The First Number You Read Sets Your Price

Open the assignment. Scroll to the total. Now decide, honestly, whether that number covers your cost.

You can't, not cleanly. The moment you read it, it became the reference point for every judgment that followed, and you spent the rest of that job negotiating against the carrier's number instead of against what the work actually costs you. That isn't inexperience and it isn't a character flaw. It's one of the most reliably reproduced findings in the study of human judgment, and it has been in print for fifty years.

The research: anchoring

In September 1974, Amos Tversky and Daniel Kahneman published "Judgment under Uncertainty: Heuristics and Biases" in Science. They described three mental shortcuts people use to estimate things they can't calculate. The third one is the one that should worry anyone who prices work for a living.

People estimate by starting from whatever number is in front of them and adjusting. And, the paper says, "adjustments are typically insufficient." Different starting points produce different final answers, biased toward wherever you started. They named it anchoring.

The demonstration is almost insulting in how simple it is. Subjects were asked what percentage of countries in the United Nations were African. First, a wheel of fortune was spun in front of them. They said whether the true figure was higher or lower than the spun number, then gave their estimate. The groups who saw 10 answered a median of 25 percent. The groups who saw 65 answered 45 percent.

A number produced by a spinning wheel, in full view, with no claim to relevance, moved the answer twenty points. And the paper adds a line worth reading twice: "Payoffs for accuracy did not reduce the anchoring effect." Paying people to get it right didn't help.

Experience is not the defense you think it is

The obvious objection is that these were students guessing at trivia, and you've written ten thousand estimates. Fair. So take the part of the same paper that deals with experts making numerical judgments in their own field.

When specialists put confidence ranges around a quantity, those ranges should contain the true value 98 percent of the time. Across the studies Tversky and Kahneman cite, the true value fell outside the stated range about 30 percent of the time. The paper is blunt about who this applies to: "This bias is common to naive and to sophisticated subjects, and it is not eliminated by introducing proper scoring rules."

Expertise narrows the error. It does not remove the anchor. That is the finding.

The restoration translation

Restoration has a structural problem most trades don't: the price list arrives before the scope is known. You get an estimate built on what an adjuster could see, on a published unit-price database, for a building nobody has opened yet. Then your crew opens the wall.

That sequence hands you an anchor at the worst possible moment. And three things follow from it.

Your cost stops being part of the conversation. Once the carrier's total is on screen, the question in your head quietly changes from "what does this job cost me to do?" to "can I make this number work?" Those are not the same question, and only one of them has an answer you can verify.

Discovered scope gets measured against the anchor, not against cost. You find more water than anyone estimated. The honest response is a documented supplement. The anchored response is to ask whether the extra work fits inside the number you already accepted, and to absorb it when it roughly does. Absorbed scope is the anchor taking your margin in real time.

Partial computation anchors too. The paper showed this with a multiplication problem: students given five seconds to estimate 8x7x6x5x4x3x2x1 answered a median of 2,250, while students shown the same digits ascending answered 512. The true answer is 40,320. Your field version is the first rough take at a job, made in the driveway, before anyone has opened anything. That number sticks too.

What actually counters it

Not defiance. The carrier is going to pay against its price list whatever you believe, so "ignore Xactimate" is advice that will cost you work. The counter is sequence.

1. Produce your number before you read theirs. A burdened cost estimate from your own closed-job data, built independently. The order is the whole intervention. A number you generated first is much harder to drag.
2. Reconcile the two line by line, not in total. Totals invite you to split the difference. Line items tell you specifically where your cost and their price list diverge, and that list is your supplement documentation.
3. Set the walk-away before the anchor lands. Your minimum gross profit per production hour is a policy, decided in a quiet month. Decided while looking at an assignment, it isn't a policy, it's a negotiation with yourself.
4. Watch your own history for the same trap. If last year's jobs were underpriced, your historical cost data is just a different anchor. Check it against what the work costs today, not what you billed for it then.

The point of job costing was never the report. It's that a number you built yourself, in advance, is the only thing in the room that the carrier's estimate can't move. That's most of what fractional CFO work looks like in a restoration shop: getting your number to exist before theirs arrives. If yours doesn't exist yet, that's worth a conversation.

General guidance for restoration and reconstruction owners, not a substitute for advice tailored to your company's numbers.

← Back to all posts