Why Managers Think a Process Works When It Doesn’t

The Problem: "It Works" Is the Most Expensive Assumption in Manufacturing

I've spent 25 years on shopfloors — from toolmaking apprenticeships to commissioning full production lines — and I hear the same thing from managers every month:

"The process works. We hit our numbers."

Here's the issue: hitting numbers doesn't mean you're not bleeding cash.

When I was setting up the Louvre production line in Sydney, the existing assembly workflow was "working." Parts were being assembled, orders shipped, customers happy enough. But when I actually stood on the floor and watched — with a video camera, not a spreadsheet — the reality told a completely different story.

What I saw:

• Operators walking 14 metres between stations to fetch parts that should have been at arm's reach
• The same part being picked up, put down, and repositioned three times before assembly
• One operator waiting 4–6 minutes per cycle while another station caught up
• Rework happening at the end of the line because errors weren't caught at the source

None of this showed up in the ERP system. None of it appeared on the production report. The process "worked" — but it was full of hidden cost.

Why managers miss it

The reasons are predictable and human:

• They manage from reports, not from the floor. ERP tells you what was produced, not how painfully it was produced.
• They confuse output with efficiency. If 100 units ship, the process must be fine — right?
• They normalise workarounds. When operators have been doing something a certain way for years, it stops looking like a problem.
• They don't measure the right things. Cycle time gets tracked. Walking time, waiting time, double-handling? Almost never.

The Cost: What "It Works" Actually Costs You

Let's put real numbers on it. Here's what I measured during a time study on a mid-sized assembly line — a line the manager said was "running well."

That's over $81,000 a year in waste on a single line — and the manager thought the process was fine.

Add scrap costs, overtime driven by inefficiency, and the knock-on delays to other jobs, and you're easily looking at $87,000–$100,000 per year in avoidable cost. On a business running $1.5M EBITDA, that's roughly 6% of your profit walking out the door.

Chart: Annual cost of hidden shopfloor waste by category. Walking, waiting, handling, and rework add up fast.

The Shopfloor Reality: What Actually Happens vs. What You Think Happens

Here's the gap I see on almost every shopfloor I walk into:

This is the fundamental problem: the process produces output, so nobody questions the process.

But output is not the same as efficiency. And in manufacturing, the gap between the two is where your money disappears.

Illustration: The difference between the report view and what's actually happening at each station.

The Solution: How to Turn Shopfloor Problems Into Measurable Cost Savings

When I set up the Louvre Line, I used a straightforward approach — no expensive consultants' frameworks, no six-month programmes. Just practical engineering thinking applied to the workflow.

Step 1: Go to the floor and watch

Put down the laptop. Stand at the line for two hours. Watch where operators walk, wait, and struggle. Time everything.

I use simple time studies — a GoPro camera, a clipboard, and the discipline to record what actually happens rather than what should happen. This is where you'll find the 11.6 minutes of waste per cycle that doesn't appear in any report.

What to measure:

• Cycle time per station (actual, not theoretical)
• Walking distance per cycle
• Wait time between stations
• Touch count — how many times is a part picked up before it's done?
• Rework frequency and where it's caught

Step 2: Map the current state — honestly

Use a Value Stream Map (VSM) to document the actual flow, including every delay, queue, and workaround. Not the idealised version — the real one.

When I mapped the Louvre line before redesign, the current-state VSM showed that only 34% of the total lead time was value-added work. The rest was movement, waiting, and correction.

Diagram: Current-state VSM showing non-value-added time vs. future-state target after redesign.

Step 3: Fix the biggest cost items first

Don't try to fix everything at once. Use a Pareto approach — tackle the 20% of causes driving 80% of the cost.

On the Louvre line, the top three fixes were:

Three changes. No major capital spend. $68,796 recovered per year.

Step 4: Standardise the improved process

This is where most improvement efforts fail. You fix the problem, but six weeks later, operators drift back to old habits because nothing was documented.

I implemented standardised work instructions at every station — clear, visual, and built into the workflow. Combined with ERP routing updates and visual controls on the line, the improvements stuck.

Step 5: Measure again and sustain

After changes went live on the Louvre line, I ran follow-up time studies at 30, 60, and 90 days. The results held — and in some areas improved further as operators refined their new routines.

Chart: Measurable improvement across cycle time, walking distance, rework, and labour cost per unit.

The DMAIC Framework: Structured Problem-Solving for the Shopfloor

The approach I've described follows the DMAIC (Define, Measure, Analyse, Improve, Control) framework from Lean Six Sigma — but applied with shopfloor common sense, not academic theory.

The DMAIC cycle: a structured approach to turning shopfloor problems into sustained cost savings.

The key is that every phase connects back to money. If you can't quantify the problem in dollars, you can't justify the fix. And if you can't measure the result, you don't know if you've actually improved anything.

Takeaway

A process that "works" is not the same as a process that works efficiently. The gap between the two is where manufacturers lose tens of thousands of dollars every year — often without realising it.

The fix isn't complicated: go to the floor, measure what's actually happening, and turn what you find into a cost number. Once you can see the waste in dollars, the path to fixing it becomes obvious.

Three things to do this week:

1. Pick your busiest line and stand there for two hours with a stopwatch or camera
2. Measure walking time, waiting time, and rework per cycle
3. Multiply the waste by your labour rate — then decide if "it works" is good enough

Ready to Find the Hidden Cost in Your Process?

Most manufacturers I work with are surprised by what a simple shopfloor study reveals. If you want help identifying where your processes are quietly costing you money, I offer a short CostDown Audit — a focused review of your shopfloor operations to uncover hidden savings and put real numbers on the opportunity.