Business Process Optimization Using the Theory of Constraints

Physicist turned management theorist Eli Goldratt’s 1984 book The Goal is one of the most influential business books of all time. In it, Goldratt lays out a framework known as the Theory of Constraints which has influenced business operations ever since. The peculiar thing about The Goal is that it’s not a typical business book; it’s actually a novel that illustrates business concepts through the plot.

Goldratt explains the Theory of Constraints through the story of Alex Rogo, a Midwestern manufacturing plant manager under immense pressure from senior leadership. He is given a very short time frame to turn things around before corporate shuts down the plant. Meanwhile, his relationship with his wife is strained to the breaking point, and he has to juggle responsibilities between family and work before he loses both. It’s a captivating story, and relatable to many executives and business owners.

With no idea what he’s going to do, Rogo happens across an old friend: a physicist named Jonah, who serves as a sort of avatar of Goldratt within the story. By applying Jonah’s methodologies, Rogo is able to turn things around in his business and save the plant. (If you want to know what happens in his personal life, you’ll need to read the novel)

While The Goal’s setting for explaining the Theory of Constraints is manufacturing, the principles can be applied to any business situation which involves process or the lack thereof. Goldratt also wrote later business novels that expanded the Theory of Constraints to other business functions like marketing, strategic planning, and project management.

In this article, we’re going to briefly introduce the Theory of Constraints as articulated in The Goal and explore some ways you can apply it in your business. Again, when we talk about things like systems or capacity, remember that the original context is manufacturing, but you can apply these same principles to whatever your business processes happen to be.

---

Every Process Has a Bottleneck

The major idea behind the Theory of Constraints is that all processes have a constraint known as a bottleneck: a point that holds up the completion of the overall process. In other words, if a business process flows like this:

Step 1 (ten minutes) → Step 2 (fifteen minutes) → Step 3 (twenty minutes) → Step 4 (five minutes) → Step 5 (five minutes)

…then the bottleneck (the process constraint) is Step 3. It doesn’t matter how quick Steps 4 and 5 are because they can’t proceed until Step 3 is completed. And if you improve efficiency in Steps 1 and 2, you still aren’t helping the overall system because everything is held up at the Step 3 bottleneck. In fact, you may actually be hurting the overall system by putting even more pressure on the bottleneck.

The entire business process therefore needs to be re-engineered around the bottleneck if you want to make any real improvements to the overall system, because nothing gets through the system without passing the bottleneck. Or as Goldratt puts it, the capacity of the system is the capacity of the bottleneck.

The Goal of Any Business System

Process improvement and profitability initiatives are often focused on what is immediately visible. As a result, the initial (often only) idea brought to the table whenever someone discusses improving profits is to start slashing costs. But it’s common for executives to cut costs and see no sustained benefit to profitability. In fact, cutting costs might even hurt profitability. The problem is that they’re viewing the costs in isolation rather than as part of the overall system.

The final step of a business’ overall system is what Goldratt calls ‘throughput,’ which is when you actually make a sale. Everything that goes into the business, including all back-office functions, ultimately should serve the final output of making a profitable sale. If the business system cannot reliably produce profitable sales, then it’s not sustainable and eventually the company will not survive. Hence, profit is ‘the goal’ for a business.

The only ways you can increase profit are to increase throughput (make more sales, assuming you’re not selling at a loss), reduce inventory (anything needed to produce your product/service, including raw materials, plants, property, equipment, etc.), or reduce operating expenses (things needed to turn inventory into sales, including all labor and overhead). Note that Goldratt uses a broad definition of the word ‘inventory,’ such that everything in the business can be considered either inventory or operating expense.

There’s also a mindset shift around inventory because the Theory of Constraints views it as a (necessary) liability, as opposed to looking at things strictly through the accounting lens where inventory appears on the balance sheet as an asset. Inventory that isn’t turning into sales efficiently (throughput) is tying up the business’ cash and is therefore damaging to the goal (profit). Note that this is essentially the same inventory concept behind Just-in-Time manufacturing pioneered in 1970’s Japan with the Toyota Production System.

Yet so often, management doesn’t take these obvious concepts into consideration when planning profitability initiatives. Cuts are made that may improve short-term, immediate profitability in one focused area of the business, but at the cost (often unseen) of some other area in the business system, or even worse, of the overall system itself. As a result, you might cut expenses and see no improvement to overall profits. And you might even make profitability worse!

This brings in Goldratt’s criticism of the cost accounting method and his preference for what he calls throughput accounting. Of course costs are important, but they need to be viewed contextually as part of the system rather than in isolation. It’s like the old adage that ‘it takes money to make money.’ If something is expensive but substantially contributes to the profitable output of the overall system, why cut it when there’s a clear ROI? Likewise, if something seems to be working well but doesn’t actually contribute to the profitable output of the system, why keep it? When cuts are made, they should be targeted at expenses that aren’t adding value to the overall system.

And as with making cuts, improvement initiatives targeted at the efficiency of this or that step might actually hurt the overall system. To return to the initial, basic example laid out above: if you focus your improvement efforts on making Steps 1 and 2 more efficient, you might tie up more cash in the inventory that’s needed for those steps, and still be unable to finish the overall process any more efficiently because of the bottleneck. Hence, throughput (the ability to make a sale) hasn’t increased and you’re reducing cash flow by tying up resources in improving the wrong steps in the process. You’re also likely putting more pressure on the bottleneck by building up the queue behind it, making your actual problem even worse.

