How much a sheet weighs, and why nobody agrees
The same A4 sheet has at least four published carbon footprints, and they do not match. Why that happens, why we count sheets instead of trees, and what cannot be claimed when you digitise.
4
published carbon footprints for the same A4 sheet, and none of them matches another.
On our sustainability page we publish one figure: 6.71 grams of CO₂e for every A4 sheet of 80 grams per square metre. It comes from a multiplication that is also written down —a sheet weighs 4.9896 grams and the paper production factor is 1,345.0779 kg CO₂e per tonne— and the source is the 2025 edition of the UK government’s emission factor database, materials-use table.
That figure covers production, from cradle to factory gate. It does not include transporting the paper to your warehouse.
What we want to tell here is what happens when you go looking for the same figure elsewhere.
Four numbers for the same object
A. C. Dias and L. Arroja did exactly that exercise and published it in the Journal of Cleaner Production in 2012: Comparison of methodologies for estimating the carbon footprint – case study of office paper, volume 24, pages 30 to 35. They took a sheet of office paper and calculated its footprint three times, with three different frameworks:
- 4.64 g CO₂e under the ISO 14040 and 14044 standards,
- 4.74 g CO₂e under the British specification PAS 2050,
- 4.29 g CO₂e under the framework of the Confederation of European Paper Industries.
Three standards, three numbers, the same object. The difference between the highest and the lowest is 10 %, and none of the three is wrong: each framework decides different things about what enters the count, how the burden is shared between products of the same process and what is done with the carbon the paper stores while it exists.
And why those three do not compare with ours either
This is where it pays not to cheat. Those three numbers are cradle to customer —they include distribution— and ours is cradle to factory gate, which is a shorter scope. Even so, ours is higher. That is not a contradiction: it is a different kind of paper, made in a different country, with a different electricity mix, and measured fourteen years earlier.
In other words: the factor per sheet is not a constant of nature. It depends on the standard, the scope, the grammage, where it was made and when it was measured. Any of the four numbers is defensible if it comes with those five things, and none of them is if it comes alone.
That is why our page states the factor, its edition, its table and what it covers. Not to show off: so you can swap it for yours if you have a better one, and redo the sum.
AI-generated image from our own references.
That is why we count sheets and not trees
The question that always comes is how many trees were saved. We do not answer it, and the reason is the same as above: turning sheets into trees requires assuming species, plantation yield and share of recycled fibre. We have none of the three, and your site’s paper supplier does not give them to you either.
Counting sheets, on the other hand, anybody can do: they are the records that used to be printed. The number that comes out is small and verifiable, and we prefer that.
What the replacement costs
Digitising is not free, and a site that does not say so is selling half the sum.
The International Energy Agency published the report Energy and AI in 2025, and there is the figure from the other side: data centres consumed 415 TWh in 2024, around 1.5 % of the world’s electricity, and the projection is that this figure more than doubles to some 945 TWh by 2030, pushed above all by artificial intelligence.
We do not publish what share of that corresponds to PAIR. We have not measured it, and an invented figure about our own consumption would be exactly the vice this note is criticising.
What we cannot say
Three things, and they are on the page too:
- We do not say PAIR cuts your emissions. It cuts printed sheets and, if you use the navigable tour, office trips. What that means in your inventory is decided by your inventory, not by us.
- We do not publish an aggregate figure of what the product has avoided. It would mean adding up sites we have not measured.
- Paper is the small figure. With our own factors, one kilometre in a site pickup weighs the same as 29.8 sheets. If somebody sells you «paperless» software as a climate initiative, they are selling you the small number.
The full arithmetic, with its factors, its sources and its assumptions written down, is under sustainability.