THE TRAINING OF CONSTRAINTS

What happens when you start seeing the one thing that actually matters

A restaurant owner spends $40,000 remodeling the kitchen. New equipment. Faster burners. Better layout. The cooks love it. Ticket times drop from twelve minutes to eight.

Revenue does not change.

The host stand seats eight tables per hour. It seated eight tables per hour before the remodel. It seats eight tables per hour after. The kitchen was never the problem. The kitchen was loud. The host stand was quiet. Forty thousand dollars went to the loud thing.

This is what it costs to not see the constraint.

01 — Finding the one thing everything else depends on

Stop looking at what is moving. Start looking at what is waiting.

Every system has one point where things pile up. In a kitchen, it is the tickets hanging before the slow station. In a company, it is decisions stacking in one person’s inbox. In a morning routine, it is the one step that makes you late no matter how fast you do everything else.

The pile is the fingerprint. Not the noise. Not the activity. The pile.

The sales team is making calls. The engineers are writing code. The floor is running at eighty percent. That is noise. It is loud and visible and rewarding to watch. None of it tells you where things are stuck.

The pile tells you where things are stuck.

Pick one system you are inside right now. Walk through it from start to finish.

  input → __________ → __________ → __________ → output
                                        ▲
                          where is the pile?

At each stage: is work waiting here, or flowing through? The stage where things pile up is the constraint. Everything before it is producing faster than it can handle. Everything after it is starving for work.

If you cannot find the pile, the constraint is somewhere you are not looking. Someone else’s inbox. A process running in the background. A decision that was deferred so long it stopped feeling like a decision.

02 — The question that catches a $40,000 mistake

A team speeds up their part of the pipeline. Impressive numbers. But the person who reviews every release before it ships is still one person, still reviewing at the same pace. The team produced more work. The reviewer cannot process it any faster. The extra work just sits in a pile.

The team worked harder to create a bigger pile.

One question separates a real improvement from an expensive illusion:

If I added one more of this, would the final result actually change?

One more person. One more hour. One more dollar. One more approval. If the answer is no, that resource is not the thing holding back the system. It does not matter how scarce it feels or how much people complain about it.

  resource: __________     changes the final result? __________
  resource: __________     changes the final result? __________
  resource: __________     changes the final result? __________
  resource: __________     changes the final result? __________
  resource: __________     changes the final result? __________

  the one that matters: __________

Most of the list will be no. The yeses are the constraints. Usually there is one.

03 — Fixing the problem once instead of forever

A team has a review bottleneck. Work piles up waiting for approval. The obvious fix: add another reviewer.

But why is so much work arriving with problems? If the work were cleaner, the reviewer would move faster. The visible bottleneck is review speed. The actual problem is quality upstream.

The pile always tells you where things are stuck. It does not always tell you why.

When you find the pile, look one step back. Ask: is the pile here because this stage is slow, or because the stage before it is pushing bad work forward?

  the pile is at: __________
  slow here, or bad work arriving from before? __________
  what would fix it before it gets here? __________

Adding another reviewer solves the symptom. Fixing why work arrives broken solves the cause. Both shrink the pile. One of them shrinks it permanently.

04 — What that rule is actually costing you

A company needs three levels of approval for any purchase over $500. The rule was written when the company had twelve people and no finance team. The company now has four hundred people and a full finance department. The approval takes nine days. It is the single thing holding back every project that needs equipment.

The problem is not money. The problem is a rule someone wrote years ago that nobody has questioned since.

Some constraints are physical. The machine can only run so fast. The bandwidth has a ceiling. The human body needs sleep. These cannot be talked away. They need resources.

Some constraints are rules. They exist because someone decided them and nobody reconsidered. These can be removed in a single conversation.

  constraint: __________     wall or door? __________
  constraint: __________     wall or door? __________
  constraint: __________     wall or door? __________
  constraint: __________     wall or door? __________

  which door could be opened by one conversation? __________

Most of the things holding back human systems are rules, not physics. The person who cannot tell the difference spends years pushing against things that could be opened with a question.

05 — More output without more resources

When you find the real bottleneck, the instinct is to throw money at it. Hire people. Buy equipment. Expand.

This is usually premature.

A factory’s slowest machine runs sixteen hours a day. It could run twenty-four. Before buying a second machine, run the first one through lunch breaks and shift changes. Move maintenance to off-hours.

A team’s bottleneck is the one person who reviews everything. Before hiring another reviewer, ask: do reviews sit in their inbox for hours between batches? Could the queue be restructured so they never have to wait for work, and work never has to wait for them?

Most bottlenecks are running at half to two-thirds of what they could handle, because nobody has treated them as bottlenecks before.

  the bottleneck: __________
  how much of its capacity is actually used? __________
  what could it handle if nothing wasted its time? __________
  what is wasting its time right now? __________
  what would let it run without interruption? __________

Squeeze everything out of what you have before you spend. You earn the right to add capacity only after you have used what is already there.

06 — What staying busy is actually producing

A factory’s slowest machine is the CNC. The CNC must never sit idle. Every other station should produce only what the CNC can handle, at the pace the CNC can handle it.

