Why Your Gut Quote Is Always Right (and Why That's the Problem)

29 October 2026

One painter quotes 16 hours. Another quotes 20 for the same job. A month later, both say it came in right on the quote. They can both be telling the truth, and that is exactly the problem.

Picture two painters walking the same job on the same morning. Same walls, same trim, same client. The first one looks around, thinks "two days," and quotes 16 hours of labor. The second thinks "two and a half days" and quotes 20. Both win work at those numbers, and a month later, both will tell you the job came in right on the quote.

They can't both be right about how long the job needed. They can both be right about how long it took. That gap is what this article is about.

Here is the idea, and a century of research backs it up: a gut quote looks accurate because the job stretches or squeezes to fit it. When that happens, a quote can be off for years and the painter never finds out, because every job seems to prove it right.

A century of research on why estimates fool us

None of this is new. Psychologists, economists and project managers have been circling the same problem since the 1920s. They studied office workers, students, software teams and billion-dollar megaprojects, not painters, but the pattern keeps showing up wherever people estimate their own work. Here's the short version, in order.

1. 1924–1933The Hawthorne studies: being watched changes the workWestern Electric's Hawthorne Works studies became famous for the idea that output shifts when workers know they're being studied. The label "Hawthorne effect" is often traced to a 1953 research-methods chapter by John French and was popularized by Landsberger (1958), and a later re-analysis by Jones (1992) found the original evidence thinner than the legend. The lesson for painters survives either way: a crew on a watched, best day is not a crew on an ordinary Tuesday.
2. 1954Meehl: simple rules versus expert judgmentPsychologist Paul Meehl reviewed studies that pitted experienced clinicians against simple statistical rules. Across the studies he reviewed, the rules matched or beat the experts.
3. 1955Parkinson's LawC. Northcote Parkinson opened an essay in The Economist with a line that became a law of project management: "Work expands so as to fill the time available for its completion."
4. 1967Bryan and Locke: the time you're given becomes a targetTwo researchers recast Parkinson's joke as a goal-setting phenomenon: the time allowed for a task acts as a goal, not just a limit.
5. 1974Tversky and Kahneman: anchoringIn a landmark paper in Science, estimates made by adjusting from a starting number stayed too close to it. In a painting business, the anchor is the client's budget, the last quote on a similar job, or the number you said out loud on the walk-through.
6. 1979The planning fallacy gets its nameKahneman and Tversky named it in "Intuitive prediction: Biases and corrective procedures." People plan from the details of this task and neglect what happened on similar tasks before. That second kind of information, how comparable tasks actually went, is the idea later turned into reference class forecasting.
7. 1979Dawes: even a crude formula beats intuitionRobyn Dawes showed that simple, equally weighted formulas beat clinical intuition when predicting a number from other numbers. The formula doesn't have to be clever. It has to be consistent.
8. 1994Buehler, Griffin and Ross: even the worst case was too shortStudents estimated when they'd finish their honors thesis. Their best guesses averaged 33.9 days, the work took 55.5, and even their worst-case estimates came in short (Buehler, Griffin and Ross, 1994).
9. 1997Goldratt: padding gets spentIn Critical Chain, a business novel, Eliyahu Goldratt argued that safety time added to an estimate gets used up: people start late because there's time ("student syndrome"), and the work expands into whatever is left.
10. 1999Parkinson's Law in the labIn three experiments, people who believed they had more time available took longer on the same task than people who didn't.
11. 2003Lovallo and Kahneman: the inside view versus the outside viewIn Harvard Business Review, they argued that forecasts built from the inside (this job, this crew, this plan) run optimistic, and that the cure is the outside view: how did jobs like this one actually go?
12. 2004Kruger and Evans: break it down and the estimate gets honestWhen people unpacked a task into its parts before estimating, their time estimates went up and the planning fallacy shrank. "The house" hides work that "forty surfaces" shows.
13. 2004Jørgensen: experts do best with a recordA review of expert estimation in software projects found no evidence that formal models beat experienced estimators across the board. Experts did well when they had domain knowledge, and the review's advice includes using documented data from previous tasks.
14. 2005Roy, Christenfeld and McKenzie: we misremember how long things tookPeople underestimate how long past tasks took, and that faulty memory is part of why they underestimate future ones. If last month's "two-day job" really took three, the next quote starts from the wrong number.
15. 2006Flyvbjerg: reference class forecastingBent Flyvbjerg turned the outside view into a method: forecast from the actual results of comparable past projects, not from the plan for this one.
16. 2009Parkinson's Law in the field: German registration officesA study of German vehicle registration offices found that more staff per case did not mean better service (Jochimsen, 2009). Extra capacity got absorbed.
17. 2009Kahneman and Klein: when you can trust your gutA famous skeptic and a famous defender of intuition agreed on the conditions: intuition is reliable only in an environment regular enough to predict, with a real chance to learn it through practice and feedback. Feeling confident is not evidence of being right.
18. 2012Halkjelsvik and Jørgensen: the big reviewA review of time-prediction research from origami to software found underestimation reported more often than overestimation in the engineering and management studies, and showed anchoring, incentives and how a task is broken down all shaping the estimate.
19. 2021Chen and Hall: early finishes disappearIn project management, work finished early routinely isn't delivered early, so it never offsets the jobs that run late. One reason they cite: if a task is finished early, less time may be allowed for it next time.
20. 2023Flyvbjerg and Gardner: almost nothing lands on the planAcross some 16,000 major projects, only 8.5% came in on time and on budget, and only 0.5% did that and delivered the benefits promised (How Big Things Get Done; figures as reported in the authors' Harvard Business Review article).

