Line-item pricing methodology and provenance

A defensible unit price is decomposed into its components rather than presented as one opaque number: the material cost (what the product or supply itself costs at a given quantity), the labor cost (a crew's time to install or perform the work, at a stated rate), equipment cost where specialized equipment is required, and a burden factor that accounts for overhead a contractor would apply to any job — insurance, disposal, mobilization, and similar fixed costs of doing the work. Keeping these components visible, rather than collapsing them into a single figure, is what lets a reviewer check whether a price is reasonable for the specific scope, rather than simply accepting or rejecting the total.

This decomposition also makes it possible to update one component without discarding the whole price. If local material cost changes but labor rates have not, only the material component needs to move, and the reasoning behind the change stays visible instead of being buried in a single revised number.

Equipment cost deserves its own line for work that requires it — a lift for high ceilings, a dehumidifier for a drying scope, specialized abatement equipment — because equipment costs typically scale with duration of use rather than with quantity of material, and folding it into a generic labor rate obscures that difference the next time the same scope is priced.

Construction labor and material costs vary materially by region — driven by local labor markets, supply availability, and transportation cost — so a unit price sourced correctly for one metro area is not automatically valid for another, even for an identical scope of work. A price without a stated region attached to it cannot be checked for this kind of error; a price with one can.

Prices also move over time, sometimes quickly during periods of material shortage or high demand. Every unit price should carry the date it reflects, so that a price pulled six months before the date of loss can be identified as potentially stale rather than assumed current simply because it appears on the estimate.

A price date and a region together let a reviewer answer one specific question — is this the right price for this place and this time — rather than the vaguer question of whether the number looks reasonable in general. That specificity is what makes a price defensible under later scrutiny, including in a dispute over the estimate's total.

Not all prices come from the same kind of evidence, and a defensible estimate keeps that distinction visible rather than treating every number the same way. A published pricing catalog reflects an aggregated, third-party pricing dataset for a region and time period. An organization's own observed data reflects prices it has actually paid or charged on prior, comparable projects. A manual entry reflects a specific quote or judgment call entered by a professional for this claim. An explicitly unverified placeholder reflects a number entered to keep a scope moving before a real price is available, and it should be labeled as such rather than allowed to look identical to a sourced price.

Labeling the source class matters most exactly when a price is questioned: a reviewer who can see that a number came from the estimator's own prior project data, rather than from a published catalog, knows what kind of follow-up question to ask and what kind of support to expect.

A price built from an organization's own observed data is only as reliable as the number of independent instances behind it — a price seen once, on a single unusual project, is weaker evidence than the same price seen consistently across many separate jobs. A defensible pricing methodology reports this as a confidence level or a minimum sample-size threshold, rather than presenting a one-off observation with the same weight as a well-supported average.

Confidence should attach to the specific price, not to the pricing system as a whole. A single organization can have high-confidence pricing for common items it prices constantly and low-confidence or no pricing at all for an unusual item it has rarely encountered, and an estimate should reflect that difference line by line.

A line item with no price entered yet and a line item that genuinely costs nothing are different facts, and an estimate has to keep them visually and structurally distinct. Showing an unpriced item as $0 silently understates the total and can look, to anyone reviewing the document quickly, like a completed and reasonable number rather than a gap. An unpriced line should be flagged as unpriced, excluded from a complete total, and resolved before the estimate is treated as final — not defaulted to zero and left there.

Overhead and profit (O&P) is a markup applied to the underlying line-item costs to compensate a general contractor for coordinating multiple trades on a job, and it is conventionally applied when a loss actually requires that kind of coordination, not as an automatic percentage on every claim regardless of complexity. Whether O&P applies, and what base it is calculated against, should be a stated, checkable decision on the estimate rather than a fixed habit carried over from claim to claim.

Source-tracked, confidence-scored pricing is listed as coming soon: the architecture — source class, reason code, and confidence per price — is built and in production, but no production pricing catalog is loaded yet, and the one connected pricing feed is an explicitly sandboxed vendor source guarded so it cannot be mistaken for real data. Treat pricing as not yet a source of real numbers on the platform today.