Quality, Change Control and Monitoring | Free Project Management Course

What you'll learn

Tell quality planning, quality assurance and quality control apart.
Use the cost of quality to justify prevention over rework.
Find the few causes behind most defects with a Pareto chart.
Run a change control process from request to updated baseline.
Report status with clear green, amber and red thresholds.

Once the project is running, three disciplines keep it on track. Quality management makes sure what you deliver meets the agreed requirements. Change control makes sure every change to scope, schedule or budget is evaluated and approved before it happens. Monitoring compares the plan with reality on a regular rhythm and raises a flag early. In this lesson you'll apply all three to the garden center construction job in Dayton, Ohio, where a single "small" owner request is worth $14,500.

3quality activities: plan the standard, assure the process, control the result

76%of punch-list items in this lesson's example come from just 3 of 7 categories

0changes that should be made before they're logged, analyzed and approved

Plan quality
Assure and control
Analyze defects
Control changes
Monitor and report

Quality: plan it, assure it, control it

Quality in project management means meeting the requirements, not adding luxury. A garden center with a plain concrete floor that meets the spec is high quality; a marble floor nobody asked for is gold plating (remember the scope lesson). Three activities work together:

Quality planning

Decide which standards apply and how you'll prove you met them.

Garden center: Ohio building code, the owner's specs, a 24-hour pressure test for plumbing.

Quality assurance (QA)

Check that the process is right, so defects don't happen.

Garden center: a pre-pour checklist before every concrete pour; weekly process audits.

Quality control (QC)

Inspect the deliverables to find defects before the client does.

Garden center: city inspections, the punch-list walkthrough, testing every circuit.

1. An electrician tests every circuit in the finished building against the electrical plans. Is that quality assurance or quality control?

Quality control, because it inspects a finished deliverable
Quality assurance, because it ensures quality
Quality planning

Inspecting or testing the output is quality control. Quality assurance would be checking that the process is sound, for example verifying that the crew follows the installation checklist on every job.

The cost of quality

Quality costs money either way. You can spend it up front on prevention (training, checklists, mock-ups) and appraisal (inspections, testing), or later on failure: internal rework before handover and external failures after it (warranty callbacks, unhappy clients). On a similar job last year, the contractor spent $7,500 on prevention and appraisal and still paid $15,000 for rework and callbacks:

Two-thirds of last year's $22,500 quality cost went to fixing mistakes. Moving money to the left side of this bar is almost always cheaper.

That's the business case for checklists and early inspections: a drywall problem found before painting costs a few hours; found after the owner moves in shelving, it costs days and goodwill.

Find the vital few with a Pareto chart

At the pre-handover walkthrough, the owner and the contractor logged 92 punch-list items. Sorting them by category and adding a cumulative percentage line gives a Pareto chart, which shows that a few causes create most of the problems:

Paint, doors and electrical fixtures account for 70 of 92 items (76.1%). Fixing the process behind those three removes most of the punch list on the next job.

The contractor dug into the top category and found that most paint touch-ups were caused by other trades damaging finished walls. The fix wasn't "paint better" but a process change: finish painting after doors and fixtures go in, and protect corners with guards. That's QA in action.

2. Using the punch-list data, what share of all items do the top two categories (paint and doors) represent?

37.0%
60.9%
76.1%

(34 + 22) ÷ 92 = 56 ÷ 92 = 60.9%. Adding electrical fixtures (14) brings it to 70 ÷ 92 = 76.1%.

Change control: every change through one door

Halfway through framing, the store owner asks for a second roll-up door and upgraded LED lighting. Saying "sure" on the spot is how projects go over budget. Instead, every change follows the same path:

Day 1Submit a change requestWritten, even if it started as a hallway conversation: what, why, who asked.
Day 1Log itIt gets an ID (CR-01) in the change log, so nothing is lost or done "off the books."
Days 2–3Analyze the impactScope, schedule, cost, risk and quality. The door needs a steel header and an engineer's sign-off; the lighting needs a panel upgrade.
Day 4DecideThe change control board (here: the owner and the contractor's PM) approves, rejects or defers.
Day 5Update and communicateNew baselines, updated schedule and budget, and a note to everyone affected. Then, and only then, the work starts.