Smart, experienced people miss their own estimates, and they miss in a consistent way. That isn't a painter problem. It's a human one. What makes painting interesting is why the misses stay invisible.

Why gut quotes seem to work

A painting job isn't a fixed block of work that takes a fixed amount of time. The walls are fixed. The hours are elastic. Give a crew 20 hours and the job tends to take about 20. Give them 16 and it tends to take about 16, one way or another. The job bends to the quote in two directions.

The stretch

Quote too many hours and nobody slacks off on purpose. The pace just settles to the time available, exactly as Parkinson joked and the lab studies later showed. Prep gets a little more careful, breaks run a little longer, the tidy-up gets thorough. The job finishes on the quote, so the quote looks right. And if the crew does finish early, there's every reason not to say so, which is the pattern Chen and Hall describe: early finishes rarely show up, so they never teach you anything.

The squeeze

Quote too few hours and the crew pushes. The owner paints the last evening and doesn't log it. Prep gets lighter than it should. The job lands on the quote, or a little over, and gets remembered as "a tough one," an exception, rather than as evidence that the quote was short. Roy, Christenfeld and McKenzie's finding makes it worse: memory quietly shortens how long it really took.

Now add Kahneman and Klein's test for when intuition can be trusted: a regular environment, and a real chance to learn it through feedback. Painting is regular enough. It's the feedback that's broken. Our reading of the research, and it's a reading rather than a finding about painters: when the job always fills the quote, the gut never gets corrected, so years of experience can repeat the same error instead of fixing it.

The hidden cost

A gut quote that "always comes in" feels free. It isn't. The cost just never shows up on one job where you can see it.

• Overquoting loses work you never hear about. The client who went with someone cheaper doesn't tell you your number was 50% high. And on the jobs you do win, if your crew is paid by the hour, you're paying wages for the stretch.
• Underquoting quietly eats margin. Unpaid owner hours, lighter prep and call-backs don't appear on the quote. They appear in a year-end profit that's lower than it should be, with no single job to blame.
• Anchoring compounds it. The last quote becomes the starting point for the next one, so an error doesn't average out. It gets copied.
• You can't hand a gut to someone else. A new estimator, a second crew or a bigger job type all need a number that lives somewhere other than one person's head.

