What You Are Actually Promising When You Quote a Price

Why this matters

When you hand over a number, the customer hears one thing: this is what it costs. You have made four promises, and only one of them is the number. The other three are promises about how much work, how long, and what result, and you made all of them after looking at the building for twenty minutes. Shops that bleed on quoted work almost never bleed on the price promise. They bleed on the ones they did not know they were making, because an unwritten promise has no boundary and a boundary is the only thing that limits what an unknown can cost you.

The number is the smallest promise in the document

A price is a forecast of your own cost plus a transfer of risk. The forecast part is arithmetic and you can check it. The transfer part is the interesting half: by naming a fixed number, you have agreed that if the job takes more work than you forecast, the extra comes out of your margin rather than the customer's pocket. That is not a defect in fixed pricing. It is the product. Customers pay a premium for a number that does not move, and a shop that sells certainty and then tries to claw it back on every job is selling something it does not have.

The question is never whether you carry risk. It is which specific unknowns you agreed to carry, and whether you knew you were agreeing.

The promises you made out loud

Four are legible on the page, assuming you wrote them:

  • Price. This number, valid until a stated date. If your quote carries no expiry, you have promised it indefinitely against a supply chain that does not care.
  • Scope. These operations, on this equipment, in this location. Scope is the boundary of the price promise, and a price promise without a scope boundary is not a price promise, it is an open account.
  • Schedule. A start window, a duration, or both. Duration is where trades under-promise least and pay most, because a duration promise is a capacity promise: every hour you spend past your estimate on this job is an hour the next job does not get.
  • Result. What the customer will have when you leave. This is the one shops write loosest and the one that decides disputes.

The promises made by silence

The other set is the reason quoted work goes wrong. Nobody wrote these. Both parties assumed them, in opposite directions.

  • Access. You quoted as though you can reach the work. The customer assumed reaching it was part of the job. Neither of you wrote down who moves the stored boxes, cuts the access opening, or pays for the two hours spent finding out that the panel is behind finished drywall.
  • Condition of what you are connecting to. You quoted the new work. The existing work it ties into is somebody's, and by default it becomes yours the moment you touch it.
  • Fitness for the customer's actual goal. They wanted the back room comfortable. You quoted equipment. If the equipment is right and the room is still wrong, whose problem is that? Silence answers this against you, because you are the professional and they will say so.
  • Restoration. The wall you opened, the landscaping you crossed, the finish you disturbed. Restoration to original condition is a promise that costs real hours and is almost never in the scope list.
  • Cleanup and disposal. Small, predictable, and consistently unwritten.

Field key: who carries it if nobody says

Promise Named in most quotes Who carries the unknown by default
Price Yes You, for the quote's validity window
Scope of operations Usually You, for anything a reasonable reader would call part of the named work
Schedule Sometimes You, and your next customer
Result the customer wanted Rarely You, because you are the one who is supposed to know
Access to the work Rarely You
Condition of adjacent existing work Almost never You, once you touch it
Restoration of finishes Almost never Contested, and contested means you

The pattern is not subtle. Every row that goes unwritten lands on the shop. That is the whole reason exclusions, assumptions and allowances exist as instruments: each one moves a specific row off the default.

The promises worth refusing

The sharpest thing on a good quote is what it declines to promise, and declining is not weakness. Three worth refusing outright:

Do not promise a result you cannot measure. "The upstairs will be comfortable" has no test. "Airflow at each upstairs register within the range noted on the attached measurement sheet" has a test, and you can be held to it because you can prove it. If you cannot state the measurement, do not state the outcome.

Do not promise the condition of work you did not do. Touching an existing circuit, an existing drain, or an existing flue does not make it yours, but silence about it does. The refusal is one sentence naming the existing component and stating that its continued serviceability is not covered.

Do not promise a duration you have not scheduled. "About a day" spoken on a driveway becomes "you said one day" in a dispute. Give a window with a named condition on it, or give a start date only.