The impact analysis for CR-01: +$14,500 and +4 workdays on the critical path. The owner approved it, so the baselines change:

Cost baseline:   $290,000 + $14,500 = $304,500
Schedule:        40 + 4 = 44 workdays to completion

ID	Request	Cost impact	Schedule impact	Decision
CR-01	Second roll-up door and LED lighting upgrade	+$14,500	+4 days	Approved
CR-02	Switch to a standing-seam metal roof	+$22,000	+8 days	Rejected (misses the April opening)
CR-03	Add an outdoor irrigation display area	+$6,800	0 days (after opening)	Deferred to phase 2

Contingency is not for new scope

The contingency reserve covers identified risks on the agreed scope. New features (like CR-01) are scope changes: they need their own approval and funding, usually new money from the client or the sponsor's management reserve. Paying for them quietly out of contingency leaves you unprotected when a real risk hits.

3. A client asks the site foreman to "just move that wall a foot to the left." What should happen first?

The foreman moves it, since it's a small change
Reject it: the scope is frozen
Log a change request and analyze its impact before deciding

Small changes can have big effects (plumbing, permits, structural loads). Logging and analyzing first lets the right person decide with real numbers. Changes aren't forbidden; uncontrolled changes are.

Monitor and report with clear thresholds

A status report is only useful if "green" means the same thing to everyone. Set thresholds in advance:

Status	Cost or schedule variance	What happens
🟢 Green	Within 5% of plan	Normal reporting
🟠 Amber	5% to 10% off plan	PM presents a recovery plan at the next check-in
🔴 Red	More than 10% off plan	Escalate to the sponsor now, with options

In week 6, the planned cost to date was $135,000 and actual spending was $142,000:

Cost variance % = (142,000 − 135,000) ÷ 135,000 = 7,000 ÷ 135,000 = 5.2% over → amber

Careful: comparing actual cost with planned cost only tells you that you spent more than planned, not whether you also got more work done. Maybe the crew is ahead of schedule. The earned value lesson in Level 4 fixes that blind spot.

4. Planned spending to date is $80,000 and actual spending is $89,600. Using the thresholds above, what's the status?

Green
Amber
Red

(89,600 − 80,000) ÷ 80,000 = 9,600 ÷ 80,000 = 12% over plan, which is more than 10%: red. The PM escalates to the sponsor with options.

Common mistakes

Relying only on final inspections.

Build quality into the process with checklists and early checks.

Fixing each defect without asking why it happened.

Use a Pareto chart to find the few causes behind most defects.

Approving changes in hallway conversations.

One process: request, log, analyze, decide, update, communicate.

"Watermelon" reports: green outside, red inside.

Agree on numeric thresholds so status can't be spun.

Practice exercise

The Front Range Fitness app in Denver (cost baseline $176,000, management reserve $8,800) receives a change request: add Apple Watch check-in. Estimated extra work: 60 developer hours at $90, 16 QA hours at $60 and 8 design hours at $75, plus about one week of schedule.

(a) What's the cost impact? (b) If approved and funded from the management reserve, what are the new cost baseline and the remaining management reserve? (c) Should the PM pay for it out of the $16,000 contingency instead?

See solution

(a) 60 × $90 + 16 × $60 + 8 × $75 = $5,400 + $960 + $600 = $6,960
(b) New cost baseline = $176,000 + $6,960 = $182,960
    Remaining management reserve = $8,800 − $6,960 = $1,840

(c) No. The contingency covers identified risks on the agreed scope (for example, the payment API risk from the risk lesson). Apple Watch support is new scope, so it goes through change control and needs the sponsor's approval and funding. The PM should also point out that it leaves only $1,840 of management reserve, and that one more week may push the launch past the summer membership drive. The sponsor might reasonably decide to defer it to version 1.1.

Summary

Quality means meeting requirements: plan the standards, assure the process, control the deliverables.
Prevention and appraisal usually cost less than rework and callbacks.
A Pareto chart reveals the few causes behind most defects; fix the process behind them.
All changes go through one process; approved changes update the baselines, and new scope isn't paid from contingency.
Report status against agreed numeric thresholds so green, amber and red mean the same to everyone.

Next lesson: Earned Value Management (EVM): CPI, SPI and Forecasts. You'll measure cost and schedule performance together and forecast the final cost.