The fix: measure, rate, record, recalibrate

The research points the same way every time. Stop estimating from the inside of this one job, and start from evidence about jobs like it. For a painter that's a four-step loop, and the loop is the whole point: each turn of it makes the next quote sharper.

1. Measure the job, area by area

Your gut sees "a three-bedroom repaint." A measurement sees every wall, ceiling, door and length of trim, each with its own size and condition. That's the unpacking effect at work: break the job into parts and the hidden work shows up before you've committed to a number, not after.

2. Apply a known rate

Price each surface from a production rate you trust, instead of from the feel of the day. A rate is the outside view in a single number: it's what jobs like this one have taken before. Dawes's lesson applies here. The rate doesn't have to be perfect to beat a guess. It has to be consistent, so that when it's wrong, it's wrong in a way you can see and fix.

3. Record what actually happened

Log the real hours on every job, and by area where you can. Count the early finishes as carefully as the overruns, because they're the ones that vanish. Make it safe for the crew to report finishing early: if finishing early only earns a tighter quote next time, people will stop mentioning it, and Chen and Hall's research suggests they already do. Time several ordinary jobs rather than one closely watched one, so your numbers reflect a normal day, not a best day.

4. Recalibrate the rate

Compare quoted hours with actual hours after every job and adjust the rate when the pattern says so. This is the step that puts the feedback back. It's exactly what Kahneman and Klein say intuition needs to become trustworthy, and it's what Jørgensen's review recommends: experience plus documented data from past jobs.

None of this throws away experience. Your eye for a tricky job, a bad substrate or a fussy client is real, and the research doesn't say otherwise. It says experience needs a memory and a scoreboard. Give it both and your gut gets better every year, instead of just getting more confident.

"But my crew does both"

Any painter who runs a crew will push back here. Some days the team squeezes a job down to the quote, other days it stretches to fill it. That's normal, and it doesn't break the loop. It's the reason the loop judges the pattern across many jobs, not any single day.

Start with the average. One job running long or short tells you very little. Ten or twenty jobs where quoted hours and actual hours keep landing apart tells you the rate is off. If they land together on average, the rate is right and the daily swings are noise.

Then remember that the swings don't cancel out. Chen and Hall found that work finished early rarely turns into time handed back, while overruns land in full. A crew that stretches on Monday and squeezes on Thursday doesn't break even. You carry the squeeze and never bank the stretch. A few habits keep both in view:

• Give each area its own hour target. "Kitchen and hall, six hours" shows the drift the same day, not at the end of the week. It's the unpacking effect again, applied on site.
• Give early finishes somewhere to go. Keep a short list of fill-in work, such as touch-ups, a small add-on or prep for the next job, so spare time becomes productive hours instead of a slower afternoon.
• Treat a squeeze as a quality question. When a crew beats an underquoted job, check what gave. Skipped prep or a thin coat often comes back later as a callback.
• Adjust from the average, never from one fast job. Cut the rate after a single quick day and the crew learns to hide the next one.

What nobody has measured yet

To be straight about the evidence: we couldn't find a published study that compares gut quotes with actual hours across painting businesses. The research above comes from students, offices, software teams and megaprojects. The pattern is consistent enough that we'd bet on it holding on a job site, but a bet isn't data.

So we're starting to collect it. The Surfacely benchmarks hub already gathers painters' anonymous answers on production rates and pricing, and we're building a free calibration tool to record gut estimates against actual time, surface by surface. When there's enough data to say something honest, we'll publish it here, including if it proves us wrong.

