Back to the on-screen lesson ·
Money already spent cannot be recovered by any choice made now, so it should not steer the choice; compare only what each option will cost and bring from here on, count what can still be sold, and price what staying on a losing course costs.
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 set aside sunk costs, compute a gain from here on, pick out the figures that count, compare two options on their future figures with resale included, price what staying on a losing course has cost, and choose between repairing and replacing.
Opportunity cost adds a cost that is not in the books. This lesson takes one out: money already spent, which feels like it should count and does not. You met the idea in the money course; here it is applied to the strategic decisions where the pull is strongest.
| Term | What it means |
|---|---|
| Sunk cost | Money already spent that no choice made now can recover. |
| Future costs and benefits | The figures that depend on the decision, from here on. |
| Resale value | What the old thing can still be sold for; it belongs to the future, not the sunk cost. |
| Sunk-cost trap | Continuing something because of what has already gone into it. |
| Fresh-start question | Starting today with nothing spent, would I choose this? |
Neighborhood Kitchen spent 4,000 dollars launching a breakfast menu: new equipment, a sign, a week of samples. Three months on, keeping it open costs 2,200 dollars a month in staff and food and brings in 1,800.
The owner's instinct: 'We've put 4,000 into this; we can't stop now.' But the 4,000 is spent whether she keeps the menu or closes it. Closing does not waste it, and keeping does not recover it.
The only question is the future: from here on, the menu brings 1,800 and costs 2,200 — a loss of 400 dollars a month. Every month she keeps it to 'get the launch money back' adds 400 to what has been lost; six more months would add 2,400. The right call, on these figures, is to close it, or to change it so its future figures work.
One part of the past does count: anything that can still be sold. If the breakfast griddle would fetch 600, closing brings in 600 that keeping does not, and that belongs on the closing side of the comparison.
Another way: table
The breakfast menu, from here on.
| Figure | Amount | Counts? |
|---|---|---|
| Launch spending | 4,000 | No: sunk |
| Cost a month from now | 2,200 | Yes |
| Income a month from now | 1,800 | Yes |
| Griddle's resale if closed | 600 | Yes |
Another way: steps
Write the sunk figure down, then set it aside. Naming it helps: once the 4,000 is written on its own line, marked 'spent', it is easier to leave out of the arithmetic.
List the future for each option. Keeping the menu: its monthly costs and income. Closing: any costs of closing — telling customers, a last stock that will be wasted — and any money that comes back, such as equipment sold.
Use the same time span for every option. A month against a month, or a year against a year; comparing a year of repairs with a month of lease decides nothing.
Compare and decide. The option with the better future wins.
Check by asking whether the sunk figure appears anywhere in the comparison. If the same number is in both options, it cancels and can come out; if it is in only one, it is either not sunk (it can still be recovered) or it has been counted by mistake.
Stopping feels like admitting the money was wasted, and continuing feels like keeping faith with it. That is why owners pour money into a failing product, a dud location or software that never works: each new payment feels like protecting the old ones. The pull grows with the size of the sunk figure and with how public the decision was — a menu announced with a banner is harder to drop than one tried quietly.
The cure is to ask one question: 'If I were starting fresh today, with nothing spent, would I choose this?' If not, the money already spent is not a reason to. Asking someone who had no part in the original decision helps too; they feel none of the pull.
The sunk-cost trap has a price, and it can be worked out. Each month a losing menu stays open adds that month's loss; each extra year of repairs on a van that should have been replaced adds the difference between the repairs and the lease. Writing that figure down — 'keeping it three more months will cost 1,200' — turns a feeling ('we can't give up now') into a choice with a price, which is much easier to decline.
The easiest time to avoid the trap is before anything is spent. When a new idea starts — a menu, a booth, a product — write down what its figures must look like by a date, and what will happen if they do not: 'if breakfast is not covering its running costs by the end of the third month, we close it.' Decided before the money is spent, the rule is easy to agree to; followed when the date comes, it saves the months the trap would otherwise take.
The breakfast launch shows what the kitchen's customers do not want, and that knowledge was bought with the 4,000. Use the lesson; do not try to earn the money back from the same idea. And the same goes the other way: a sunk cost is not a reason to stop something that is now working, just because it was expensive to start. A costly launch that now makes 300 a month from here on should be kept, whatever it cost to begin.
