1. What measurement is, and what is at stake
Measurement is the calculation of the quantities of work to be carried out, derived from the project drawings. It is expressed in physical units: square metres of façade, linear metres of beams, cubic metres of concrete, numbers of doors. It is what fills the quantity column of the priced bill and of the estimate.
Its reliability depends on two inseparable factors: the quality of the drawings supplied, an uncoordinated drawing without dimensions cannot support precise measurement, and the method applied, since every trade has its own measurement rules and applying them incorrectly distorts the result even from perfect drawings.
ℹ The fundamental rule: quantities cannot be more precise than the drawings Quantities taken off outline design drawings are accurate to ± 15 %, and that is normal. Trying to reach ± 5 % from outline design drawings is an illusion that produces false confidence in the estimate. Good practice is to state the accuracy of the estimate explicitly by stage, and to size the contingency accordingly. A client who knows their outline estimate is accurate to ± 15 % will not be surprised by a 12 % movement at developed design.
2. Four methods, one per stage
| Stage | Method | Accuracy | Basis of calculation | Recommended contingency |
|---|---|---|---|---|
| Feasibility and concept | Overall rates per m² or per unit | ± 25 % | Floor area × rate for the building type | 15 – 20 % |
| Outline design | Approximate quantities by package | ± 15 % | Site plan plus schematic sections | 10 – 15 % |
| Developed design | Semi-detailed quantities | ± 8 % | Dimensioned developed design drawings at 1:100 | 8 – 10 % |
| Detailed design and tender | Detailed quantities, item by item | ± 3 – 5 % | Dimensioned detailed design drawings at 1:50 or better | 5 – 8 % |
The overall rates method is the fastest and the least precise. It is perfectly suited to an initial economic go or no-go, but it must never be used as the basis of a contractual budget or of a tender. The detailed item-by-item method is the only one that supports a reliable comparison of tenders.
3. The detailed method: five steps
Step 1, identify and organise the drawings
Before any calculation, list the drawings available with their revision number and date. Working from unidentified or superseded drawings is the first source of error. Organise them by floor level and by type: site plan, floor plans, sections, elevations, details.
Step 2, calculate the reference areas
Calculate or verify the reference areas from the drawings: habitable area, gross floor area, gross usable area. Those areas serve as the check: if the quantities for a package produce a rate per square metre that is inconsistent with market ranges, that is usually the sign of an error in the take-off.
Step 3, measure by trade
The general rule is to measure each item in its natural unit, square metres for areas, linear metres for runs, cubic metres for volumes, numbers for discrete elements, applying the rules for deducting openings set out in the relevant technical standard.
Step 4, check the rates
Once calculated, check each package subtotal against the reference area. The resulting rate has to be consistent with market ranges for the type of project and the region. An aberrant rate signals either an error in the take-off or an atypical item that deserves an explanation.
Step 5, contingency and residual uncertainty
Apply a contingency proportionate to the accuracy of the stage. That contingency is not padding, it is an honest quantification of the uncertainty that remains. It has to be explicit in the document given to the client, with its justification.
4. The items that get forgotten
Several items recur in incomplete take-offs. Here they are in order of how often they are missed.
External works and services: access earthworks, foul and surface water drainage, water supply, roads, fencing, landscaping. These represent 3 to 8 % of the works budget and are absent from most outline estimates because the site plan is not yet detailed enough.
Openings and coordination items: holes cut through concrete walls for ducts, sleeves in slabs, fire stopping, making good. These are often priced as lump sums but they have to be budgeted; missing them out generates unavoidable variations.
Circulation, vertical and horizontal: stairs, landings, corridors, entrance halls. These areas produce no habitable area but they consume structure, tiling, paint and electrical work, sometimes 20 to 30 % of the real quantities.
Flat roofs and upstands: frequently missed on multi-storey buildings because the roof level does not appear among the floor plans. The waterproofing package is 3 to 6 % of the budget.
Enabling works and site set-up: asbestos removal, selective demolition, site establishment, skips, hoarding. Often treated as a lump sum, but leaving them out distorts the total cost.