When the station before it runs faster, it creates a pile that costs money and space. When the station after it falls behind, the CNC runs out of work. Neither is helping. Both feel productive.

This is the hardest idea to accept. The station running at fifty percent is doing its job perfectly. The station running at a hundred and twenty percent is creating waste. The hardest-working part of the system is causing the most damage.

  the bottleneck: __________

  part of system: __________     serving the bottleneck? __________
  part of system: __________     serving the bottleneck? __________
  part of system: __________     serving the bottleneck? __________
  part of system: __________     serving the bottleneck? __________

  what should slow down? __________
  what should speed up? __________

The instinct says everyone should be busy. The system says everyone should be aligned to the one thing that limits the output. These are different instructions. Confusing them is how teams stay stuck while every person in them is exhausted.

07 — What starting more is actually doing

A team has five projects running. Each one takes about a week of focused work. But nobody works on anything full-time. Everyone is splitting attention across all five. Context switching. Status meetings. Updates on everything, progress on nothing.

Eight weeks later, all five finish in the same week. The math says five weeks of work should take five weeks. It took eight. Three weeks vanished into the switching.

Run the same five projects differently. Finish the first before starting the second. One at a time. The first delivers in week one. The second in week two. By week five, all five are done. Same work. Same people. Three weeks faster. And each project delivered value the moment it finished, not eight weeks later.

The mechanism is not discipline. It is queue math.

When a system has a bottleneck, adding more work to the queue does not make the bottleneck move faster. It makes the queue longer. Every item waits longer. Nothing finishes sooner. The only thing that changes is how long everything sits.

  things in progress right now: __________
  can your bottleneck handle all of them at once? __________
  what would finish first if you stopped starting and started finishing? __________

Teams that limit how many things they work on at once routinely finish more total work than teams that try to do everything in parallel. Not by working harder. By not clogging the queue.

The counterintuitive result: doing less at once produces more in total.

08 — Seeing the next wall before it hits

Fix the slow machine in the factory. Now shipping is the bottleneck. Fix shipping. Now sales cannot keep up. Fix sales. Now cash flow cannot fund production.

The bottleneck does not disappear when you fix it. It moves.

One organism treats every new bottleneck as a crisis. Strain, breakthrough, relief, next wall, frustration. Another organism expects the move. Fixes the bottleneck. Immediately asks: where did it go? The next wall is not a surprise. It is the system doing what systems do.

Same walls. Different experience. The difference is not toughness. It is having a model that includes the move.

  current bottleneck: __________
  plan to fix it: __________
  when fixed, the next bottleneck will be: __________

The first predictions will be wrong. That is the point. The surprise tells you where your understanding of the system has a gap. After enough predictions, you start seeing the next bottleneck before it arrives.

09 — Breaking a stall without adding anything

Not every constraint is imposed. Some are chosen.

The writer who limits herself to five hundred words. The founder who will not take outside money. The chef who cooks only with what is in season.

Chosen limits work because open space is too large to move through. A blank page with no limit, no topic, no deadline offers total freedom and produces nothing. Every open option is a decision. Every decision uses energy. Enough open options and the whole thing freezes.

A chosen limit removes decisions. It says: not this. Not that. What is left is small enough to actually explore.

There is a deeper mechanism. A team given a week to finish a two-hour task will use the week. Not because the team is lazy. Because without the time constraint, every subtask gets equal attention. Every decision gets equal deliberation. Every output gets equal polish. The work expands not through malice but through the disappearance of the signal that separates what matters from what does not.

The constraint is the focusing function. The deadline tells you what to skip. The budget tells you what to prioritize. The word limit tells you what to cut. Remove the constraint and the signal disappears. The work expands to fill whatever space is available.

  what is stalled? __________
  what has too many open options? __________
  one limit to set: __________
  what signal does that limit give you? __________

The limit is not the enemy of output. The limit is the thing that makes output possible. The blank page with total freedom feels open. The blank page with a word count and a topic produces writing.

10 — The question that stops you from removing the wrong thing

The instinct when something feels limiting is to remove it. This is wrong more often than it is right.

A team removes their weekly check-in because it feels like a waste of time. Three weeks later, nobody knows what anyone else is doing. The check-in was the constraint. The constraint was producing coordination.

A person drops their budget because it feels restrictive. Six months later, no savings, no idea where the money went. The budget was the constraint. The constraint was producing awareness.

Before removing any constraint, one question: What is this holding up?

If nothing, remove it. It is dead weight. If something you depend on, it is load-bearing. Removing it takes down whatever it supports.

  constraint you want to remove: __________
  what does it hold up? __________
  what falls apart without it? __________

The shift

The person who has done this enough times does not think about systems differently. They see them differently.

In the first week, the question is slow. Where is the pile? You have to stop yourself from solving the loudest problem and look for the quiet one.

In the first month, the piles start showing up on their own. A number on a dashboard that has not moved. A person who is always busy but nothing moves after them. A process everyone complains about but nobody has examined.

In the first quarter, you walk into a room and see the constraint the way a musician hears the key of a song. Not figured out. Recognized.

Where is the pile?

Find it. Fix the constraint. Watch the system move. Find the next pile.

The mechanism this training stands next to lives in The Machinery of Constraints and The Machinery of Constraints (Business).