Why a Published Building Rate Is Not What You Pay

There are at least three different numbers attached to any piece of building work, and they are not close together. There is the published rate for the work itself. There is the price a main contractor puts in a tender. And there is the sum a homeowner actually pays, which is the one people are usually asking about when they look a figure up.

Most published domestic cost writing quotes the first number and answers a question about the third. The gap is not a markup, a rip-off or regional variation. It is a set of exclusions that the price books state openly in their own front matter, and which almost nobody carries through to the doormat.

This note walks the chain from one to the other, using a single edition of one price book so that every step can be checked against the same source.

What a Measured Rate Contains

A measured-works rate is the price of doing a defined unit of work: laying a square metre of tiles, forming a metre of verge, hanging a door. It is built up from the material at its prime cost, the labour hours to fix it, a share of the plant, and a thin allowance for the contractor’s overheads and profit.

What it does not contain is stated in plain words. Spon’s Architects’ and Builders’ Price Book 2011 sets out its own structure and then says of the two measured-works sections: “All prices in Parts 3 & 4 exclude the main contractor’s preliminaries costs.” On tax, the same front matter says prices “do not include for VAT, which must be added if appropriate”. On fees, its approximate estimating rates are described as “exclusive of fees for professional services”.

Three exclusions, all declared, all compounding. The book is not hiding anything. The figures that circulate online are simply lifted out of it, or out of something derived from it, without the exclusions attached.

A note on the money. The figures below come from the 2011 edition, so they are 2011 prices and should not be used as current rates. What carries forward is not the money but the relationships: the proportion preliminaries add, the split between labour and material, the shape of the fee scale. Those have not moved much, and they are what the argument rests on.

Step One: Preliminaries, Around 11%

Preliminaries are the most misunderstood line in construction pricing, usually because the word sounds like head office overhead. It is not. Preliminaries are the site.

Spon’s prices a worked example of the preliminaries for a £3,000,000 project. The total is £333,265, which its editors describe as “approximately an 11% addition to the value of measured work”. They also give the market range: “The value of the Preliminaries items in recent tenders received by the editors varies from a 10% to 13% addition to all other costs.”

What Preliminaries Actually Buy

The itemisation is more useful than the percentage, because it shows what a measured rate leaves out:

Management and staff alone is 4.5% of the whole measured value. So a rate that excludes preliminaries excludes the site manager, the scaffold and the toilet. Comparing that rate against a builder’s quotation and calling the difference a price rise is comparing work with work plus the site it happens on.

That worked example is a commercial-scale job, and the percentage should be treated as an order of magnitude rather than a constant. For a homeowner the item list is the durable content: when you read a per-square-metre figure somewhere, ask which of those lines it is carrying. Our renovation budgeting note puts a main contractor’s preliminaries and profit at 10–15% on a domestic job, which is the same correction arriving from the other direction.

Step Two: Professional Fees

Fees are excluded too, and the addition is not a constant. The Valuation Office Agency publishes a scale for exactly this purpose, because its own replacement costs “exclude professional fees and other charges”:¹

The scale covers feasibility study, architect, CDM coordinator, quantity surveyor, structural and civil engineer and mechanical and electrical engineer, with a further 2 to 4 percentage points for more complex buildings. The VOA’s own footnote to the top band is worth reading: schemes up to £750,000 “would be charged on a time basis”, which is to say that at domestic scale a percentage is a budgeting device rather than how anybody actually invoices.

Two cautions. This is a rating valuation tool built for a different market, so take the shape of the correction from it, not a domestic rate. And note the direction: fees fall as jobs get larger, which means the smallest jobs carry the heaviest fee percentage, at the same time as they carry the heaviest size premium in the section below. The two corrections push the same way on a house.

Step Three: VAT

Most domestic renovation work is standard-rated at 20%, and it applies to labour as well as materials.² Reduced and zero rates do exist, and they matter a great deal on conversions and long-empty properties, but they are the exception; we cover which work qualifies in our note on VAT on home renovations and the DIY reclaim scheme.

For the purposes of this chain, the point is simply that VAT sits at the end of it and multiplies everything before it, including the preliminaries.

Why You Hear Lower Numbers

Three legitimate reasons a figure someone quotes at you sits well below your own bill:

• Trade counters display ex VAT. The Price Marking Order 2004 defines the price shown to a consumer as the final price including VAT, and it binds consumer sales only.³ Trade merchants selling to business customers sit outside it and generally display ex-VAT figures, so the same bag of cement carries labels 20% apart depending on which counter it sits at.
• Subcontract invoices carry no VAT at all. Since March 2021 the domestic reverse charge applies to most CIS-reported supplies between VAT-registered construction businesses, so the customer accounts for the VAT instead of the supplier charging it.⁴ A price overheard from the subcontract side of a job is silently 20% below what the same work bills at to a homeowner.
• Not every trader is registered. Registration is compulsory only above £90,000 of taxable turnover in twelve months, so a trader below it need not charge VAT at all (some register voluntarily anyway).⁵ The real gap is smaller than it looks: an unregistered trader cannot recover the VAT on their own materials and has to carry that cost in the price, so on a job split evenly between labour and materials the unregistered quote lands around 8% below the registered one, not 20%. That is a tax position, not an efficiency, and it is worth knowing before you read it as a saving.

