth over abstraction is the same instinct behind a Gemba walk, a practice with roots in Japanese manufacturing where a leader goes to where the work actually happens and observes it firsthand rather than reviewing it secondhand. It is worth building into a redesign before the first workflow gets drawn rather than after it fails.
The Numbers Behind What Practitioners Already Suspected
McKinsey & Company, in research titled The People Power of Transformations, found something worth sitting with. Among large-scale change efforts that failed to engage line managers and frontline employees, only 3 percent of respondents reported success. Where those groups were engaged, success rates climbed to 26 and 28 percent respectively.
That gap is not small. It is the difference between a redesign that sticks and one that gets quietly dismantled by the people who have to live inside it every day.
Executives tend to treat frontline input as a courtesy step, something to check off after the real design work is finished. The research suggests the opposite. Involvement is not a courtesy. It is a leading predictor of whether the new process survives its first year. The input session that gets scheduled for the last week of a project should be scheduled for the first, because design decisions made before that input arrives tend to harden fast, and by the time the feedback session happens, the redesign team is defending choices instead of genuinely reconsidering them.
The Objection Every Executive Raises First
The pushback is predictable and reasonable on the surface. Pulling frontline employees into design sessions takes them off the floor, adds coordination overhead, and slows a project that already has a deadline attached to it. In a market where speed gets rewarded, adding steps to a redesign reads as the wrong move.
That objection confuses two different kinds of speed. There is speed to a finished deck and speed to a process that actually holds under real conditions. A redesign that skips the people doing the work can hit the first kind of speed easily. It rarely hits the second, because the rework that follows a failed rollout costs more time than the input would have taken up front. Slower at the start is not the same as slower overall.
Why Committees Are Not the Answer Either
Faced with the case for frontline input, most leaders overcorrect. They respond by adding a large committee, a rotating advisory group, or a series of listening sessions that stretch the timeline by months and produce a report nobody reads. That is not what effective input looks like. Effective input is narrow, specific, and built into the design work itself rather than bolted onto the end of it.
A better approach includes:
Conducting a structured Gemba walk before process mapping begins, so executives and process engineers observe the actual workflow alongside the people running it rather than drafting from a summary of it
Pulling two or three respected practitioners into the actual design sessions to stress-test proposed changes in real time, instead of holding a separate feedback round afterward
The proposed redesign should be tested against the worst week of the year, not the average week, since averages hide the exceptions that break new processes
Opening a direct, fast channel for frontline staff to report friction during the first thirty days of rollout, with real authority to trigger an adjustment rather than wait for the next review cycle
The new process gets piloted with the same group that helped design it, before it rolls out broadly
Notice what is missing from that list. There is no steering committee. There is no six-month engagement plan. The goal is proximity to the work, not the appearance of inclusion.
What Changes When the People Doing the Work Help Design It
I have watched redesigns of similar scope and budget land in opposite places depending on a single variable, whether the people running the process helped build the fix or simply received it. The ones built with the team tend to hold. The ones built for the team and handed over as a finished product tend to erode, quietly and predictably, as the old workarounds return.
That difference tends to track leadership behavior as much as process design. Organizations that stay close to how work actually happens, rather than managing it from a distance through periodic reports, are the ones where a redesign becomes something employees own rather than something they tolerate.
Process improvement that starts with the people closest to the work moves slower in the design phase and faster in the adoption phase. Process improvement that skips them moves fast on paper and stalls the moment it meets reality.
Executives who want the second outcome should ask themselves a direct question before the next redesign begins. Who is in the room, and who should be?
Where Redesigns Actually Get Tested
Aspirations Consulting Group was built for the moment when a redesign looks finished on paper and untested everywhere else. Operations decisions made without the people who run them tend to resurface as the same problem wearing a new name six months later, and catching that gap early is far less costly than rebuilding a process twice. If your organization is heading into a redesign, a reorganization, or any initiative that depends on the people doing the work to actually adopt it, a confidential conversation with ACG is a reasonable next step. Reach Aspirations Consulting Group at https://www.aspirations-group.com.
Every Weekday, One More Way to Think Ahead
ACG Strategic Insights publishes each weekday to more than ten million current and aspiring executives around the world, covering the decisions senior leaders are working through right now. Request a complimentary subscription to ACG Strategic Insights at https://www.aspirations-group.com/subscription and keep the perspective coming.
Thanks for reading!
~ Jerry Justice
Living to Serve, Serving to Lead™
Comments