How to read a construction estimate and spot the hidden risks

A professional construction estimate is a structured document setting out the forecast cost of a project together with its assumptions, its sources and the limits of its reliability. Several elements are indispensable.

• A stated reference area: habitable area, gross floor area, net usable area, every building type has its standard reference area. An estimate with no stated area cannot be read.
• The date of the estimate and the price base: rates move with inflation. A 2021 estimate applied without updating in 2025 can understate costs by 10 to 20 % depending on the package.
• A breakdown by trade package: groundworks, structure, roof, façades, mechanical, electrical and so on. Without a breakdown the overall rate cannot be checked.
• The items outside the works: design team and consultant fees, VAT (5,5 % to 20 % depending on the project), external works, insurance, finance costs. Fees, VAT, insurance and finance costs do not appear in the design team's estimate, which covers the forecast works cost: they belong to the budget of the operation, which the client sets. A design team estimate that does not contain them is therefore not deficient; an operation budget that forgets them is. As a benchmark, the MIQCP indicated in 2008, for design fees, 10 to 15 % of the works envelope for small operations and 7 to 10 % for larger ones, and observed, in very general terms, that the works envelope including VAT represents some 50 to 60 % of the provisional financial envelope including VAT, excluding land and furniture, at delivery-date value.
• The contingency: of the order of 10 % at outline design, 8 % at developed design. These are assumptions of this page: no published scale sets these rates, and the MIQCP refers to a provision "to be assessed case by case". A contingency that is missing or below 3 % is the single most serious warning sign.
• The documented assumptions: specification level (standard or premium), site configuration, whether there is a basement, the regional cost factor applied.

The reality is that most estimates clients receive, especially at concept stage, do not contain all of these. The checklist of fifteen warning signs below lets you spot the gaps and gauge the associated level of risk.

Run every estimate you receive through this grid. Each warning sign detected is a potential overrun to be quantified.

The six five-star warning signs are the critical ones. Signs 1 (no contingency), 5 (items outside the works missing) and 13 (refurbishment with no survey) are the most frequent and the most expensive. On an operation budget, signs 1 and 5 combined can account for a gap of several hundred thousand euros for a project with 1 M€ of works, an order of magnitude given by the fictitious example in the next section.

Here is a fictitious example. A client receives an estimate for a scheme of eight dwellings (around 650 m² habitable area), presented by the architect at a concept design meeting. The document shows a total of 983 000 € excluding VAT. Checking the estimate gives the following table.

This example illustrates the most classic and most dangerous mechanism of all: the estimate shows a works cost consistent with the market, but it leaves out the fees, the VAT, the contingency and the price fluctuation allowance entirely. The client has budgeted 983 000 € for a scheme whose full cost is 1 638 000 €. The 655 000 € gap is a 66,6 % understatement, not of the works, but of the total cost of the operation. The works account for only 75 000 € of it; the fees and VAT, about 432 000 €, are not an error in the design team's estimate, which covers the forecast works cost, and the contingency and inflation allowances make up the rest, about 148 000 €.

ℹ Works cost against total cost: the most expensive confusion there is The fundamental error in reading an estimate is to confuse the cost of the works with the total cost of the operation. The works cost is what the contractors cost. The total cost is what the whole operation costs: works plus fees plus taxes plus finance charges. For a residential scheme the total cost is typically 1,5 to 2 times the works cost in urban areas, excluding the land. An investor who compares the scheme budget with the works figure in the estimate is out by 50 to 100 % in the feasibility analysis.

A quick check on an estimate starts by comparing the overall rate proposed with the market range for the building type concerned. A rate more than 15 % below the bottom of the range is a strong signal of an anomaly.

These are outline-design ranges applicable to standard projects in the regions. For the Paris region add 15 to 25 %. For the overseas territories add 20 to 40 %. For seismic zones 3 and 4 add 8 to 15 % on the structure. For refurbishment the range is wider because it depends on the condition of the existing building, and a contingency of the order of 20 % is a prudent assumption, to be adjusted according to the surveys of the existing building.

If you receive an estimate and want to gauge how reliable it is before committing to a funding decision, here are the five questions to put to whoever produced it.

• Question 1, which area is this rate calculated on? Habitable, gross floor, usable or something else? If the answer is vague or undocumented, the estimate cannot be verified.
• Question 2, what contingency is included? It must be explicit, stated in euros, and separate from the rates. An answer along the lines of "the contingency is in the rates" is a warning sign.
• Question 3, is this figure a works estimate or an operation budget? If it is the design team's estimate, the absence of fees, VAT and insurance is normal: it is for the client to add them to arrive at the cost of the operation.
• Question 4, are the assumptions documented? Specification level, site configuration, presence of a basement, regional factor. With no assumptions document the estimate cannot be defended if there is an overrun.
• Question 5, what is the date of the price base? An estimate built on prices more than 18 months old needs updating. In an inflationary period, the gap over eighteen months is measured package by package with the corresponding INSEE BT index, between the value date of the estimate and the date of the tenders; the BT01 index, all trades, stood at 138,9 in July 2026 (2010 base, published on 15 September 2026).

Three situations typically call for an independent audit.

A feasibility estimate produced by the architect in line for the commission is structurally optimistic; the architect has a commercial interest in the project being judged viable. An independent second opinion on economic feasibility, before signing a purchase contract worth several million euros, is the most profitable investment of the whole operation.

This is the last opportunity to recalibrate the budget before the architectural decisions become hard to change. An independent outline-design estimate, produced on the final concept drawings, confirms that the scheme is still financially viable before committing to detailed design and application fees.

When the tenders received diverge significantly from the design team economist's estimate, more than plus or minus 15 %, an external audit establishes whether the estimate was wrong or the tenders are abnormal, and lets you decide whether to award or to re-tender on an informed basis.

A reliable estimate has six elements: a stated reference area, a date and price base, a breakdown by trade package, the items outside the works (fees, VAT, external works) where it is an operation budget, an explicit contingency (of the order of 10 % at outline design), and documented assumptions. The fifteen warning signs surface the hidden risks, and the six five-star signs are the critical ones. The annotated example reveals a 66,6 % gap between the works figure and the real total cost of a scheme of eight dwellings. Confusing works cost with total cost is the most frequent and most expensive error there is. Three situations justify an audit by an independent economist: before buying a site, before submitting the planning application, and after an abnormal tender.

The nature of the figures quoted

The percentages and amounts quoted on this page are orders of magnitude for framing, not measurements. They illustrate mechanisms and proportions, and they substitute neither for a tender exercise nor for an estimate prepared on drawings.

• The economic risks of a construction project: the complete picture
• The outline stage cost estimate: the first budget assessment of a construction project
• Construction costs in France: rates per m², building types and the factors that move them