Every claim needed a smaller number

The job was substantiation, not marketing

Twenty two claims were grouped into five themes: raw material, production, transportation, building, and market and community. Each was taken separately and tested against citable evidence, with a numbered source list on every one.

The comparison work was done through published environmental product declarations rather than through vendor literature, which matters, because declarations are made to a common set of standards and can therefore be set beside one another. Four were used, one for the client product and three for competing structural materials, all declared under the same three standards.

That is a slower method than quoting a brochure and it is the only one that survives a procurement question.

Where the evidence was stronger than the claim

On energy the gap is an order of magnitude, not a percentage. Total energy use per cubic metre of finished product runs at 17,390 megajoules for the engineered wood product, 4,524 for precast concrete, and 251,593 for welded and coated steel, with cross laminated timber at 10,064. Renewable share of that energy is 82.3 per cent for the wood product against 9.3 for concrete and 6.7 for steel.

On water the ratio is similar in shape. 2.09 cubic metres per cubic metre of product against 6.22 for concrete and 157 for steel.

On raw material efficiency the margin is single digit and worth stating honestly as such. The product needs 2.27 cubic metres of logs per cubic metre of output, against 2.5 for laminated veneer lumber generally and 2.7 for glulam and cross laminated timber. That is a comparison against other wood products, not against steel or concrete.

Where the claim had to get smaller

Three claims came back true only in a narrower form, and narrowing them was the useful part of the work.

The negative carbon figure is real and it only holds while stored biogenic carbon is counted. Strip that out and the product sits at 130 kilograms of carbon dioxide equivalent per cubic metre rather than minus 655. Against concrete at 454 and steel at 21,274 that is still a strong number. Published as an unqualified negative it invites exactly one question, and the answer to that question is better given first.

The green building certification advantage was capped. Material choice touches three of the six certification categories and five criteria inside them. Scoring wood against steel across those five gives a net advantage of ten points against five, which on a forty point certification threshold is 12.5 per cent of what a project needs. Real, useful, and a long way short of the way that claim is usually made. The same analysis records that the product scores poorly on sound isolation, which is a point it loses.

And on cost the evidence was mixed and reported that way. On one reference building, research found that concrete would have been 4 per cent cheaper, while steel would have cost about the same and taken about the same time. Concrete construction was four weeks slower. A claim of cheaper does not survive that. A claim of faster does.

The carbon number, and why it carries a range

The central carbon figure comes from a synthesis of twenty one international studies, calculating the reduction in carbon dioxide over a full life cycle from replacing a cubic metre of non wood material with a cubic metre of wood.

It is not one number. The low estimate is 0.74 tonnes, assuming all waste to landfill and inefficient manufacturing. The high is 4.05 tonnes, assuming renewable energy, full recycling and best insulation. The medium is 1.83 tonnes, which is the figure normally quoted.

Quoting 1.83 without the words medium estimate, or without the range beside it, is the difference between a substantiated claim and one that falls over under questioning. The study is also from 2010, which needs saying out loud rather than discovering later.

The transport claim is modelled, and the model is stated

The freight comparison is a calculation on declared assumptions rather than a measurement of client shipments. A non load bearing partition wall, 2.7 metres high, 250 metres long, 675 square metres in total, moved on standard trailers rated at 20,400 kilograms.

On those assumptions the wood wall weighs 28.8 kilograms per square metre and takes about one truck. Steel weighs 39.0 and takes about two. Concrete weighs 200.1 and takes about seven.

It is a fair comparison because the assumptions are on the page. It is not a result, and it should never be presented as one.

What the client was left holding

Twenty two claims, each with the evidence behind it and the exact wording it can safely carry, plus the qualifier that has to travel with it. Four like for like material comparisons drawn from published declarations rather than from marketing. A carbon figure with its range attached and its date attached.

There is no measured client result in this work and none is claimed. Nothing was built, nothing was sold, and no performance improved because of it. What it produced was a set of claims that hold up when somebody pushes back, which is the only kind worth making.