The numbers,
and where they come from

This site sets out to sit in an awkward but useful place: detailed enough to satisfy a curious reader, but never simplified to the point where the reader would have to unlearn something later. Holding that line depends entirely on the numbers being right, so the sourcing discipline below is not an afterthought — it is the thing that makes the rest of the site worth reading.

How a page gets made

Each material page starts as research, not writing. Before a page is drafted, its production process, its statistics, and its technology trends are gathered from published sources and consolidated — one working record per material, with every source link kept alongside a note on what it does and does not cover. Only once the figures are gathered and cross-checked does the page get written around them.

That order matters. It means the prose is built to fit verified numbers, rather than numbers being found later to prop up prose already written. Where a figure could not be sourced to the standard below, the page says less rather than guessing.

Where the numbers come from

The statistical backbone of the site is the U.S. Geological Survey's Mineral Commodity Summaries, published annually and freely available in full. It covers most of the materials here on a consistent basis — production, reserves, trade and prices — which makes it the natural spine to hang each page on and the reference against which other figures are checked.

Around that spine, each material draws on the bodies that know it best:

• Steel — the World Steel Association (worldsteel), the OECD Steel Committee, and the IEA for decarbonisation.
• Aluminium — the International Aluminium Institute (IAI).
• Copper — the International Copper Study Group (ICSG) and the International Copper Association.
• Nickel — the Nickel Institute.
• Energy and emissions across all materials — the International Energy Agency (IEA), including its Global Critical Minerals Outlook.
• Cement, glass, lime and the rest — the relevant industry associations, national statistical agencies, and peer-reviewed process literature.

The principle is simple: prefer the original source over anything that summarises it. A production figure comes from the body that compiles it, not from a news article reporting that body. Where a page relies on a trade association — which has its own interests — the figure is treated as a claim from a named party, not as neutral fact, and checked against an independent number where one exists.

The rules every figure follows

These are the working rules behind the numbers on every page. They exist because each one, at some point, caught a mistake.

Global and regional are never mixed

This is the single biggest source of error in writing about industrial materials, and the one guarded against hardest. A process split, a list of top producers, or a per-tonne figure means nothing without saying where. Global steelmaking is roughly 70% blast furnace and 30% electric arc; in the European Union the split is close to even. Quoting one of those as the other is simply wrong, and the site is written to state the geography every time.

Estimates are labelled as estimates

Anything rounded, projected, or spanning a range of sources carries a small tag — approx or range — so it is never mistaken for a figure a source states outright. Anything untagged is a number a named source gives directly. Forward-looking figures from scenario models, such as the IEA's projections, are described as scenario results, not predictions, because that is what they are.

Derived numbers show their working

Some figures on the site are arithmetic rather than reported — an industry's value estimated from volume times price, say, or idle capacity from capacity minus production. Wherever that happens, the page says so derived and shows the inputs, so a reader can see exactly how the number was reached and check it against any official figure that measures the same thing a different way.

Sources sit next to the figure, not just in a footer

Attribution appears inline, beside the number it supports, as well as in the sources block at the foot of every page. This is partly for the reader, who should not have to hunt for where a striking figure came from, and partly a deliberate design choice for how the site is read by AI systems: a sentence that carries its own source and units survives being quoted out of context, which a bare number does not.

A wrong "fact" is flagged, not smuggled in

When a source, a draft, or a plausible-sounding claim turns out to contradict the verified figure — a trade tonnage that doesn't reconcile, a widely repeated number that traces back to nothing — the page uses the correct figure and, where it helps, notes the discrepancy rather than quietly papering over it.

The connections between materials

Most explanations of how things are made treat each material in isolation. Industry does not work that way, and neither does this site. The materials were researched not only individually but as a network — mapping which ones feed into which — and those dependencies are a deliberate thread running through the pages.

The clearest example is sulphuric acid, which turns out to be the busiest hub of the lot: it is an input to a large share of the other materials covered here. Copper and sulphuric acid form a genuine loop rather than a one-way chain — copper smelters make the acid from their own captured waste gas, and that acid is then used to leach copper from ore too poor to smelt. Those connections are mapped from the same sourced figures as everything else, and where a page claims one material feeds another, that link is one a source supports.

What to trust more, and less

No honest methodology claims every number is equally solid. Here is a fair guide to how much weight different figures on the site can bear.

If you find a figure on this site that you believe is wrong, or a source that has been superseded, that feedback is genuinely welcome — accuracy here is a standard to keep meeting, not a box already ticked.