Back to the on-screen lesson ·
Defects grow with volume and cost twice, in the redo and in lost goodwill; write the standard down, check where errors are cheapest to catch, test whether a check pays, and watch the defect rate as the business grows.
Paper packet. Every task here also exists on screen, where it is checked automatically; answers written on paper are not assessed by Nydus. When you are back at a device, enter your answers there.
You will count and price the defects a larger volume brings, test whether a check pays for itself, write the cost of defects as a rule, work a growing firm's quality figures, and mark a defect rate.
Every growth step so far has added volume: a second stand, a new product, a second site. Quality is what volume tends to cost when nobody guards it: new people, more jobs, and less of the owner's own eye on each one. This lesson puts a number on that cost, using the same arithmetic as before — a rate, a count and a price — so that protecting quality can be weighed like any other decision.
| Term | What it means |
|---|---|
| Defect | A job done wrong: a return visit, a flawed mug, a rejected plate, a late delivery. |
| Defect rate | Defects as a share of jobs, such as 3 in every hundred, or 3 percent. |
| Cost of a defect | What it takes to put right, plus what it loses in goodwill. |
| Check | A planned look at the work at the point where errors are cheapest to catch. |
| Written standard | A description of what a right job looks like, clear enough for a new hire to follow. |
Bright Home Cleaning used to do 100 cleans a month, and about 3 in every hundred needed a free return visit: 3 returns. Growing to 400 cleans with new cleaners, even at the same rate it expects 4 × 3 = 12 returns.
Each return costs about 60 dollars: the cleaner's time and travel, and the chance that an annoyed customer does not rebook. At 100 cleans that was 180 dollars a month; at 400 it is 720.
$$m = \frac{v r c}{100}$$
And the rate rarely holds. New cleaners without the owner beside them make more mistakes; if the rate rises to 5 in every hundred, returns reach 20 and cost 1,200 a month. The growth that was meant to add profit has quietly added a cost that did not exist at the smaller size.
Another way: table
Returns as Bright Home Cleaning grows.
| Cleans a month | Rate | Returns | Cost |
|---|---|---|---|
| 100 | 3 in 100 | 3 | 180 |
| 400 | 3 in 100 | 12 | 720 |
| 400 | 5 in 100 | 20 | 1,200 |
Another way: steps
Count the defects. Take the volume in hundreds and multiply by the defects in each hundred. The rate must be measured, not remembered: count every return, remake or complaint for a month.
Price them. Each defect costs the work of putting it right — time, travel, materials — and some of the customer's goodwill. The goodwill is hard to measure, but leaving it out prices defects too low. A simple allowance, such as the contribution of one lost repeat order, is better than nothing.
Test the higher rate. The rate at the old size is a floor, not a forecast. Work the cost at a rate a couple of points higher, because that is where it tends to go while new people learn the job.
Test the check. A check pays if the defects it prevents cost more than the check does.
Check each figure by sense. Doubling the volume at the same rate doubles the defects and their cost; a check that halves the defects saves half their cost, never more.
Quality that lived in the owner's head has to move onto paper before it can move into new people. The sequence: write down what a right job looks like; train new staff to it; check at the point where errors are cheapest to catch — the kitchen pass, the kiln shelf, the end of each clean; count the defects every week; and when the count rises, fix what causes it rather than just redoing the work.
The written standard does not need to be long. A cleaning checklist of twenty lines, a photo of a correctly plated dish, a card listing Maya's glaze thickness and drying times: each turns the owner's judgment into something a new person can follow and a supervisor can check against.
The same mistake costs different amounts depending on when it is caught. A badly plated dish caught at the pass costs a few minutes to fix; served to the customer, it costs a remade meal and perhaps a lost regular. A flawed glaze caught before the kiln costs a wipe and a redip; caught after firing, it costs the mug. A missed room caught by the cleaner's own checklist costs two minutes; caught by the customer, it costs a return visit.
So checks belong at the last cheap point before the work leaves the business. A check at the end, after the customer has seen the result, is not a check at all: it is a complaint.
A check has a cost, so test it like any decision. A spot check costing 300 dollars a month that halves 720 dollars of returns saves 360: it pays, by 60 dollars a month. And the goodwill it protects is worth more than the arithmetic shows, because a customer lost to poor quality takes their future spending, and their friends', with them.