Not every past payment is sunk. A deposit that will be returned, stock that can still be sold, equipment with a resale value, a prepaid service that can be transferred: each is money that has left the account but can still come back, and each belongs in the future comparison. The test is simple. If a choice made now changes how much of the payment returns, the recoverable part is not sunk; only what is gone whatever happens is.
This matters most when a business is deciding whether to stop something. A café that closes its breakfast service can sell the griddle; a booth that gives up its space may be able to sublet it; a business that cancels a software contract early may get part of the year's fee refunded. Leaving these out makes stopping look worse than it is, and keeps the business on a losing course longer. Counting them sometimes turns the decision around.
The opposite mistake is to count as sunk a cost that is still to come. A lease signed for two years is not sunk in its first month; its future payments are real future costs, unless the lease cannot be ended and the space cannot be used any other way, in which case they are committed whatever is decided and cancel out of the comparison. The question to ask of every payment, past or future, is the same: does this decision change it? If yes, it counts; if no, it does not.
Keeping a short written list of commitments — leases, contracts, subscriptions, with their end dates and exit terms — makes this much easier. When a decision arrives, the owner can see at once which future payments the decision can change and which it cannot. It also shows when commitments are about to end, which is the cheapest moment to reconsider any of them, and it prevents the unpleasant surprise of a contract that renewed itself automatically for another year.
A bakery opened a second location in a new shopping development, spending 85,000 dollars on the fit-out. A year later the development was still half empty, and the second store lost about 3,500 a month even after its share of the shared costs. The owners kept it open for another year because, as one of them put it, 'we can't walk away from 85,000'.
A new accountant asked them to look only forward. Keeping the store another year would lose about 3,500 × 12 = 42,000 on the development's current traffic. Closing would cost 9,000 to break the lease early, but they could sell the ovens and display cases for about 20,000, a net gain of 11,000 from closing. Over the next year, closing was better than keeping by about 53,000 — and the 85,000 appeared on neither side.
They closed the store, moved the best staff to the original location, and used the ovens' sale to add a wholesale line from the original kitchen, which covered its costs within four months. The fit-out was still lost, as it had been from the day the development failed to fill; what they stopped losing was the 3,500 a month.
Researchers who study decisions found that people and organizations reliably keep spending on projects in proportion to what they have already spent, even when the future figures say stop. Naming the pull, and asking the fresh-start question, is the simplest defense.
Stopping wastes what was spent. It is spent either way.
Keep going until it earns back the investment. The investment is not recoverable by continuing.
Sunk costs mean the idea was a mistake. They say nothing about what to do next; only future figures do.
Everything in the past is sunk. What can still be sold belongs to the future.
Sunk costs teach nothing. They teach; they just do not decide.
A committed future payment is sunk. A lease signed but not yet paid is a future cost; it only cancels out of the comparison if no choice can change it.
Admitting a sunk cost is admitting a mistake. Plenty of good decisions turn out badly because the world changed; stopping them promptly is the good decision now.
The fresh-start question is only for big decisions. It works just as well for a small subscription, a supplier or a weekly market booth, and the small ones add up.
The money already spent should be earned back before the business changes course. The only way to earn anything is from the choices still open, so choose the best of them.
Stopping wastes what was spent. The spending is gone whether the project continues or stops; stopping only avoids spending more on something that will not pay. The waste, if any, happened when the money was committed, not when the project is ended.
Set the build cost aside.
$6000 \text{ spent}$
Sunk.
Read the app's future cost.
$150 \text{ a month to host}$
If kept.
Read its future benefit.
$60 \text{ a month of phone time saved}$
Customers rarely use it.
Work the gain from here on.
$60 - 150 = -90$
A loss each month.
Decide on the future alone.
$\text{switch it off; a 20-dollar calendar instead}$
Whatever the app cost to build.
Set the launch aside.
$4000$
Sunk.
Work six more months of keeping it.
$6 \times (1800 - 2200) = -2400$
Future income less cost.
Work closing now.
