Leveraging Your Data As a Foundation for Manufacturing ROI

Data and analysis don’t have to be complicated to yield bottom-line benefits

Janelle Farkas

November 12, 2019

According to the International Institute for Analytics, businesses that use data will gain $430 billion in productivity benefits over competitors who aren’t using data by 2020. As an industrial engineer for the Northeastern Pennsylvania Industrial Resource Center, part of the MEP National Network, I tell small-business owners and manufacturers that this quote does not say you have to use “big” data. You don’t have to use complex analysis methods and the latest and greatest technology. It just says in order to get a piece of that productivity pie, you have to do something.

Unfortunately, many small- to medium-sized manufacturers (SMMs) are still not taking advantage of data to boost their bottom-line margins. This is often due to a common misconception that utilizing data requires a Ph.D. in statistics or a state-of-the-art ERP system to crunch the information for you. Utilizing data for manufacturing does require a willingness to experiment and a time investment to realize bottom-line benefits, but it doesn’t have to be complicated.

Data demystified for competitive advantage

Some smaller and family-owned companies fear that data could make certain manufacturing skills obsolete or even take away someone’s job. In reality, data can generate more jobs by providing the necessary information to increase production efficiency and capacity.

Data and analysis support SMMs’ success through:
• Clear and focused communication that improves understanding
• Objective information that supports both short- and long-term decision making
• More efficient operations with reduced lead times and improved quality
• Improved customer satisfaction

As I noted earlier, data and analysis don’t have to be complicated to yield bottom-line benefits. Some examples that may be relevant to your organization are:
• Daily production quantity
• Defects in production
• Manual process time
• Time per production run
• Planned time vs. actual time for production

To further illustrate this concept, let’s look at a theoretical case. Smaller manufacturers often don’t do enormous runs of one product. A large portion of their production involves smaller runs that may repeat a few times a year, but it’s not a consistent flow. Measuring and comparing these different runs can provide many insights.

Say you’re a small manufacturer that has quoted a project at 50 hours to complete 50 widgets. After measuring two production runs, the data reveal:

Production Run

Total Time for 50 Acceptable Widgets

Average Time/Widget

Variation in Time/Widget


55 hours

1.1 hours

0.5 hours


40 hours

0.8 hours

0.25 hours

Based on these data, you can see that there has been an improvement across the board from production run 1 to run 2—why? Were process improvements made that should be documented for future use? Could some of these improvements be replicated across other products or production lines? If the run measurements were reversed, showing a decrease in performance from production run 1 to run 2—why? Was a critical machine down, requiring employees to use an older machine? Was there something different about the raw materials?

The trick to getting real ROI from your data is understanding what types of information are most beneficial for your specific manufacturing operations. You have to ask the right questions based on your current processes. These data can serve as a lean baseline and the foundation for a continuous improvement (CI) strategy.

Using data as a lean baseline

Data are an excellent starting point for assessing success—whether it be productivity, cost, or quality. Using data at the beginning of your lean journey may seem a bit intimidating at first, but it doesn’t take many data sources to get started. For example, hospital emergency rooms typically use just four pieces of data to assess patient priority. When flying a plane, a pilot requires six or fewer data sources to keep a plane in the air and determine where she needs to go. For these seemingly complex situations, you would think data would also be complex, but a limited amount of information is used to make critical decisions.

For SMMs, a few pieces of data can serve as important components of a lean journey. However, it’s no secret that beginning that journey can be tough. Have you ever been tasked with improving a production process without a strategy? Perhaps you had to dive in without a life jacket because someone demanded it. The effort probably wasn’t a tremendous success because no one had agreed to what “success” looked like. This sort of experience is enough to make anyone hesitant to try again.

If you’ve collected data from your processes before you begin a new lean initiative, you can determine quantitatively (and not just anecdotally) where your biggest opportunities are, what success looks like from a quantitative perspective, and the ROI of the initiative. This information builds the momentum of your lean journey, rather than detracting from it.