Checks also pay back in a way the monthly figure misses: they show where defects start. A supervisor who sees the same missed step three times can retrain one cleaner, fixing the cause, and then the check itself can be relaxed.
Redoing a defective job fixes that job; it does nothing for the next one. When the weekly count rises, ask what changed: a new hire, a new supplier, a rushed schedule, a step missing from the standard. Growth that damages quality eats the reputation that made growth possible, and the damage shows in reviews and lost regulars long after the defect rate has been brought back down.
A defect rate is only useful if it is measured the same way every week. Decide what counts as a defect before counting: a return visit, a remade plate, a mug that fails inspection, a delivery more than fifteen minutes late. Write the definition down, so a busy week does not quietly redefine a defect as something smaller.
Count the jobs as well as the defects. Twelve returns in a week of 300 cleans is a better week than eight in a week of 120, although the count is higher. The rate — defects over jobs — is the figure that can be compared from week to week and from one crew or cook to another.
Record the rate for each person or team where the numbers allow it. A business-wide rate of 4 in every hundred can hide one new cleaner at 12 and the rest at 2. The fix is then not a new check for everyone, but training for one person, which costs far less.
Plot the weekly rate on a simple chart on the wall or in a spreadsheet. A single bad week is noise; three weeks rising in a row is a signal. Owners who watch the chart see a slip early, while it is still a few defects, rather than late, when it shows up as bad reviews and lost regulars.
Finally, measure after every change: a new hire, a new supplier, a busier season, a new product. Each is a moment when the rate is likely to move, and a moment when a quick check costs little compared with the defects it can prevent.
The cost of a defect has two parts, and the second is the one owners most often leave out. The first is the direct cost of putting it right: a cleaner's hour and the drive back, a remade plate's ingredients and the cook's time, a replacement mug. That part is easy to count.
The second is the goodwill lost. Some customers who suffer a defect say nothing and simply do not come back. Some leave a review that the next hundred customers read. A simple way to put a number on it is to estimate the share of affected customers who stop buying, and multiply by what a regular customer is worth in contribution over a year. If one customer in ten who needs a return visit never rebooks, and a regular brings 600 dollars of contribution a year, each return carries about 60 dollars of lost goodwill on top of its direct cost.
The figure is rough, but it makes the comparison fair. A check that looks marginal on the direct cost alone often pays handsomely once the lost customers are counted.
A landscaping company in Atlanta added a second crew to keep up with demand, going from about 200 yard visits a month to 400. The owner had led the first crew personally; the second was led by a new hire.
Within two months, callbacks — customers asking the crew to come back for a missed edge, clippings left on a driveway, a bed not weeded — rose from about 2 in every hundred visits to 6 in the second crew's work. Each callback cost about 45 dollars in time and fuel, and the owner guessed a few customers a season were quietly switching to a rival. At 400 visits, the callbacks were 4 from the first crew and 12 from the second: 16 in all, about 720 dollars a month, against 180 before the growth.
The owner wrote a one-page checklist for every visit, with photos of a finished edge and a clean driveway, and had each crew leader walk the yard with it before leaving. That walk took about five minutes a visit, roughly 350 dollars a month of crew time. Within six weeks the second crew's callbacks fell to 3 in every hundred, bringing the monthly cost down to about 400 dollars. The check more than paid for itself before counting the customers it kept.
Large manufacturers formalize the same ideas: written standard work, checks built into each step rather than an inspection at the end, and defect rates tracked daily. A small business does not need their paperwork, but it needs the same habits once the owner is no longer beside every job.
Quality takes care of itself if the staff are good. New staff need a written standard and checks.
The defect rate stays the same as volume grows. It usually rises.
A defect costs only the redo. It also costs goodwill.
Checks are a cost with no return. Test them; many save more than they cost.
The best check is at the end. Check at the last cheap point, before the customer sees the work.
Count the hundreds of repairs.
$300 \div 100 = 3$
A busier month.
Count the comebacks.
$3 \times 4 = 12$
Four in every hundred.
Price the comebacks.
$12 \times 50 = 600$
Fifty dollars each.
Price a check that halves them.
$600 \div 2 - 250 = 50$
A final test drive at 250 a month.
Decide on the figures.
$50 > 0$
The check pays, before goodwill.