The Whole Chain, in Order

Take £40,000 of published measured work and walk it to the doormat:

About a third, before professional fees. And because preliminaries are a range rather than a number, the honest version of the bottom line is a band: at 10% it is £52,800, at 13% it is £54,240. Then fees go on top of that as a separate, scale-dependent line.

This is the single most expensive misreading in UK domestic cost data, and it is entirely avoidable, because every exclusion in it is printed in the front of the book the rate came from.

Two Things the Rate Does Not Know: Where, and How Big

Even corrected for the chain above, a published rate still assumes a place and a project size.

Where. Every rate in Spon’s is an Outer London rate, struck at a tender price index of 468, and the book prints the correction as a table: Inner London +9%, the South East -9%, Scotland -11%, Wales -17%, the West Midlands -20%, the North and Yorkshire and Humberside -21%, East Anglia, the East Midlands and the North West -22%, and Northern Ireland -45%. The prose puts it more bluntly: “Regional variations for certain inner London boroughs can be up to 14% higher, while prices in the North and Yorkshire and Humberside can be as much as 21% lower.” Top to bottom that is a factor of 1.98, so the identical specification is worth nearly twice as much in one part of the United Kingdom as in another.

How big. The VOA publishes the size effect as a scale, because “small projects will usually cost more per unit than larger ones”.¹ Its rates are struck for a notional £4,000,000 contract, and the adjustment runs +10% at a quarter of a million or less, +4% at one million, zero at four million, and down to -10% above forty million. Top to bottom that is a factor of 1.222, so the same specification legitimately costs about 22% more at domestic scale than at the largest.

Price books are explicit about their own band when you look. Spon’s Minor Works section says it is “intended to apply to a small project in the outer London area costing about £160,000 (including preliminaries)”. A rate taken from a book struck for a £4m contract and applied to a £30,000 kitchen is understating it before anything else has been done to it.

The Small-Works Premium Lands on Buying, Not Hours

Everyone knows small jobs cost more per unit. The usual explanation is that the labour is less efficient: fewer metres a day, more setting up and clearing away, more travel per unit of output. The clearest public evidence in a UK source points the other way, and it is a rare controlled comparison: the same book, the same year, the same specification, priced twice.

Spon’s prices concrete interlocking tile roofing, Marley Eternit Anglia, 387 mm by 230 mm to a 75 mm lap on battens and reinforced underlay, in both its Major Works and its Minor Works sections:

The rate is 10.4% higher on the small-works page. The labour hours are identical, the labour cost is identical to the penny, and every one of the £2.35 is in the material column. The prime cost of the tiles themselves rises 23.1% purely for buying in small quantity.

One caveat, stated because the book states it: the two sections carry different mark-up conventions, 3.5% for overheads and profit on built-up rates in Minor Works against 2.5% in Major Works, so roughly a percentage point of the ten is the book’s own convention rather than the market. That does not touch the finding, because the labour line is identical across the two and the entire gap sits in material either way.

Two things follow for anyone pricing a small job:

1. Modelling a small-works uplift by adding labour hours gets the right total for the wrong reason, and the wrong total as soon as the labour and material mix changes. Apply it to a labour-heavy operation and you overstate; apply it to a material-heavy one and you understate.
2. On a small job, the saving is in buying, not in the programme. Consolidating orders, buying full packs, taking delivery in one drop and using a merchant account rather than a retail counter is where the small-works premium actually lives. Our note on reading a builders’ merchant quote covers the mechanics.

The same convention trap catches anyone adding their own margin: a book rate already carries a thin overheads-and-profit allowance, and putting a further percentage on top of it double counts.

How to Read Any Published Figure

Before you compare a number you have found against a number a builder has given you, five questions settle whether they are the same kind of thing at all:

1. What basis is it on? Measured work, contract sum, or delivered price including everything?
2. Ex or inc VAT? Assume ex VAT unless it says otherwise, and check whether the source is a trade or a consumer one.
3. Are preliminaries in or out? If the figure came from a price book’s measured-works section, they are out.
4. Are professional fees in or out? Usually out, and the addition moves with project size.
5. Where and how big? Which region the rate is struck for, and which contract size.

A published figure that sits 30% below your builder’s quotation is very often not a cheaper builder. It is a different question, answered correctly, about work rather than about a project. Once the basis is recorded at every step, the price books are excellent and the comparison holds; it is the unlabelled number, lifted out of its own front matter, that does the damage.

If you are at the stage of getting prices in, our notes on getting and comparing building quotes and on the difference between a quote and an estimate deal with the other half of the problem: making sure the numbers you are given are comparable with each other.