September 8, 2026 · 7 min read
TL;DR: Control chart options are keyed 3 times and wrong 11 times in this bank. The recurring test is not how to read a chart, it is where the two kinds of limit come from. Specification limits come from the customer and say what is acceptable. Control limits are calculated from the process and say what it does when nothing unusual is happening. Every distractor that confuses the two is describing a real mistake practitioners make.
Candidates who learn the control chart as a picture with lines on it get the reading questions right and the reasoning questions wrong. The bank tests the reasoning.
Different places, and that is the whole answer.
A record in the bank puts a senior project manager in the position of coaching a junior one who asks where the control limits sit. Keyed is within the specification limits, and the explanation states the distinction as a rule: specification limits are set by the customer or the requirement, so they state what is acceptable, while control limits are calculated from the process itself, conventionally at three sigma either side of the process mean, so they state what the process actually does. It then adds the line worth memorising: nobody chooses the control limits, you choose a process whose limits fall where you need them.
A second record asks what defines the acceptable range of variation for a process. Keyed is upper and lower control limits, and three near misses surround it. The mean is the centre line the limits are measured from, not a range. Standard deviation is the unit those limits are expressed in rather than the boundary itself. Specification limit comes from the customer rather than from the process. Each distractor is a true statement about control charts, placed where it does not answer the question.
| Control limits | Specification limits | |
|---|---|---|
| Set by | The process, calculated from its own variation | The customer or the requirement |
| Say what | The process does normally | Is acceptable to receive |
| Typical construction | Three sigma either side of the mean | Whatever the requirement states |
| Can you move them? | Only by changing the process | Only by renegotiating the requirement |
| Where they should sit | Inside the specification limits | Outside the control limits |
More than a point outside the limits.
The bank tests three signals, and they escalate in subtlety. The first is the outlier: a single measurement falling outside the band between the upper and lower control limits, which signals the process may be out of control and that an assignable cause should be investigated. A distractor offers that outliers are insignificant results, often measurement errors, whose causes need not be investigated, and the explanation rejects it directly: outliers are not to be dismissed without investigation.
The second is the run. Studying a bottling line chart, a project manager notices nine consecutive points falling on the same side of the mean. Keyed is identifying the assignable cause. The explanation gives the statistical reasoning rather than the rule alone: with a stable process, each point is about equally likely to fall above or below the mean, so a long run on one side is very unlikely by chance. Two distractors are instructive. Redrawing the chart around the new mean treats the symptom as the new normal. Ignoring it under the rule of seven inverts what the rule of seven means, which is a mistake that only survives if you learned the phrase without the content.
The third is the one candidates miss. A control chart for a machined component's diameter shows the process mean has drifted noticeably closer to the upper specification limit over the last several measurements, though every individual point still falls within the control limits. Keyed is treating the drift as a signal and having the team investigate before it crosses a limit. Taking no action because every point remains within the control limits is offered, and it is exactly what a chart read literally would tell you to do. The explanation says why it fails: a mean steadily drifting toward a specification limit is itself a signal, because it shows the process trending toward becoming out of control rather than staying stable, and waiting for a breach means acting only after the defect risk has already materialised.
Common trap: widening a limit to make a problem disappear. Two distractors in that drift record do this, and they are worth naming because both look like action. Recentring the control chart around the new, higher mean redefines the drift as normal, which loses the signal that made it visible. Widening the specification limits so the drifting measurements no longer appear as a risk changes what the customer is told to accept, without changing anything about the part. The bank offers both as plausible next steps and explains why neither is one, which is the difference between learning that they are wrong and understanding what makes them tempting.
When the question is why, not whether.
A design project has a significant defect. The management team blames poor software, the engineering team blames incorrect input data, and the question asks which tool would best help identify the cause. A control chart is offered. Keyed is a fishbone diagram, and the explanation puts the boundary in one clause: control charts reveal whether a process is out of control but not why. Checklists confirm that a process is followed correctly, and check sheets tally how often defects occur.
That sentence is the shape of most control chart distractors in this bank. The chart detects. It does not diagnose. When a stem describes a disagreement about causes, a chart adds nothing, and the keyed answer is whatever structures the search for the cause.
One more boundary is worth knowing. A record asks which two tools a team would not be using while controlling and verifying the quality of deliverables. Control charts and statistical sampling examine the deliverables themselves, so they belong to that work. Quality audits and Design for X do not, because both review the process and the design approach rather than the output. The explanation notes that PMBOK 8 keeps that assurance-versus-control distinction inside Manage Quality Assurance rather than running two separate processes.
What is the difference between control limits and specification limits? Specification limits are set by the customer or the requirement and state what is acceptable. Control limits are calculated from the process itself, conventionally at three sigma either side of the process mean, and state what the process actually does when nothing unusual is happening. Nobody chooses control limits; you choose a process whose limits fall where you need them.
Where should control limits sit relative to specification limits? Inside them. A process whose natural variation falls entirely within what the customer will accept is capable of meeting the requirement. If the control limits sit outside the specification limits, the process produces unacceptable output as part of its normal behaviour.
What is the rule of seven on a control chart? A pattern test. Seven or more consecutive points falling on the same side of the mean is treated as a signal that something has changed, even though no individual point has breached a control limit. With a stable process each point is about equally likely to fall either side, so a long run on one side is very unlikely by chance.
Does a point inside the control limits always mean the process is fine? No. The bank keys investigation when a process mean drifts steadily toward a specification limit even though every individual point stays inside the control limits, because a trend is itself a signal. Waiting for an actual breach means acting only after the defect risk has materialised.
PMP Practice's 2,141 questions are certified against PMBOK 8 and the July 2026 ECO, with every wrong answer explained rather than just marked wrong. Start the free 20-question sample — no card, no signup required to try it.
Related reading: Quality on the PMP Exam: Prevention, Inspection, and the Cost of Getting It Wrong and Root Cause Analysis: 13 Keyed, 14 Distractors, and One Word in the Stem Decides It. Quality tasks are mapped on the Process domain study page.
What is the difference between control limits and specification limits?
Specification limits are set by the customer or the requirement and state what is acceptable. Control limits are calculated from the process itself, conventionally at three sigma either side of the process mean, and state what the process actually does when nothing unusual is happening. Nobody chooses control limits; you choose a process whose limits fall where you need them.
Where should control limits sit relative to specification limits?
Inside them. A process whose natural variation falls entirely within what the customer will accept is capable of meeting the requirement. If the control limits sit outside the specification limits, the process produces unacceptable output as part of its normal behaviour.
What is the rule of seven on a control chart?
A pattern test. Seven or more consecutive points falling on the same side of the mean is treated as a signal that something has changed, even though no individual point has breached a control limit. With a stable process each point is about equally likely to fall either side, so a long run on one side is very unlikely by chance.
Does a point inside the control limits always mean the process is fine?
No. The bank keys investigation when a process mean drifts steadily toward a specification limit even though every individual point stays inside the control limits, because a trend is itself a signal. Waiting for an actual breach means acting only after the defect risk has materialised.