Count the returns at the old rate.
$4 \times 3 = 12$
Four hundred cleans.
Price those returns.
$12 \times 60 = 720$
A month.
Count the returns at the new rate.
$4 \times 5 = 20$
With new cleaners.
Price the new returns.
$20 \times 60 = 1200$
A month.
Price a spot check that halves them.
$1200 \div 2 - 300 = 300$
It pays at the higher rate.
Plan the fix.
$\text{a written checklist for every clean}$
Fixing the cause.
Measure Maya's own rate.
$2 \text{ in every hundred}$
Her standard.
Measure the assistant's first week.
$5 \div 50 \times 100 = 10$
Ten percent.
Price the flawed mugs.
$5 \times 12 = 60$
Twelve dollars of clay, glaze and kiln time each.
Write the standard.
$\text{glaze thickness; drying times}$
What was in Maya's head.
Check at the cheap point.
$\text{each tray before the kiln}$
A wipe, not a lost mug.
Measure again after four weeks.
$3 \div 100 \times 100 = 3$
Three flawed in a hundred.
Read the trend.
$10 \to 3$
The fix is working.
Count the remade plates.
$20 \times 4 = 80$
Two thousand plates a month, 4 in every hundred.
Price the remade plates.
$80 \times 6 = 480$
Six dollars a plate.
Price a pass check that halves them.
Bright Home Cleaning is growing to $400$ cleans a month with newly trained cleaners. About $6$ in every hundred cleans needs a free return visit, and each return costs $42$ dollars in time, travel and lost goodwill. Complete the sentence.
The larger volume is h hundred cleans, so it should expect d return visits a month.
Complete the worked solution: Bright Home Cleaning is growing to $500$ cleans a month with newly trained cleaners. About $4$ in every hundred cleans needs a free return visit, and each return costs $46$ dollars in time, travel and lost goodwill. A supervisor's spot check would cost $150$ dollars a month and halve the returns. Find what the check adds to the month's profit.
Multiply the hundreds of cleans by the rate.
$5 \times 4 =$ d
Returns a month.
Multiply the returns by their cost.
$(\text{returns}) \times 46 =$ m
What the returns cost a month.
Halve that cost.
$(\text{cost}) \div 2 =$ h
What the check saves.
Subtract the check's cost.
$(\text{saving}) - 150 =$ n
What the check adds.
Read the sign.
$\text{positive, so the check pays}$
Before counting the goodwill it keeps.
Bright Home Cleaning is growing to $700$ cleans a month with newly trained cleaners. About $3$ in every hundred cleans needs a free return visit, and each return costs $66$ dollars in time, travel and lost goodwill. Fill in the returns and their monthly cost, in dollars, at $100$ cleans a month and at $700$.
| Returns a month | Cost a month, dollars | |
|---|---|---|
| 100 cleans a month | ||
| The larger volume |
Bright Home Cleaning is growing to $600$ cleans a month with newly trained cleaners. About $5$ in every hundred cleans needs a free return visit, and each return costs $78$ dollars in time, travel and lost goodwill. A supervisor's spot check on each new cleaner's work would cost $300$ dollars a month and halve the returns. Does the check pay for itself?
A business does $v$ jobs a month. About $r$ in every hundred need putting right, and each costs $c$ dollars. Write the monthly cost of defects, $m$.
Answer:
A bike-courier firm in Chicago doubled in a year, from 2,000 deliveries a month to 4,000. Its late deliveries rose from $4$ to $8$ in every hundred, and each late delivery costs about 15 dollars in refunds and lost goodwill. Fill in the working sheet.
| Amount | |
|---|---|
| Late deliveries last year | |
| Late deliveries this year | |
| Their cost this year, dollars | |
| Rise in monthly cost, dollars |
Lesson test: one question per skill, one attempt each, no hints. Your answers are checked when you submit.
Maya's new assistant glazed $50$ mugs last week, and $2$ came out of the kiln flawed. Mark the defect rate, as a percentage.
0 |——————————| 12
Mark the position with a cross, then write the value:
You can grow without letting quality slide unnoticed. Tell someone why a defect costs more than the redo. Next: choosing what to standardize.
18. Your turn: Neighborhood Kitchen's new cooks, step 3
$480 \div 2 - 200 = 40$
It pays.