Identifying and Understanding Process Bottlenecks

It’s important to correctly identify the bottlenecks, and they may not always be obvious. Goldratt thinks of constraints in three categories: resources, policies, and permanent. Permanent constraints are things you can’t necessarily do anything about (like overall market conditions), so they aren’t the focus of the Theory of Constraints. Resource constraints might be money, time, labor, raw materials, technical capacity, or anything similar. Policy constraints are created by non-resource matters (like poor management decisions) that don’t align with the goal of the overall business system. The focus of The Goal (and this article) is on resource constraints.

A process bottleneck happens wherever demand is greater than capacity. So when looking at a complete business process holistically, you’re seeking to identify where the whole system is being stalled. To use another classic adage, it’s the weakest link in the process chain, and the chain is only as strong as the weakest link. There may be several bottlenecks, but usually you’ll find there’s just one or two.

It’s also important to understand that you can’t eliminate bottlenecks — to do so in one area would just shift the bottleneck somewhere else — so the key is to leverage and work around the bottlenecks.

Balance Your Business System with Market Demand

Optimizing the system doesn’t mean balancing its capacity with market demand. In other words, don’t try to engineer your business so that you’re only capable of producing or outputting exactly what you need to sell at exactly the right time. Performance fluctuations happen for a variety of reasons, and if you try to balance your capacity with demand it will eventually cause backlog that you won’t be able to outrun through averages. Absent some future technological development, it’s impossible to perfectly balance system capacity with demand.

Instead, you need to balance your use of the system with the market’s demand. And because the capacity of the overall system is controlled by the capacity of the bottleneck, the correct action is to balance the workflow through the bottleneck itself to the demand of the market. That way, you aren’t idling at the bottleneck but are instead delivering what you can actually sell when you can actually sell it. But that also means you will have idle time in the system; it will just be at non-bottlenecks.

Many business leaders see any idle time and immediately want to spring into action to rectify it, but this is wrong. Idle time is a necessary part of a truly efficient business system. When you optimize to reduce idle time, you’ll wind up focusing efforts and resources into optimizing steps that don’t improve the overall throughput of the system and may even make it worse. You need to be comfortable with idle time in the non-bottleneck steps of your business processes. That means every non-bottleneck in the process should have excess, unused capacity (i.e., idle time).

Using the Theory of Constraints in Your Business

Using the Theory of Constraints to overcome resource constraints can be distilled into five steps, or what Goldratt calls the Process of Ongoing Improvement:

1. Identify the bottlenecks/constraints.

2. Focus improvement efforts on maximizing the bottleneck’s utilization.

3. Re-engineer the entire system around the bottleneck. This means don’t tie up cash on things that can’t quickly be turned into throughout/sales (even if it means you will have idle time), while also making sure the bottleneck is never lacking work to do. In addition, it means the system has to be balanced with market demand, so non-bottleneck capacity should always exceed demand while the work flowing through the bottleneck should be timed to match demand. This is resolved by building a proper buffer, meaning the bottleneck should have a queue in waiting at all times, but not too much.

4. Increase the capacity of the bottlenecks. This might be through new investment or the like, but only after process changes have tapped out potential improvements from what you already have available. You should always seek to make use of what you have before a commitment to new investment.

5. Manage around potential new constraints. In time, this may cause the bottleneck to shift elsewhere, creating a new issue that needs to be managed and the cycle to repeat. There will always be a bottleneck somewhere, so once you have things under control, manage the system around where you would like it to be. Often the easiest answer is to maintain the current bottleneck so that you don’t have to go through the entire process again.

Practical Takeaways for Continuous Improvement

Though it was released nearly 40 years ago, The Goal is still extremely relevant to business today. While it may not be entirely as unique as some devotees of the Theory of Constraints seem to imply, it does provide a fantastic framework for business process optimization and conveys it in a relatable and meaningful way.

The most important lessons here for any business leader are think about the big picture and realize that this is a process of continuous improvement. Any improvement initiative, profitability efforts, cost cutting, or investments need to be viewed through the lens of the entire system. Even if an initiative appears to show improvement in a certain area, if it doesn’t improve the overall system, then it’s a waste at best and a hindrance at worst. And as the system is improved, new issues will arise that require attention. This is not so much a game to be won as it is a road map to be followed.

This has profound implications for how we think about management. If your KPI’s, compensation structures, and strategic decisions are not made based on things that serve the overall system, then your business is going to misfire. You’ll direct resources and attention to the wrong things and ultimately lead to poor results, scrambling to solve problems, and struggling to make money. Your management systems need to all be aligned to serve the goal.

Another key point is that when re-engineering your processes, don’t waste the bottleneck’s time on things that could be done elsewhere. For example, let’s say your system requires Actions A, B, C, and D to complete, and you have Resources I, II, III, and IV available. All four Resources can accomplish Actions A, B, or D but only Resource III can accomplish Action C. Because Resource III is the bottleneck, wherever possible, don’t waste its time on the other three actions even though it’s capable of handling them. Offload those non-bottleneck actions to the other resources in the system so that the entire process isn’t stalled.

One final point is that your quality control efforts should be greatest in the steps before the bottleneck. If something goes wrong before the bottleneck, it’s not as big of a deal. If something goes wrong at or after a bottleneck, then you may have to expend more time and resources at the bottleneck to fix it, which is substantially worse for the whole system. Solve as many problems as possible before they pass through the bottleneck.

Remember that ‘a chain is only as strong as the weakest link,’ but also that you will always have a weakest link, so everything needs to be managed around it in order to improve the entire system. Get comfortable with accepting imperfections in some (or many) areas of the business if it serves to improve the output of the entire system, i.e., increase profit. Your business may rightly have other ends than making as much money as possible: to provide jobs, serve customers, elevate your community, etc. But no business can survive to do any of those things without making money; that is the goal.

Previously published on NickGausling.com See licensing details.