$+600$
The griddle sold.
Compare the two futures.
$600 - (-2400) = 3000$
Closing is better by 3,000 over six months.
Ask the fresh-start question.
$\text{would we launch this today?}$
No.
Keep the lesson.
$\text{customers want coffee, not breakfast}$
Bought with the 4,000.
Set this year's repairs aside.
$2400 \text{ spent}$
Sunk.
Work a year of keeping the old van.
$350 \times 12 = 4200$
Repairs expected.
Work a year of leasing a new one.
$420 \times 12 = 5040$
The lease.
Take off the old van's sale.
$5040 - 1500 = 3540$
Replacing's cost from here on.
Compare the two costs.
$4200 - 3540 = 660$
Replacing saves 660 over the year.
Check the sunk figure is absent.
$2400 \text{ in neither option}$
It cancels.
Decide on the van.
$\text{replace the van}$
Whatever the repairs cost this year.
Set the display stand aside.
$1200 \text{ spent}$
Sunk.
Work each fair from here on.
$350 - 200 = 150$
Contribution less the booth fee.
Decide whether to keep going.
Neighborhood Kitchen has spent $5724$ dollars launching a breakfast menu. From here on, keeping it open costs $2000$ dollars a month in staff and food, and brings in $2100$ dollars a month. What is the menu's gain, in dollars a month, from here on? Write a loss with a minus sign.
Answer:
Complete the worked solution: Bright Home Cleaning has spent $2800$ dollars on its old van's repairs this year. Keeping it another year should cost about $200$ a month in repairs. Leasing a replacement would cost $700$ a month, and the old van would sell for $1400$. Compare the next year.
Work a year of repairs on the old van.
$200 \times 12 =$ k
Keeping's cost from here on.
Work a year of the lease, less the old van's sale.
$700 \times 12 - 1400 =$ q
Replacing's cost from here on.
Take replacing's cost from keeping's.
$(\text{keeping}) - (\text{replacing}) =$ d
Above zero, replacing is cheaper.
Leave this year's repairs out.
$2800 \text{ is sunk}$
Spent whichever is chosen.
Ask the fresh-start question.
$\text{starting today, which would I choose?}$
The question that breaks the trap.
Neighborhood Kitchen has spent $3974$ dollars launching a breakfast menu. From here on, keeping it open costs $2000$ dollars a month in staff and food, and brings in $2800$ dollars a month. Should she keep the breakfast menu open?
Maya is deciding whether to keep selling at a craft fair. Mark every figure that should count in the decision.
This task has no paper form; do it on a device.
A gym has spent $8085$ dollars on custom booking software that is still unfinished. Finishing it would cost $3500$ dollars more and be worth $4800$ dollars to the gym; switching to a ready-made system would cost $3300$ dollars and be worth $6000$ dollars. Fill in each option's gain from here on, in dollars.
| Gain from here on, dollars | |
|---|---|
| Finish the custom software | |
| Switch to a ready-made system |
Neighborhood Kitchen's breakfast menu cost $2700$ dollars a month to run and brought in $1200$ from its third month on. The owner kept it for another $8$ months 'to earn back the launch' before closing it. How much did those extra months cost, in dollars?
Answer:
A café's espresso machine has had $1600$ dollars of repairs this year and will need about $150$ a month more to keep going. A new machine can be leased for $300$ a month, and the old one would sell for parts for $450$. Fill in the next year's figures, in dollars.
| Amount | |
|---|---|
| Keeping, next year's cost | |
| Replacing, next year's cost | |
| Saving from replacing |
Lesson test: one question per skill, one attempt each, no hints. Your answers are checked when you submit.
Bright Home Cleaning has already spent $2125$ dollars on repairs to its old van this year. Keeping it running for the next year would cost $1800$ dollars more; leasing a replacement for the year would cost $1100$ dollars. Both would do the same work. By how many dollars does the cheaper option cost less from here on?
Answer:
You can stop money already spent from steering the next decision. Tell someone the question that breaks the sunk-cost trap. Next: finding what limits a business's growth.
17. Your turn: Maya's craft fair, step 3
$\text{keep going: each fair gains 150}$
The stand's cost does not enter.