Sources

1. Roethlisberger, F. J., and Dickson, W. J. (1939). Management and the Worker. Harvard University Press. Background: Harvard Business School, Baker Library, Hawthorne Studies collection.
2. Landsberger, H. A. (1958). Hawthorne Revisited. Cornell University.
3. Jones, S. R. G. (1992). Was there a Hawthorne effect? American Journal of Sociology, 98(3), 451–468. doi.org/10.1086/230046
4. Meehl, P. E. (1954). Clinical versus Statistical Prediction: A Theoretical Analysis and a Review of the Evidence. University of Minnesota Press.
5. Parkinson, C. N. (1955, November 19). Parkinson's Law. The Economist. economist.com
6. Bryan, J. F., and Locke, E. A. (1967). Parkinson's Law as a goal-setting phenomenon. Organizational Behavior and Human Performance, 2(3), 258–275. doi.org/10.1016/0030-5073(67)90021-9
7. Tversky, A., and Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185(4157), 1124–1131. doi.org/10.1126/science.185.4157.1124
8. Kahneman, D., and Tversky, A. (1979). Intuitive prediction: Biases and corrective procedures. TIMS Studies in Management Science, 12, 313–327. An earlier version circulated as a 1977 technical report.
9. Dawes, R. M. (1979). The robust beauty of improper linear models in decision making. American Psychologist, 34(7), 571–582. doi.org/10.1037/0003-066X.34.7.571
10. Buehler, R., Griffin, D., and Ross, M. (1994). Exploring the "planning fallacy": Why people underestimate their task completion times. Journal of Personality and Social Psychology, 67(3), 366–381. doi.org/10.1037/0022-3514.67.3.366
11. Goldratt, E. M. (1997). Critical Chain. North River Press.
12. Brannon, L. A., Hershberger, P. J., and Brock, T. C. (1999). Timeless demonstrations of Parkinson's first law. Psychonomic Bulletin & Review, 6(1), 148–156. doi.org/10.3758/BF03210823
13. Lovallo, D., and Kahneman, D. (2003). Delusions of success: How optimism undermines executives' decisions. Harvard Business Review, 81(7), 56–63. hbr.org
14. Kruger, J., and Evans, M. (2004). If you don't want to be late, enumerate: Unpacking reduces the planning fallacy. Journal of Experimental Social Psychology, 40(5), 586–598. doi.org/10.1016/j.jesp.2003.11.001
15. Jørgensen, M. (2004). A review of studies on expert estimation of software development effort. Journal of Systems and Software, 70(1–2), 37–60. doi.org/10.1016/S0164-1212(02)00156-5
16. Roy, M. M., Christenfeld, N. J. S., and McKenzie, C. R. M. (2005). Underestimating the duration of future events: Memory incorrectly used or memory bias? Psychological Bulletin, 131(5), 738–756. doi.org/10.1037/0033-2909.131.5.738
17. Flyvbjerg, B. (2006). From Nobel Prize to project management: Getting risks right. Project Management Journal, 37(3), 5–15. doi.org/10.1177/875697280603700302
18. Jochimsen, B. (2009). Service quality in modern bureaucracy: Parkinson's theory at work. Kyklos, 62(1), 44–64. doi.org/10.1111/j.1467-6435.2009.00422.x
19. Kahneman, D., and Klein, G. (2009). Conditions for intuitive expertise: A failure to disagree. American Psychologist, 64(6), 515–526. doi.org/10.1037/a0016755
20. Halkjelsvik, T., and Jørgensen, M. (2012). From origami to software development: A review of studies on judgment-based predictions of performance time. Psychological Bulletin, 138(2), 238–271. doi.org/10.1037/a0025996
21. Chen, B., and Hall, N. G. (2021). Incentive schemes for resolving Parkinson's Law in project management. European Journal of Operational Research, 288(2), 666–681. doi.org/10.1016/j.ejor.2020.06.006
22. Flyvbjerg, B., and Gardner, D. (2023). How Big Things Get Done. Currency. Figures as reported in Flyvbjerg, B., and Gardner, D. (2023), How Frank Gehry delivers on time and on budget, Harvard Business Review.

← Back to all articles