Field Note: Finding the Bottleneck
Date: 2025-01-20 Sector: OPERATIONS Read Time: 4 minutes
Every system has a bottleneck. The constraint that limits throughput. The thing that, if you improved it, would improve everything. The thing that, if you ignore it, makes all other improvements irrelevant.
The Observation
Client had a manufacturing operation. They'd invested heavily in automation on the production floor. State-of-the-art equipment. Impressive cycle times. And yet, output was flat.
Everyone had theories:
- "We need more floor space"
- "The equipment isn't running at capacity"
- "We need more shifts"
We walked the floor. Watched the flow. Asked one question at every station: "What are you waiting for?"
The bottleneck wasn't on the production floor. It was in the quality lab. Every batch needed certification. The lab was understaffed, working overtime, and still couldn't keep up. Production sat waiting. All that automation produced material that queued for inspection.
The Principle
The Theory of Constraints: A chain is only as strong as its weakest link. A system only produces as fast as its slowest process.
Investing anywhere except the bottleneck is waste. No matter how fast the other stations get, the bottleneck determines output. If inspection can handle 100 units per day, you can never ship more than 100 units per day, no matter how much production capacity you have.
Finding Your Bottleneck
Method 1: Follow the Queue
Where does work pile up? Where do people wait? Where do things sit before moving to the next step? That's probably the bottleneck.
Method 2: Calculate Capacity
For each step in your process, calculate how much it can handle per unit time. The step with the lowest capacity is the bottleneck.
Method 3: Ask the People
The people doing the work usually know. Ask: "What's the biggest barrier to getting more done?" Listen without defending.
What to Do With the Bottleneck
Once you've found it:
Exploit it: Make sure the bottleneck is never starved for input or blocked from output. It should never wait.
Subordinate to it: All other processes should operate at the pace the bottleneck can handle. Producing faster upstream just creates inventory.
Elevate it: If you need more capacity, invest in the bottleneck first. Not elsewhere.
Repeat: Once you fix this bottleneck, something else becomes the bottleneck. Find it and start again.
The Resolution
The client didn't need more production equipment. They needed two more lab technicians and a faster testing protocol.
Cost: ~$200K per year in salary Result: 40% increase in output ROI: Measured in weeks
The automation investment was fine. It just couldn't realize its value until the bottleneck moved.
The Meta-Lesson
The bottleneck is often not where you expect it. And it's almost never where the organization is investing.
High-performing operations don't ask "how do we improve?" They ask "what's the bottleneck?" and then improve only that.
Efficiency everywhere is waste. Efficiency at the bottleneck is leverage.