Too Many Priorities? Find the One Constraint Holding Your Business Back

Find the one step where work waits, then make the rest of the business serve it

The Goal, a business novel by Eliyahu M. Goldratt and Jeff Cox, explains the ideas behind the Theory of Constraints, which Goldratt developed ¹. Goldratt's five focusing steps are: identify the system's constraint(s), decide how to exploit them, subordinate everything else to that decision, elevate the system's constraint(s), and return to step one if the constraints were changed ². We have turned those five steps into five moves you can make with last month's numbers and a sheet of paper. Each step below says what to do and how you know it worked.

Put last month's number on every stage from first lead to cash collected

Do this. Draw four boxes in a row: new leads, sales, work delivered and cash collected. In each box, write last month's number, and next to it how long work sat waiting before that stage moved it on.

It worked when every box holds a real number from last month, and you can show where each number came from.

Why. Our rule of thumb is that numbers come before opinions: if a box has no number, getting that number is the first job. In the Marine Corps maintenance case later in this article, management went looking for the constraint by asking people. When management asked across the organization where the bottlenecks were, opinions varied, and every major activity in the Center was believed to be an important bottleneck by at least someone ². We read that as a warning for a business of any size: ask your team where you're stuck and you can get a different answer from each person.

Circle the one box where work waits longest or the number drops most

Do this. Circle one box: the stage where work waits longest before it moves, or where the number falls most from the box before it. If two boxes compete, circle the one where work waits longer.

It worked when you can say in one sentence which box is circled and which number put it there.

Why. The TOC Institute uses the image of a chain: no chain can be stronger than its weakest link, and every chain has one, and only one, weakest link ³. The TOC Institute puts it this way: the output of the constraint governs or restricts the output of the organization as a whole ⁴. In our view, that is why one circle beats a long list: work on any other box and the business delivers no more than before.

Example. In the Federal Reserve Banks' 2025 Small Business Credit Survey, reaching customers and growing sales was the most common operational challenge, reported by 57 percent of employer firms, followed by hiring or retaining qualified staff at 46 percent ⁵. In the same survey, 30 percent of employer firms reported supply chain issues, 29 percent utilizing technology, and 11 percent no operational challenge at all ⁵. The 2025 survey was fielded from September 3 to November 14, 2025, and yielded 6,525 responses from small employer firms with 1 to 499 employees ⁵. We read that split as a reason to test both sides before you circle anything: whether you are short of buyers, or short of the people and hours to serve them.

Get more out of the circled box before you spend money on it

Do this. List everything that slows the circled box: rework, waiting for answers, idle time, and jobs it does that another stage could do. Remove the cheapest of these first, and plan the box's week at 80 to 85 percent of its available hours.

It worked when the circled box moves more work over the next two weeks than its rate last month, with no new hire and no new software.

Why. For the exploit step, the TOC Institute advises maximizing the utilization and productivity of the constraint, not of the non-constraints ⁴. Robert Kaplan and Steven Anderson suggest, as a rule of thumb, assuming that practical full capacity is 80 to 85 percent of theoretical full capacity ⁶. We read that as a planning rule for the circled box: a week planned at every available hour leaves no room for the job that runs late.

Example. Tesla said it believed that improving utilization of existing lines and making selective improvements to address bottlenecks, rather than building entirely new duplicated lines, would be the most capital-efficient approach ⁷.

Make every other part of the business work at the circled box's pace

Do this. For the next two weeks, start nothing new that doesn't help the circled box, and stop sending it more work than it can finish. Write down every request, idea and project you say no to.

It worked when after two weeks you have a list of at least three things you said no to, and less work is waiting in front of the circled box than when you started.

Why. The TOC Institute's advice for this step is plain. For the subordinate step, it says it is crucial to avoid producing more than the constraint can handle ⁴. The researchers behind the Marine Corps case point to a cost of juggling. The authors note that when people are moved between jobs before they finish them, each job tends to take more time than if the jobs were completed one at a time from start to finish ². Our rule of thumb follows from both: nothing new starts unless it helps the circled box and has a named person who owns it, and your list of no's is the measure of your focus.

Example. Toyota describes the Toyota Production System as resting on two pillars: jidoka, "automation with a human touch", and Just-in-Time, "making only what is needed, when it is needed, and in the amount needed" ⁸. Toyota credits Taiichi Ohno with helping establish the Toyota Production System ⁸. We read Just-in-Time as the same rule seen from a factory floor: work made before the next step can use it just waits in a pile.