5. The eight commonest sources of error
| Source of error | Trades affected | Typical impact on the estimate | Severity |
|---|---|---|---|
| Incomplete or undimensioned drawings | All | 10 to 30 % gap between the estimate and the tenders received | ★★★★★ |
| Drawing revision not identified | All | Quantities taken off a superseded revision, changes not picked up | ★★★★★ |
| External works and services omitted | External works | An unpredictable 3 to 8 % of the works budget left unbudgeted | ★★★★★ |
| The wrong rule for deducting openings | Façades, partitions, painting | 5 to 15 % overstatement on the finishing packages | ★★★ |
| Measuring gross instead of net area | Tiling, flooring, painting | 10 to 20 % overstatement depending on the complexity of the plan | ★★★★ |
| Openings and coordination items not counted | Structure, mechanical, electrical | Coordination lump sums missing, disputes on site | ★★★★ |
| Circulation areas understated | Floors, partitions | Circulation is 15 to 25 % of the habitable area and is often forgotten | ★★★ |
| Basement or flat roof left out | Structure, waterproofing | A whole item unbudgeted, up to 15 % of the structural budget | ★★★★★ |
6. The checklist before any take-off
Do not start a take-off before every point below has been ticked. A take-off begun on an incomplete set of drawings has to be done again, which doubles the time spent.
| Check | Ask | If missing or incomplete | Blocking? |
|---|---|---|---|
| Drawings identified by revision and date | Architect | Measuring the wrong revision means doing it all again | Yes |
| Areas per floor shown on the drawings | Architect | Calculating them yourself introduces error | Recommended |
| Dimensioned sections and elevations | Architect | Uncertain storey heights, structural volumes understated | Yes for structure and façades |
| Junction and upstand details | Architect | Interfaces cannot be quantified, lump sums have to be increased | Recommended |
| Geotechnical investigation available | Geotechnical engineer or client | Foundations calculated on a standard assumption, a strong risk | Yes with a basement or difficult ground |
| Structural engineer's design note | Structural engineer | Concrete sections unknown, structure estimated on rates alone | Recommended at developed design |
| Mechanical and electrical loads defined | Services engineers | Services packages estimated on rates, ± 20 % accuracy | Recommended at developed design |
| Client's brief signed off (areas, levels, uses) | Client | Scope creep if the brief moves after the take-off | Yes |
7. Measurement and building information modelling: where are we?
Digital models have been promising for ten years to automate measurement. In practice, extracting quantities automatically from a model works reliably only where the model has been built with strict discipline, every element correctly categorised, dimensioned and parameterised.
The reality in French architectural practices is that most models are not yet built to the level of detail required for reliable quantity extraction. The economist often has to rework the extracted quantities by hand, which does not always save time against traditional measurement from PDF drawings.
The underlying trend is nonetheless clear: practices that model rigorously produce quantity extractions that can be used directly. On projects run in coordinated modelling, ask systematically for the model export and the quantity schedules alongside the drawings.
In summary
Quantities are the basis of the whole of construction economics, and their precision governs the precision of the estimate, of the priced bill and of the tender analysis. Four methods match the four main stages: overall rates at feasibility (± 25 %), approximate quantities at outline design (± 15 %), semi-detailed at developed design (± 8 %), detailed item by item at detailed design (± 5 %). The five items most often forgotten are external works, openings through the structure, circulation areas, the flat roof and the site set-up. The three most serious errors are incomplete or unidentified drawings, a basement or roof left out, and measuring from a superseded revision. The eight-point checklist has to be cleared before any take-off begins.
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.
Articles liés
- The construction economist's tools and methods: the complete picture
- Measurement rules by trade: the practical guide
- Priced bill, pricing schedule, specification: the three documents explained
- Underestimating the budget at concept stage: why the first figures are always too low
- Construction costs in France: rates per m², building types and the factors that move them
Sources: the BT construction cost indices published monthly by INSEE on its 2010 base, used to update the figures · construction cost statistics from SDES, the statistical service of the ministry responsible for construction · regulatory texts published in the Journal officiel and consolidated on Légifrance for the requirements cited.