What is Process Improvement?

As Process Improvement Consultants, the first hurdle we must overcome is giving a definition to the services and process we use to improve the processes in our client’s operations. While the phrase is fairly descriptive, the science and practice of Process Improvement, and how it can help your business, is easy to lose sight of. So then, what is Process Improvement? Put simply, Process Improvement is the science of using feedback loops to produce continuously better results. These feedback loops can take different forms in different frameworks, but usually include data gathering, hypothesis formation, data analysis, improvement, and control stages.

Let’s Start with a Beer


While not many would think beer would be the way we’d begin our primer on Process Improvement, there’s a delicious anecdote involving beer that will help us explain how this all works. Near the turn of the century, the Guinness Brewery, rather ahead of its time, made a practice of hiring the brightest graduates of Cambridge and Oxford. Such was the case with William Sealy Gosset, a bright statistician, who Guinness Brewery hired in 1899.

At Guinness, Gosset sought a way to identify the best yielding varieties of barley using only small sample sizes. Through his work, Gosset developed the t-test (under the pseudonym “Student”) and in many ways revolutionized the field of Statistics and by extension, the nascent science of Process Improvement. His work allowed Guiness to focus their barley operations on the most productive and robust varieties, laying the groundwork for a more efficient process from seed to bottle.

How does it work?

We’ve already defined Process Improvement as “the science of using feedback loops to produce continuously better results.” Process Improvement is a tool that makes the input of a process (like the type of barley you’re growing in the field) a function of the output of that process (the beer you just brewed). It’s a way of measuring and improving every part of a process for a superior, more efficiently generated product.

Another simple example to help understand the science of Process Improvement is to look at the pen probably sitting on the desk in front of you. Your pen is, in all likelihood, the same size and shape as all other pens that come off that assembly line. But the factory doesn’t know that by taking a ruler to every single pen their process makes. Instead, they measure and calibrate the pressure of the extruder that makes the pen the size and shape it is, and correlate that to a small sample of final pen measurements. By monitoring that input correctly, they ensure an expected and consistent output.


Figuring out where to start improvement can be daunting. We're here to help.

Tools for Process Improvement

In order to successfully improve processes though, you need more than abstract concepts of what it is. You need tools. Fortunately, scientists and engineers have developed a whole bunch of them to tackle process problems of all shapes and sizes. Here are a few of our favorites.

Feedback Loops

The first and arguably most important tool in the Process Improvement Professional’s toolkit is in the very definition we gave above: Feedback Loops. The concept is simple enough: (1) do something, (2) measure it, and (3) make a hypothesis on how you might do it better and go back to the first step to measure the change, if there is one.

Data collection seems like the easiest part, second only to the business activity itself, but many businesses don’t have a systematic way to collect data that eliminates bias and many don't collect data at all. Even so, you might be surprised how much data you’re already sitting on. Do you track your sales efforts? Have you collected employee surveys? Do you collect customer feedback? In the right hands, all this data can be used to improve your processes, saving--and making--you money.

Root Cause Analysis

It’s one thing to recognize that something is going wrong in a process, but something else entirely to identify what’s causing it. That seems straightforward-and it is. But instead of expedient, albeit informed guesses, Process Improvement brings to the table a whole set of tools to narrow down the root cause using science.

A simple example of a Pareto Chart. Columns are based on number of instances while the line plotted represents the cumulative percentage of all faults.

A simple example of a Pareto Chart. Columns are based on number of instances while the line plotted represents the cumulative percentage of all faults.

Pareto Chart

There’s a common rule of thumb in Process Improvement: 80% of problems are produced by 20% of causes. The Pareto Chart is a tool used to identify those 20%. Causes are plotted on the x axis, and the percentage of times that cause leads to the problem in question is plotted on the y. At this point you have a typical bar graph, but this is where it gets fun.

First, arrange causes from the most common to the least common. Then, create a point above each where you add the the sum of the current cause to the sum of all previous causes. In other words, if cause 1 creates 50% of the problems, and cause 2 is the source of 20% of the problems, the point above cause two would be 70%. Connect the dots to create your Pareto Chart, and you’ve visualized what’s causing the biggest problems in your process, and more importantly, how much of a reduction in problems you’ll have if you address those root causes.

An Ishikawa or Fish-bone Diagram helps structure ideas on what is causing a defect. Image courtesy of  Wikipedia

An Ishikawa or Fish-bone Diagram helps structure ideas on what is causing a defect. Image courtesy of Wikipedia

Fish-bone or Ishikawa Diagram

When you need to identify the possible causes of a problem, there’s really no way around a good brainstorm. But the output of many brainstorms is often a list at best, alphabet soup at worst. Enter the Ishikawa “Fish-bone” Diagram.

An Ishikawa Diagram begins by writing the problem. Draw a horizontal line away from the problem. Then adding at least four diagonal lines pointing away from the problem, like ribs on a fish. Each one of these ribs should be linked to a cause. From there, you and your team can brainstorm possible reasons that cause manifested.


Processes are Everywhere

While many mistakenly think that Process Improvement can only be used in manufacturing, the science has successfully optimized healthcare applications, government departments, and startup business models. We’ve personally used it to improve the sales process of an energy efficiency provider by calibrating the discounts given to each client based on a series of criteria. Now, more deals sell for more money.

The boundaries are limitless when you realize that virtually everything we do is part of a process ripe for improvement. From your morning slice of toast, to setting your alarm before you go to sleep, we engage in processes and informally improve them constantly. But only by engaging in Process Improvement with a scientific rigor can you achieve breakthrough results.

If you’re ready to take your business’s operations up a notch, let us know. We’d love to help you figure out how Process Improvement can help you earn more, with less work.

Need an expert?

Sustainabilist uses Data Science, Machine Learning and Artificial Intelligence to make businesses more capital efficient and less carbon intensive using the science of Process Improvement.