Add capacity only after you've squeezed the box and paced the rest, then draw the boxes again

Do this. If the circled box is still the limit after you have squeezed it and paced everything else to it, add capacity there: a hire, a tool or a second team. Then redraw the four boxes with fresh numbers and circle again.

It worked when you redrew the boxes after the change and either circled a new box or confirmed the old one with the new numbers.

Why. The TOC Institute is firm about the order. It advises elevating the constraint only once exploiting and subordinating are fully complete, if at all ⁴. It also warns that once a constraint is elevated, the weak link may not remain the weakest ⁴. In our view, this step is what makes the method a cycle rather than a one-time fix: once the circle moves, the next round starts.

Example. In its second-quarter 2018 update, Tesla wrote that over the past 12 months it had overcome bottlenecks across various stages of the Model 3 manufacturing process ⁷. Tesla wrote that it had become clear its main general assembly line, GA3, would likely become a production constraint if certain issues were not addressed ⁷. To address the short-term issues with GA3, Tesla built a second assembly line, GA4, to help it reach its 5,000-a-week target earlier ⁷. Tesla reported that during July 2018 it had repeated weekly production of approximately 5,000 Model 3 cars multiple times ⁷. We read Tesla's letter as the squeeze and add-capacity steps at a large scale: get more from the lines you have, add a line where the constraint sits, and keep looking for the next bottleneck.

We think of your business as a chain: it delivers no more than its narrowest step allows

In the 1980s, Goldratt and Cox presented the Theory of Constraints as a method for managing production planning and scheduling ². Goldratt and Cox told the idea as a story. In the novel, plant manager Alex Rogo has ninety days to save his plant, or it will be closed by corporate headquarters with hundreds of job losses ¹. The book's publisher quotes The Economist: a survey of managers' reading habits found that the one management book they had actually read from cover to cover was The Goal ⁹. When Goldratt died in June 2011, The Goal had sold more than 5 million copies and had been translated into 35 languages ¹⁰.

The Theory of Constraints rests on the principle that the goal of a business is to make money, now and in the future, and that a system's constraints determine its capacity to make money ². In our view, that is the light-bulb moment for an owner with too many priorities: the priorities are not equal, because only the one at the constraint changes how much the whole business can deliver. In our view, the easy move is to add people where the team feels busiest. We think busy and narrow are different things: a stage can be busy and still have room, while the step that limits output sits somewhere else.

Little's Law, L = λW, says that the average number of items in a system equals the average arrival rate of items multiplied by the average time each item spends in the system ¹¹. John D. C. Little published his proof of the formula in Operations Research in 1961 ¹¹. In operations terms, Little writes the law as throughput equals work-in-process divided by cycle time ¹¹. Little notes that the law holds under remarkably general conditions ¹¹. We read that as a way to find the constraint without a consultant: if work keeps arriving at the same rate, a growing pile in front of one step means a longer average wait for the jobs in it. One person whose team had just grown from five to ten put it this way: "The worst part is I am becoming the bottleneck. Every question routes to me because I am the only one who knows how everything is supposed to work." The same person described what it cost: "Client onboarding that used to take 3 days is now taking over a week because nobody is sure whose job each step is." In our view, that is Little's Law at work: more work waiting on one person, and a longer wait for each client.

In our view, this builds on picking a few priorities each quarter and saying no to the rest: you choose them by the constraint, so the priority that frees the circled box goes first, and the others wait their turn.

At a Marine Corps maintenance center, nearly every department looked short of capacity, and the constraint was a scheduling rule

In 2001 the Marine Corps Maintenance Center in Albany, Georgia, was overhauling only five MK-48 heavy haulers a month against a demand of 10 ². At the Center, asking for additional time to complete the work had become a normal way of doing business ². The Center had been plagued by apparent capacity shortages in virtually every department, and additional manpower was being considered ². As we saw in step one, each major activity was seen as a bottleneck by at least someone.

The analysis showed that, contrary to everyone's opinion, the Center had more than enough capacity to repair 10 MK-48s a month ². The bottleneck was not a physical resource but a policy constraint introduced by the scheduling process, which pushed work onto the shop floor without regard for the status of the resources ².

After the change, repair cycle time for the MK-48 fell from an average of 167 days to 58 days, and for the LAV-25 from 212 days to 119 days ². The cost to repair products fell by 25 to 30 percent, mainly because fewer delays meant more throughput with no increase in the cost of repair ². The MK-48 line could then work at anywhere from 10 to 23 units a month ². The authors wrote that the Center was ahead of, or on schedule for, 99 percent of the production lines where the principles had been implemented ².

