For most companies, Scope 1 and Scope 2 are tractable. You have meter readings and fuel invoices, the boundaries are clear, and the arithmetic is straightforward.
Scope 3 is a different exercise. It typically accounts for the large majority of a company’s footprint, the data belongs to other people, and the fifteen categories in the GHG Protocol range from reasonably measurable to genuinely difficult.
The result is that companies either avoid it or attempt everything at once and produce something they cannot defend. There is a better route.
Do not start by measuring
Start by screening. Work through all fifteen categories and make a rough assessment of which are likely to be significant for your business. This can be done at a coarse level with existing financial data and a few assumptions.
For most companies, the answer clusters quickly. A manufacturer’s footprint is dominated by purchased goods and services and by upstream transport. A professional services firm’s is dominated by purchased services, business travel and employee commuting. A retailer’s often sits in purchased goods and downstream use.
Three or four categories usually account for the overwhelming majority. Screening tells you which, and everything after that is proportionate effort.
Then use spend
For a first inventory, spend-based estimation is a reasonable place to start. Take your purchase ledger, map spend to categories, and apply average emissions factors.
It is imprecise. It is also defensible as a starting point, it uses data you already have, and it produces a number within weeks rather than quarters. Most importantly it tells you where the concentration is, which is what determines where to invest in better data.
The known weakness is that spend-based methods make a cheaper supplier look lower-carbon than an expensive one for the same physical output. This is why you move off spend data for material categories once you know which they are.
Then improve where it matters
For the categories that dominate, move towards primary data.
Supplier-specific data for purchased goods and services, starting with your largest suppliers by spend. Many will already have a footprint; those that do not may be willing to work on one if a significant customer asks.
Activity data rather than spend for transport and travel: distances, modes, weights, rather than invoice values.
Better assumptions for commuting and homeworking, ideally from a staff survey rather than a national average.
You will not get all of this in year one, and you should not try. Improving one category a year, starting with the largest, is a credible programme. Attempting all fifteen at once produces a spreadsheet nobody trusts.
Document as you go
This is the part that determines whether the work is usable later.
For every category: the method used, the data source, the emissions factors and their version, the assumptions made, and any exclusions with the reason. Write it as you do it, not afterwards.
If your inventory is ever assured, this documentation is what the practitioner examines. If it does not exist, the number cannot be assured however carefully it was calculated.
Expect the number to move
Your Scope 3 figure will change as your method improves, sometimes substantially and sometimes upwards. This is normal and should be explained rather than hidden.
The honest framing is that early estimates are directional and will be refined. That is a much better position than restating a number without explanation, or quietly changing method and hoping nobody compares.
Disclose the method, disclose the changes, and restate comparatives where the change is material.
What good looks like in year one
A screening assessment covering all fifteen categories. Measurement of the three or four that matter, on a spend basis. Documented methods and exclusions. A named owner. A plan for which category gets better data next year.
That is a credible first inventory. It is achievable in a few months, and it is a substantially stronger position than a partial number produced without a method.