Fixing One Bottleneck at a Time: The Theory of Constraints Applied to Solo Work
Published
The slowest step sets the pace for everything
The Theory of Constraints holds that the output of a whole workflow is set by its slowest step. That step is called the constraint. Speeding up any other step changes little, because work still piles up at the constraint.
Physicist Eliyahu Goldratt popularized the idea in The Goal (1984), a business novel he wrote with Jeff Cox. The setting is a factory, but the logic fits any work that moves in sequence: taking an inquiry through to delivery, or turning an idea into a published article.
The five focusing steps
Goldratt laid out a five-step routine for dealing with a constraint:
- Identify the constraint.
- Exploit it: make sure it never wastes time.
- Subordinate everything else to the constraint's pace.
- Elevate the constraint: add capacity by hiring, buying tools, or outsourcing.
- If the constraint moves, go back to step 1 so that old habits do not become the new limit.
The order matters. Step 4 usually costs money, so try steps 2 and 3 first. They often free more capacity than people expect.
Finding it in your own work
Start by listing the steps in one workflow. For a service business that might be inquiry, intake call, quote, production, review, and invoice.
Then look for where work waits. Queues form in front of the constraint, while the steps after it often sit idle. If projects stack up before the quote stage, quoting is your candidate.
You can add rough numbers: how long each step takes and how many items are stuck before it. Precision is not required. Often your own sense of where things jam is right. Sometimes the constraint is your own time. In that case, check whether hours that should go to the constrained step are leaking into other tasks.
What removing a bottleneck can look like
Suppose quoting is the jam. For step 2, protect a fixed block of time for quotes so other errands cannot eat it. For step 3, change the intake stage so it gathers everything the quote needs in the first exchange, which cuts back-and-forth.
If you reach step 4, options include building reusable quote templates, or handing part of the drafting to a tool or a contractor. Before that, check whether steps 2 and 3 were enough.
Once one constraint is eased, another appears. After quoting, production may become the slowest step. That is step 5 at work. Do not stop after one fix; review the flow every few weeks.
Limits to keep in mind
The approach suits work that flows in sequence. If your tasks are independent of each other, or differ greatly every time, the constraint is harder to pin down.
Also, "exploit the constraint" does not mean working yourself to the limit. The goal is to remove wasted time at the constrained step, not to add hours to your day.
Goldratt's wider claims about accounting and production planning have drawn debate, and not all of them have been tested equally well. The most practical use is as a way to find where your work jams.
What tends to happen after you ease a constraint
When a constraint opens up, work that was stuck in front of it can flood the next step. If you speed up quoting, production requests may arrive all at once and put deadlines at risk. Check the next step's spare capacity before you make the change, and consider throttling how much work enters.
Another trap is keeping the non-constrained steps busy. Idle time at those steps is not waste; it is a result of matching the constraint's pace. Filling it with new work only lengthens the queue in front of the constraint. Spend the spare time supporting the constraint instead, for example by building the quote templates.
When the constraint is outside your workflow
Sometimes the constraint lies in the market or with a client, not in your own work. If inquiries are scarce, tuning your steps will not raise output. The constraint is then demand or referrals, and you look somewhere else. Keep in mind that a constraint can sit outside the flow you are mapping.
What to do next
- Pick one workflow, such as order to delivery, and list five to eight steps.
- In front of each step, write how many items or days are waiting.
- Choose the step with the longest wait as your candidate constraint.
- Remove one task that gets in its way, or block a fixed time slot for it.
- Review the flow in two weeks and check whether the jam has moved.