In late 2022 a procurement manager at a food company asked us a reasonable question: what is the carbon figure behind the forty containers we just bought from you? And we gave her an answer that, written down, amounted to considerably less than new.
She was polite about it. We were not comfortable about it. We had been making an environmental argument for a decade and, when asked to put a number on it, we had a direction rather than a figure.
What we did about it
We spent the next several months doing three unglamorous things: weighing containers, finding defensible published coefficients, and writing down the assumptions we were making so that somebody could disagree with them.
1. Weighing
We stripped and weighed forty containers across four manufacturers. A 275-gallon composite IBC averages 18.1 kg of HDPE in the bottle and 25.0 kg of steel across cage and base, plus fittings. Those are our own measurements from our own intake, not a datasheet, and they are the only inputs in the whole model that are genuinely ours.
2. Coefficients
| Coefficient | Value | Applies to | Basis |
|---|---|---|---|
| HDPE carbon intensity | 1.9 kg CO₂e/kg | Bottle | Cradle-to-gate average, virgin resin |
| Tube steel carbon intensity | 1.9 kg CO₂e/kg | Cage, base | Blended BOF/EAF, galvanised tube |
| Polymer primary energy | 76 MJ/kg | Bottle | Polymerisation to moulding |
| Process water | 24 L/kg polymer | Bottle | Plus steel finishing allowance |
| Oil-equivalent feedstock | 1.75 kg/kg HDPE | Bottle | Naphtha feedstock plus process fuel |
| Landfill airspace | 1.4 yd³/container | Whole unit | Uncrushed container volume |
3. Route factors
This is where the judgement lives. Reuse is credited at 1.00 because it displaces a whole new container. Rebottling at 0.62 because it retains the cage and base. Granulation at 0.34, a material credit only, because it displaces virgin resin at a discount and displaces no container at all.
Those three numbers are arguable and we would rather argue about them in public than have them buried in a spreadsheet.
Why publish the coefficients
Because a green number without a method is decoration, and because the people who most need this figure — sustainability teams, procurement, assurance providers — cannot use a number they are asked to trust.
A number you can interrogate is worth more than a number you have to trust. Publishing the method costs us the ability to flatter ourselves, which is exactly why it is worth doing.— From the methodology note we send with every report
It also means that if you think our steel factor is wrong — and it is the weakest link, because BOF and EAF production differ by roughly a factor of three and we do not know which mill made any given cage — you can substitute your own and redo the arithmetic. Several customers have. Two came back with lower figures than ours and we adopted their reasoning for their reports.
What the tool deliberately does not do
- It does not produce an audited disclosure. It is a screening tool and it says so on the page.
- It does not claim removals. Every output is framed as avoided relative to a new-container baseline, which is the only framing that survives assurance.
- It does not include the landfill emissions of the container that was displaced, even though a more generous model would. That is a deliberately conservative omission.
- It does not sell offsets, and it never will. Avoided emissions inside your own inventory boundary are not tradeable and anyone telling you otherwise is selling something.
How it actually gets used
Not the way we expected. We built it imagining sustainability teams. In practice the heaviest use is procurement: buyers print the receipt and staple it to a purchase memo to justify specifying reconditioned containers over new ones to somebody who has never thought about it.
The second-heaviest use is internal persuasion — an operations manager showing a plant manager what the current disposal habit costs in something other than dollars. It turns out a printed receipt is remarkably effective in a meeting.
The procurement manager who asked the original question now gets a line-item calculation with her serial numbers on it, quarterly. It took us four months and a certain amount of wounded pride to get there, and it was the most useful thing we built that year.