We read the Center's story as the lesson that matters for a business of any size: the bottlenecks people named were not the constraint, and hiring into them would have added cost without adding output. In our view, the rule in your business might be how jobs are scheduled, who has to approve a quote, or how much work you start before the last job is done.

A 2003 review by Mabin and Balderstone pooled the results of more than 80 successful Theory of Constraints applications reported in the literature and found significant improvements in both operational and financial performance ¹². The review's authors wrote that despite extensive searches they found no reports of failures, and that the reports came mainly from manufacturing organisations ¹². In our view, finding no failures says more about what gets published than about the method, so treat it as a strong tool, not a guarantee, and test it on your own numbers.

The method can stall when the constraint is you, or when demand and delivery look tied together

Sometimes the circled box is you. In our view, when every box waits on your answer, the constraint is the owner, and the fix belongs in two other articles: how to replace yourself in your business, and what to fix first when you are overwhelmed.

Sometimes two boxes seem to hold each other back. One person running a family food factory said: "I need both — more clients AND the ability to actually produce for them." The same person described the loop: "It’s a frustrating loop: I need sales to generate cash, but I need cash for packaging before I can produce, and I need to produce before I can sell." In another reply, the same person conceded: "But you're probably right, no amount of systems or spreadsheets will fix this if nobody's out there selling." In our view, a loop like this still has one narrow point. Run the circling step and let last month's numbers decide which box you circle.

One software consultant warned a small business asking about a new system: "If you don't have operational limiting factors your bottleneck is not going to be affected by an ERP, maybe it will even slow you down." Our rule of thumb: don't buy software or hire for a box you haven't circled. In our view, money spent on a stage with spare capacity buys more spare capacity.

Sometimes you circle the box and still find yourself working on everything else. Ask yourself plainly what is stopping you, and deal with that first. Our rule of thumb is to give the larger share of your own working time to the circled box until it is no longer the circle. In our view, the constraint also changes as the business grows: in one year it may be getting enough help with the workload, in another it may be trying to serve everyone.

Within fourteen days, four boxes and a list of no's can show you where to work

Here is the whole method as one small task, sized to fit in your next fourteen days.

You started with too many priorities and now have one to work on first

You began with more priorities than hours, each one looking urgent. You now have a self-check that points to your constraint: four boxes with last month's numbers, and one circled where work waits longest or the number drops most. You have the five steps in the order to take them: number the stages, circle one box, squeeze it, pace the rest to it, then add capacity and look again. And you have a list of what to say no to while you work on it: anything that doesn't help the circled box, written down for the next two weeks.

Read before: picking the three things that matter this quarter, saying no to good opportunities, How to Replace Yourself in Your Business (Before You Can Afford a Team), Overwhelmed as a Business Owner? What to Fix First, and What Can Wait and How to Stop Micromanaging Your Team: Delegate Decisions, Not Tasks.

To see whether operations, technology or media holds your business back most, take the Scorecard.

Sources

1. The Goal: A Process of Ongoing Improvement, 30th Anniversary Edition · North River Press (Eliyahu M. Goldratt and Jeff Cox), publisher page
2. Applying Theory of Constraints Principles and Lean Thinking at the Marine Corps Maintenance Center · Defense Acquisition Review Journal, Defense Acquisition University (Mandyam Srinivasan, Darren Jones, Alex Miller), 2004 (archived copy)
3. Theory of Constraints · TOC Institute, undated page
4. Five Focusing Steps · TOC Institute (Theory of Constraints training firm), undated page
5. Small Business Credit Survey: 2026 Report on Employer Firms (2025 survey) · Federal Reserve Banks, 2026
6. Rethinking Activity-Based Costing (Harvard Business Review excerpt) · HBS Working Knowledge (Robert S. Kaplan and Steven R. Anderson), 2005-01-24
7. Tesla Second Quarter 2018 Update (shareholder letter) · Tesla, Inc., published with Q2 2018 results, 2018-08-01
8. Toyota Production System · Toyota Motor Corporation (global.toyota), undated page
9. The Goal: A Process of Ongoing Improvement, 3rd edition · Routledge (Eliyahu M. Goldratt and Jeff Cox), copyright 2004, publisher page
10. Eli Goldratt, Author of The Goal and Creator of TOC, Dead at Age 64 · iSixSigma (Paul Arnold), 2011-06-15
11. Little's Law as Viewed on Its 50th Anniversary (reprint of Operations Research 59(3), 2011) · Project Production Institute (John D. C. Little), reprinted 2016-12-18
12. The performance of the theory of constraints methodology: Analysis and discussion of successful TOC applications · International Journal of Operations & Production Management 23(6) (V. J. Mabin and S. J. Balderstone), 2003, abstract