The scene
The scene
Elias and his partner budgeted $40,000 for their kitchen renovation. They had saved for two years. The plan was clear. The contractor was hired.
Week two: the original floors did not match the new kitchen finish. They would look strange. The contractor offered to redo the floors for another $4,800. It felt obvious. They said yes.
Week three: with the new floors, the doors looked dated. Another $3,200. Already in scope, basically.
Week four: the wall paint clashed with the new doors. Another $1,800.
Week five: while the walls were open, it made sense to update the wiring. Another $5,600.
Week seven: the bathroom door, just visible from the new kitchen, suddenly looked terrible. Another $2,400.
Week eight: a small leak was discovered in the laundry. Already there, already a job to do. Another $4,200.
By the end of the project, the renovation had cost $68,000. Each individual addition had felt small and reasonable in context. None of them would have been approved if the original budget had included them. All of them got approved because the project was already underway.
Elias and his partner were not careless with money. They had saved for two years specifically to avoid going over budget. The over-budget happened anyway, decision by small defensible decision.
The same pattern shows up in renovation projects studied by industry researchers. Across multiple surveys, the average renovation in Australia, the UK, and the US comes in 15 to 25 percent over budget. The mechanism is consistent: each addition feels small relative to the total already committed.
What your brain just did
What your brain just did
Our minds use the amount already spent as the reference for whether the next addition is worth it. Elias is not extravagant. His brain compared each new $4,000 expense not against his original $40,000 budget, but against the much larger evolving total, the way all our brains do when we are already committed to a project. This behaviour has a name: Sunk Cost Fallacy (often combined with the closely related effect of incremental escalation).
What to do instead, in one move
What to do instead, in one move
The fix is to set a contingency at the start, in writing, and treat any addition beyond it as a separate project to decide on later. Twenty percent contingency on top of the original budget is a common figure. Once the contingency is gone, the project stops at the next natural pause point, and any further work goes into a fresh decision with a fresh budget, in a few months. The pause is the only mechanism that breaks the chain.
TL;DR
- Situation: You start a renovation or major project with a clear budget. As the project progresses, additions accumulate. The final cost is significantly above the original budget.
- What your mind does: It evaluates each addition against the growing total of what has already been spent, not against the original budget (this is called Sunk Cost Fallacy, see below).
- Consequence: Projects routinely come in 15 to 70 percent over budget, not from any single bad decision, but from a chain of small additions that felt reasonable in context.
- What to do: Set a 20 percent contingency at the start in writing. Once the contingency is gone, pause the project at the next natural break and treat any further work as a separate decision with a fresh budget, weeks or months later.
What to do
- Set a contingency at the start of the project. Twenty percent of the original budget is a starting point used by many renovation professionals. Write it down before starting.
- Track additions against the contingency, not against the original budget. The contingency is the buffer. Once it is gone, the buffer is gone.
- When the contingency is depleted, pause. Stop work at the next logical break point. Any further additions become a separate project with a separate budget, decided fresh in a few weeks or months.
- Get fixed-price quotes where possible, not estimates. Estimates are negotiable. Fixed prices are not. The brain treats them differently.
What not to do
- Do not compare additions to the growing project total. Compare them to the original budget. The original was the decision.
- Do not say yes to additions on the spot, in the middle of the project. The brain is in commitment mode. Ask for a day to think.
- Do not assume that "while we are at it" is a saving. It usually is not. The work would have cost the same as a separate project later.
A renovation that grows by ten percent each week feels small each week. At the end, it is seventy percent over budget.
Want to understand why this happens?
The Sunk Cost Fallacy is the brain's habit of letting past investments (money, time, effort) drive future decisions, even when the past investments cannot be recovered.
For renovations, the bias takes a specific form. Each new addition is evaluated against the total already committed, not against the original budget. The original $40,000 fades from view. The current $52,000 becomes the new reference point. The next $4,000 addition feels small relative to $52,000, so it gets approved. Then the new $56,000 becomes the next reference point, and the cycle continues.
The mechanism is partly emotional commitment. Once a project is underway, walking away from any part of it feels like wasting the parts already done. The brain prefers to complete the project at any cost rather than stop with the project in a half-state. So additions get approved to "finish properly" rather than rejected to "stay within budget".
It is not you. It is how every human brain handles commitments that are already in motion.
What the research found
Hal Arkes and Catherine Blumer's 1985 study documented the Sunk Cost Fallacy across multiple decision types. In one experiment, participants who had paid for a season ticket attended more games (including ones they did not particularly want to attend) than participants who received the same ticket for free. The money already spent altered the future decision, even though it should not have.
Subsequent studies extended the finding to projects, relationships, careers, and household renovations. The pattern is consistent: past investments distort future decisions. The fix is not to ignore the past, but to evaluate future decisions against future criteria, with the past treated as separate.
For renovations specifically, multiple industry surveys have documented that the average project comes in 15 to 25 percent over budget. The Houzz Australia 2024 Renovation Report found 38 percent of homeowners reported renovation costs exceeded their original budget. The mechanism, across all the surveys, is the same: incremental additions that felt reasonable in context. None of the homeowners reported a single big bad decision. All of them reported a series of small ones.
The fix is the contingency, set in advance, treated as the real ceiling. Twenty percent buffer at the start is the single most consistent recommendation from renovation professionals and behavioural economists alike.
"The money already spent should not affect the next decision. Yet it does, consistently, across every domain we have studied." — Hal Arkes and Catherine Blumer (paraphrased from The Psychology of Sunk Cost, 1985)
This is called Sunk Cost Fallacy. Hal Arkes and Catherine Blumer, Organizational Behavior and Human Decision Processes (1985).
Get the next pattern before it gets you.
Get free weekly patterns explained in 60 seconds.
Free forever · Unsubscribe anytime
Related decisions
I have been saving these loyalty points for two years. The trip I wanted now costs 50 percent more points than when I started.
Behind this pattern: Endowment Effect (Daniel Kahneman, Thinking, Fast and Slow (2011)).
The annual plan said save 30 percent. I clicked annual. I cancelled three months in.
Behind this pattern: Hyperbolic Discounting (Dan Ariely, Predictably Irrational (2008)).
I have been meaning to set up automatic savings for two years.
Behind this pattern: Present Bias (Richard Thaler & Cass Sunstein, Nudge: Improving Decisions About Health, Wealth, and Happiness (2008)).
I moved $5,000 to a 0 percent card. I planned to clear it before the promo ended. I did not.
Behind this pattern: Planning Fallacy (Daniel Kahneman & Amos Tversky, Intuitive Prediction: Biases and Corrective Procedures (1979)).