The inventory runs across Scope 1, Scope 2 and the parts of Scope 3 that genuinely move the total. Combustion on your own sites and the electricity you buy are the straightforward half of it. The rest is a question of how far down the supply chain the data can be made to reach before the cost of chasing it stops being worth the precision it buys, and we put that line in writing before any counting starts.
Every figure carries its emission factor, the source it came from and the date that source was published. Where a number is modelled rather than metered it is marked as modelled, and the assumption behind it is written down in language a non-specialist can argue with. That log is handed over with the inventory, because the first question an assurer asks is which part of the total you actually measured.
Boundary questions are settled first and written down, because moving them halfway through invalidates everything counted up to that point. Leased assets, joint ventures and franchises each have a defensible treatment under the GHG Protocol, and more than one of them may be defensible for you. We choose one, record why it was chosen, and hold to it for the rest of the engagement.
If a figure cannot survive somebody reading the working behind it, it is not a figure worth publishing in the first place.
Four to six weeks from the first boundary conversation, you receive the inventory, the factor register, the assumption log and the model itself. The model comes in a format that opens without a licence we hold, because a figure you cannot rebuild yourself next year is a figure you are renting rather than one you own.
The figures also arrive as a short statement your board can read without opening the spreadsheet, with the caveats set in the body of it rather than in a footnote.
What you receive
Every figure arrives like this one: the factor it used, where that factor came from, and whether the number was measured or modelled.
Metered
Read off a meter, not estimated.
Factor
The published source, with its date.
Uncertainty
How far the figure could move.
Inventory · line 014
Specimen
Scope 2 · purchased electricity
Period
1 Jan – 31 Dec 2025
12 months, no gaps
Site
Dunmore plant
IE-CRK-02 · meter A, half-hourly
1,284,000
kWh
Metered
×
0.207
kgCO₂e / kWh
DESNZ 2025 · 26.06.2025
=
265.8
tCO₂e
Reported
Method
Location-based
Assumption
None
Uncertainty
±1.2%
Prepared by R. Oakes · checked 14.01.2026
At the end of the engagement these files move to you, in formats you can open without us, with every figure traceable to the line it came from.
Turnaround
4–6 weeks
from the first boundary conversation
Files handed over
4
openable formats, not a portal login
Signed by
1
named analyst, still here next year
Document
What it answers
Share of the file
Pages
Status
Full inventory
What the total is, line by line
48
Handed over
Factor register
Where every factor came from
26
Handed over
Assumption log
What was modelled, and why
14
Handed over
The model itself
So you can re-run it next year
—
On request
Page counts are from the last engagement of this size. If a figure in any of them cannot be traced back to the line it came from, it does not leave the building.
