There is science ­- and there is pseudo-science -­ and the danger with pseudo-science is that it attempts to reduce complex issues that involve experience and intuition to formulae. So it is with valuation based on ‘price per square foot’ which, in the property market, has the capacity to produce some spectacular mistakes and anomalies if it is wielded by the naïve or the unwary.

It is beguiling for both buyers and sellers. Buyers, particularly those with limited local knowledge, love the lifejacket it gives them in working out where they are in a sea of (mis)information that threatens to drown them. If they know that another flat in the building has gone for £1000 per square foot, then why not open the bidding at £950? What they don't know is that the other flat was in immaculate condition, fully furnished, overlooking the river and on a longer lease. Theirs overlooks a railway and has seen better days. The seller, or his agent, has been economical with the truth­ which is not a crime while caveat emptor prevails.

This focus on ‘price per square foot’ is relatively recent and has much to do with the influx of Asian buyers into London in the mid-1990s. In Asian markets it makes some sense. Ceiling heights are pretty uniform, new developments and flats the norm and the product more commoditised. As most Asian buyers in London were buying in new developments - ­and often off-plan from a developer’s road-show on a Saturday afternoon in the Mandarin Hotel in Hong Kong ­- there was not much else to go on. No-one came to too much harm as a rapidly rising market and a two-year building programme covered up all but the most egregious price anomalies. However, the habit of referring everything back to ‘price per square foot’ stuck and is now used universally to justify just about anything. The patina of science has given it respectability.

Some guidance around the pitfalls may be helpful.

First, define the terms. Are we talking external, gross internal or net internal?

External is measured around the outside of the walls and used primarily by builders particularly in new-build houses when talking about building costs. There could be a 25% difference between this and the net internal. Gross internal includes staircases and landings and is the one used mainly by estate agents. Any flat, on one floor, is going to feel a lot bigger than a house of the same gross internal -­ especially if it is on five floors -­ as so much of the house is taken up with stairs. Net internal excludes stairs and landings but will probably give a better comparison between flats and houses.

Then there is what is being measured. Up until a year ago, loft rooms were not included. Now, if the roof height is above 1.5 metres, it is included -­ which means that you have to be careful comparing the current ‘price per square foot’ with anything that was sold over a year ago. Roof terraces are a grey area and are not normally included, though they can add a considerable amount to a flat’s value. As a rough rule of thumb, you could value a roof terrace at about 50% of internal space in order to compare like with like. A larger than usual garden, or access to communal gardens, will always add value.

On top of this, you have to bear in mind that the companies that provide the measuring service vary in quality and diligence. Hangovers, Friday afternoons and malfunctioning measuring machines mean that you should allow around a 10% margin of error, whether you are talking external, gross or net. Shaky science, this.

And all this before you get to issues peculiar to London, the main one being ceiling heights. Central London is predominantly a city of houses ­- or was, as most of the largest are now converted into flats, particularly in Belgravia and South Kensington with their stucco-fronted squares that are as characteristic of London as rain at Wimbledon. Typically, the grand rooms of these houses are on the first floor, with the next best usually on the ground and second floors. The higher up you go, the lower the ceilings and the longer the walk, if there is no lift; and when you get to the top, unless you are vertically challenged, the usable floor space is cut back by sloping ceilings. Basements, often damp, dark and vulnerable to burglars, are normally chosen only by those wanting more space for their money and a garden. Not surprisingly, prices vary accordingly because, subconsciously, buyers are actually buying cubic, not square, footage and it is perfectly possible to find a building where the price per square foot for the basement is half that of the first floor.

Add to this the variables of views, layout and, of course, condition.

Though they may not admit it, the average buyer is a sucker for the house or flat that is in immaculate condition, with the sound of Mozart on the B&O music system and the smell of cedar in the bedroom cupboards - which is why developers drive expensive cars. What is less well known is that the marker flats and houses are often sold lock, stock and barrel with the Philippe Starck furniture and the Picasso lithographs included. Needless to say, the ‘price per square foot’ quoted by the developer to establish the benchmark for his next (minimalist) creation is unlikely to be qualified by the fact that there was the equivalent of £300 per square foot of designer furniture in the price of the benchmark.

So does ‘price per square foot’ have any use in valuation? Yes, but only if you compare apples with apples and pears with pears but the ability to select the correct fruit only comes with knowledge and experience. Long may this last, as it is one of the reasons we have a business. We use it - but mainly as a back-check to make sure that our apples are truly apples. Ultimately, the truism remains, well … true, that valuation is an art not a science and that you should treat attempts to make it otherwise with healthy scepticism. As H.L. Mencken rather nicely put it, “For every complex problem, there is usually a simple answer and it’s usually wrong.”