Control Charts: Everything You Need To Know

Control charts are a key part of the management reporting process that have long been used in manufacturing, stock trading algorithms, and process improvement methodologies like Six Sigma and Total Quality Management (TQM). The purpose of a control chart is to set upper and lower bounds of acceptable performance given normal variation. In other words, they provide a great way to monitor any sort of process you have in place so you can learn how to improve your poor performance and continue with your successes.

The control chart serves to “sound the alarm” when a process shifts (for instance, a machine suddenly breaking on a factory floor) or if someone has a breakthrough that needs to be documented and standardized across the larger organization. Simply put (without taking anomalies into consideration), you'll know something needs to be fixed if you're below your lower control limit or above your upper control limit. See the control chart example below:

Control Charts At Work In 2 Industries

In industrial settings, control charts are designed for speed: The faster the control charts respond following a process shift, the faster the engineers can identify the broken machine and return the system back to producing high-quality products. At a factory, a lag in testing could mean that thousands of parts are produced incorrectly before anyone notices the machine is broken, which results in wasted time and materials, as well as angry customers.

In nonprofit organizations, a control chart could be used to determine when an online donation system has broken down. If the website goes offline, halting critical donations, the leadership team can quickly alert IT and ensure the page gets back up and running quickly. Alternatively, seeing a major jump in donations likely means something good is happening—be it world events or a successful marketing campaign. Either way, leadership should know as soon as possible when donation activity changes.

Control Chart Examples: How To Make Them Work In Your Organization

• Budget: You can use your control charts to examine your percentage of spend each month. If you spend over 15% of your budget in one particular spring month, that is extremely helpful to know right away so you can cut back over the rest of the year. Or, if you spend less than 8% of your budget for a couple months in a row, you'll know you may have a little wiggle room in the months to come.
• Retention rate: Some organizations feel like they need a little turnover to keep the organization healthy. If you're retaining your talent at a rate above your normal control limit, you'll know that you may not be evaluating staff very selectively. You'll want to be sure to identify the reasons you may be retaining so many employees to see if this is positive news or if an HR process is broken. But if your retention rate is increasing or it drops below your lower control limit, you'll be able to determine how to move that trend the other direction and dedicate more resources to recruiting for a period of time.
• Employee or citizen surveys: At ClearPoint, we do quarterly customer support feedback surveys to see how our clients feel we’re doing. If we're doing something that is having a positive effect, we want to know what it is and continue to do it well. But if we're falling below our normal control limit, we'll want to note that something needs to change. This could be anything from having better customer service response time to changing a particular feature in our software that is frustrating or difficult to use.

"Can I Create A Control Chart In Excel?”

Because of Excel’s computing power, you can create an Excel control chart—but in order to do so, you need to know how the upper and lower limits are calculated. There are different statistical analysis tools you can use, which you can read more about here.

Control Charts & The Balanced Scorecard: 5 Rules

Control charts can be used as part of the Balanced Scorecard approach to account for an acceptable range or variation of performance. If you choose to do this, there are five key quality control rules to keep in mind when considering using control charts at your organization:

1. Give it time and update your limits accordingly. You can't expect to see immediate results or instant insights from a new control chart (that is measuring something new to your organization). It takes a number of months—or even years—to understand natural variation and baseline “normal” performance.Don't be afraid to adjust if necessary, and don't rest on your laurels if something you've been tracking has been steadily improving over time. These are good indications that your upper and lower limits may need to be updated.
2. Watch for "big movers."  There is going to be a certain amount of variation as part of normal operations, and small variation is nothing to worry about. Instead, focus your attention on major jumps or falls. These are the places where your organization needs to concentrate its efforts.
3. Results matter and should be visible. Process improvement initiatives should cause a particular metric to rise above the upper control limit, demonstrating that there was a statistically significant shift in the objective’s measure. Control charts give you a clear way to see results and act on them in the appropriate way. Over time, you may need to adjust your control limits due to improved processes.
4. Don’t get bogged down. Take a moment to remember that control charts can be complicated. (They were, after all, developed by engineers!) But your organization can keep your control charts as simple as you need. Extremely complex math is still being developed in the operations research field to better understand process variation and how to account for it via control charts, but the typical leader at an organization does not need to worry about going into that level of detail. Instead, try to identify the acceptable upper and lower limits for each key metric that you want to track, and keep the overall theory of limits in mind when reviewing your control charts.
5. Update your limits according to new data. Remember that controls charts are based on historical data—so as time progresses and new data is collected, these limits need to change. Don't be afraid to adjust if necessary, and don't rest on your laurels if something you've been tracking has been steadily improving over time. These are good indications that your upper and lower limits may need to be updated.
6. Set your RAG evaluations as ranges. On your control bars, within 5% of your target is green. Outside of 5% but within 10% is yellow, and outside of 10% is red. You can adjust the percentages, but the RAG status help show that you are getting more out of control.

The key with control charts is to recognize when anything is happening outside the norm. Be it good or bad, you will want to develop an action plan for how to respond when the latest measure lands outside the acceptable limits.

FAQ:

What are the different types of control charts?

The different types of control charts include:

- X-Bar and R Chart: Used for monitoring the mean and range of a process over time.
- X-Bar and S Chart: Similar to the X-Bar and R Chart but uses standard deviation instead of range.
- P Chart (Proportion Chart): Used for monitoring the proportion of defective items in a process.
- NP Chart: Similar to the P Chart but used for a fixed sample size.
- C Chart: Used for monitoring the number of defects per unit.
- U Chart: Used for monitoring the number of defects per unit when the sample size varies.

How do I interpret a control chart?

To interpret a control chart:

- Center Line: Represents the average value of the process.
- Control Limits: Upper and lower control limits (UCL and LCL) indicate the threshold for variation. Data points within these limits suggest the process is in control.
- Data Points: Individual data points plotted over time. Points outside the control limits indicate potential issues or special causes of variation.
- Patterns: Look for patterns such as trends, cycles, or clusters, which may indicate underlying problems.

What are some common mistakes people make when using control charts?

Common mistakes people make when using control charts include:

- Misinterpreting Random Variation: Treating random variation as a sign of a problem can lead to unnecessary adjustments.
- Ignoring Special Causes: Failing to investigate points outside control limits can overlook significant issues.
- Incorrect Control Limits: Using incorrect calculations for control limits can lead to misinterpretation.
- Overcomplicating the Chart: Adding too much data or unnecessary details can make the chart difficult to read.
- Lack of Training: Inadequate training on how to use and interpret control charts can lead to misinformed decisions.

How can I use control charts to improve my processes?

To use control charts to improve your processes:

- Monitor Performance: Regularly use control charts to track process performance and stability.
- Identify Variations: Detect and investigate any points outside control limits or unusual patterns.
- Implement Improvements: Use insights from control charts to make data-driven decisions and implement process improvements.
- Continuous Monitoring: Continuously monitor the process after implementing changes to ensure improvements are sustained.
- Train Employees: Ensure employees are trained in using and interpreting control charts to foster a culture of quality and continuous improvement.