Refusing in writing reads as competence to the customer who is comparing three quotes, because the other two refused nothing and therefore thought about nothing.

Worked example: one quote, decomposed after the fact

A three-line quote for work in an older single-family home. Sold hours, by line:

  • Line A, equipment swap in an accessible location: 6.0 hours
  • Line B, tie-in to existing distribution through a finished ceiling: 5.0 hours
  • Line C, controls, commissioning and customer walkthrough: 3.0 hours
  • Total sold: 14.0 hours

The job delivered in 19.0 hours. That is 5.0 hours over 14.0 sold, a 36 percent overrun in delivered hours against sold hours. Those 5.0 hours are unsold technician time the shop absorbed; they are not a margin figure and they do not net against anything, because nobody paid for them.

Where they went is the finding. Lines A and C landed on estimate at 6.0 and 3.0. All 5.0 extra hours sat on Line B, which ran 10.0 against 5.0 sold, a 2.0x multiple on that line alone. The ceiling had a framing member exactly where the tie-in had to pass, and a previous remodel had left a second, abandoned run in the same cavity that had to be identified before anything could be cut.

So the exposure on this quote was never spread across 14.0 hours. It was concentrated in one line, and that line was the only one whose access could not be verified from the walk-through. Lines A and C were priced against things the estimator could see and touch, and both came in exactly where they were priced. The estimator was not bad at estimating. The estimator was accurate on 9.0 of 14.0 sold hours and completely blind on the other 5.0, and priced all 14.0 with the same confidence.

The instrument that would have moved this was one sentence on Line B: this line assumes the tie-in path through the ceiling cavity is clear of framing and abandoned material, and if it is not, the additional work is quoted before it proceeds. That sentence costs nothing to write. If the assumption had been wrong and stated, the correction happens as a phone call before the ceiling is opened, and the customer decides whether to pay for the extra 5.0 hours or to accept a different routing. If the assumption is wrong and unstated, the correction happens after the ceiling is open, at which point the shop cannot stop without leaving a hole and cannot proceed without eating the difference.

Note what the sentence does not do: it does not make the 5.0 hours disappear. Someone still spends them. It decides who, and it decides before the decision gets expensive.

What would flip the reading

Two conditions change the call.

Repeat work in a housing stock you know cold. If you have done forty tie-ins in the same tract-built vintage and thirty-eight of them were clear, pricing Line B firm is a reasonable bet and the assumption sentence is belt-and-braces. The bet is defensible because you have a base rate. It stops being defensible the moment the address is outside the stock you know.

A customer who will not accept conditional lines. Some buyers, particularly on competitively bid commercial work, treat any conditional language as a non-responsive quote. Then the conditionality has to move into the price rather than the text, which is a contingency decision and belongs in a separate calculation. See the related articles on contingency sizing rather than trying to solve it in the scope text.

How to check a quote before it leaves the office

Read your own quote as the customer's attorney would, and answer four questions out loud:

  1. What result am I on the hook for, and can I measure it? If the answer names a feeling rather than a measurement, either add the measurement or delete the result language.
  2. Which line has the unknown? Every quote has one. Name it. If you cannot name it, you have not looked hard enough, because a job with no unknown is a job you have done in that exact building before.
  3. What did I touch that I did not install? Every existing component in the path gets either an exclusion, an assumption, or a condition note.
  4. If this job runs long, which line runs long? If you cannot answer, your duration promise is a guess wearing a suit.

The check takes a few minutes on a quote that will govern a week of work, and it catches the concentrated-exposure case in the worked example above, which is the common shape: not a quote that was wrong everywhere, but a quote that was right nearly everywhere and blind in one place.

References

  • U.S. Small Business Administration (SBA), guidance on contracting and pricing for small businesses
  • Contract terms, warranty language and remedies for breach are governed by state law and differ for consumer and commercial customers; have your own attorney review your standard quote form
  • See related: The Difference Between an Estimate, a Quote and a Contract; The Conditions Clause: Protect the Quote; Confirming Scope Before You Roll the Truck