Maybe you’re trying to reduce the number of product defects that occur in a week’s time. Once you’ve collected the data on the number of defects in one week, you can use that information as your baseline. When the data show the defects have gone down from 15 to five a week, you’re beginning to see the ROI for your continuous improvement efforts.

That’s just the tip of the ROI iceberg when it comes to data. Another opportunity for SMMs to reap bottom-line results is through using Industry 4.0 measurement and monitoring capabilities.

Cost-effective data monitoring and measuring

Of the SMMs that collect data, many are still collecting measurements and conducting quality inspections manually. This takes employees away from other potentially productive activities to take those measurements, reducing the bottom-line benefits. In addition, we’re all human, and people are prone to making mistakes. Analysis and insights are only as good as the data used to generate them, so if there are mistakes in the measurements, your insights may lead you to a poor decision.

Thanks to the automated measuring and monitoring capabilities available in today’s market, you can get accurate and repeatable data. This information is unbiased and often real-time. Automation removes the temptation to manually “nudge” the numbers just a bit in your favor. When the entire measuring and monitoring process is automated, you can see the raw truth of what’s going on in your processes.

SMMs willing to invest in measurement and monitoring capabilities are going to see more bottom-line results in their continuous improvement processes.

Using data to advance continuous improvement

Once you’ve got actionable data, you can continue your quest for continuous improvement, which may include:
• Allowing someone to assess, analyze, and draw conclusions from your data
• Using the information to make decisions
• Discovering, interpreting, and communicating meaningful patterns in the data

With actionable data in hand, you may be ready for Six Sigma. Keep in mind that Six Sigma is entirely data driven. If your facility has already implemented lean policies, you have been using a visual approach that works on improving process flow and eliminating waste. At some point in your journey, a visual approach isn’t going to cut it. That’s when you know your company is ready for Six Sigma, which is a quantitative approach to improve processes and process capability.

Many people are daunted by the thought of using Six Sigma and data in general. If you monitor your income vs. your spending each month, you’re already using data to monitor the success of your financial decision-making process. If you monitor your daily steps, water intake, calories, or weight, you’re using data to monitor the success of your health decision-making process. Many small manufacturing clients see benefits fairly quickly after applying Six Sigma to opportunities or problems that can’t be addressed visually. It can be very approachable, if the foundation and framework are built correctly.

Ready to become data literate?

Data literacy simply means you have the ability to read, interpret, and communicate your data in the appropriate context.

The first step is to ask yourself some questions:
• What questions do your company constantly need answered?
• Are you sure they’re being answered the same way every time?
• Is there an opportunity for standardization?
• Is there an opportunity for automatically generated reports?
• Does your manufacturing operation have a data strategy to chart the course to your desired future state?

Once you’ve become data literate, you realize that data are triggers for root cause analysis. This is not finger pointing. It does make management stop and say, “What caused this defect? Did we get a bad batch of material? Does our machine need preventive maintenance? Did something happen weather-wise that could have damaged the materials?”

Your data are triggers to have those conversations and really start to understand what’s going on with your processes.

Want to learn more about how your operation can benefit from data and other Industry 4.0 tools? Connect with the experts at your local MEP Center.

About The Author

Janelle Farkas’s picture

Janelle Farkas

At the Northeastern Pennsylvania Industrial Resource Center, part of the MEP National Network, Janelle Farkas works closely with clients to improve the three Ps that apply to any organization: Processes, People, and Products. Her specialties include Six Sigma, statistical analysis, continuous improvement strategy, data strategy, design and implementation of operational metrics, and Lean Enterprise.


Big Profits from Small Data

A recent TED video showed that Big Data is a $122 Billion Dollar business, but 73% of projects failed to return a profit. That’s a one-sigma performance. You don’t need big data. You can get big profits from small data.

Defects, mistakes and errors devour one-third of a  typical company’s expenses. The "Pareto pattern" driving these errors can be discovered in a few hundred records. Then, a little root cause analysis will determine how to fix the problem forever.

Using data analysis and charts like control charts and Pareto charts will solve most problems facing service and manufacturing businesses. Data science is on the rise.