Carbon accounting

Scope 3 is almost the whole footprint.

A fashion brand’s emissions are overwhelmingly not its own. They happen in mills and dye houses it does not own, in countries it may never have visited, months before a garment reaches a warehouse. Everything difficult about carbon accounting in this industry follows from that one fact.

The short version

Scope 1 is the emissions a company produces directly, Scope 2 is the energy it buys, and Scope 3 is everything else in its value chain: for an apparel brand, around 90% of the total, dominated by the purchased goods and services it sources from suppliers. Green Story builds Scope 3 up from real product footprints and traced orders rather than from a factor applied to spend, which is what makes the company footprint and the product footprints reconcile.

The frame

Only one of the three scopes is yours.

The GHG Protocol splits a company’s emissions into three scopes, and the split is about control rather than about size. Scope 1 is what you burn: gas in your own buildings, fuel in vehicles you own. Scope 2 is the energy you buy, principally electricity, counted where it was generated. Scope 3 is everything else in the value chain, upstream and downstream, from the cotton that became your fabric to the washing machine a customer will run for the next five years.

For a service business the first two scopes are most of the story. For a fashion brand they are a rounding error. An apparel company that leases its offices and outsources its manufacturing has almost no Scope 1 at all, and its Scope 2 is a few floors of lighting, while the fifteen categories of Scope 3 hold the mills, the dye houses, the freight, the shop floor and the customer’s laundry.

This is why a brand can decarbonise its own operations completely and move its total footprint by a percentage point. It is also why Scope 3 is where every regulator, retailer and target-setting body now looks first, and why the accounting question in this industry is really a supply chain question wearing different clothes.

  1. Scope 1

    What you burn

    Fuel in owned or controlled sources: boilers, company vehicles, refrigerant leaks. For most brands, small. For a vertically integrated manufacturer, substantial and directly actionable.

  2. Scope 2

    The energy you buy

    Purchased electricity, steam, heat and cooling. Reportable two ways (location-based on the local grid, market-based on what you contracted for) and both are usually required.

  3. Scope 3

    Everything else in the chain

    Fifteen categories, upstream and downstream. In apparel, category 1 (purchased goods and services) typically dwarfs the other fourteen put together.

Inside Scope 3

One category does most of the work.

Scope 3 has fifteen categories and a first inventory usually spends its effort in the wrong ones. Business travel is legible, countable and satisfying to reduce; it is also, for a clothing brand, almost never material. Purchased goods and services (the fibre, the yarn, the fabric, the wet processing, the make-up) is the category that decides the answer.

The consequence for how you spend your time is stark. Improving the precision of a category worth two per cent of the total cannot move the number, however good the data gets. Improving the precision of a category worth seventy can move it enormously and, more to the point, is the only thing that can tell you where a reduction is actually available.

Downstream matters too, and it is the half most brands leave out entirely. Use (washing, drying, ironing over a garment’s life) and end of life are genuinely large for clothing, and they are governed by what customers do rather than by procurement. They belong in the inventory with their assumptions stated, which is a different job from pretending they are not there.

The fork

Two ways to get the number.

There are two ways to get a Scope 3 number and the difference between them decides what you can do with it afterwards. The first takes what you spent with each supplier and multiplies it by an industry factor for that kind of spend. It is fast, it can be done from the finance system in a fortnight, and for a first-ever inventory it is a legitimate way to find out roughly where the mass sits.

Its problem is that it is a model of your purchasing, not of your production. Two suppliers charging different prices for the same garment get different footprints. A supplier who switches from coal-fired steam to a cleaner fuel gets exactly the same footprint as before. Negotiating a discount reduces your reported emissions. The number responds to commercial events and is deaf to physical ones, which is precisely backwards for a measurement whose job is to tell you whether anything improved.

The second builds the inventory up from the products you actually bought: the traced orders, the materials, the processes, the facilities. It is slower to stand up and it is the only version that can survive a reduction target, because it is the only one in which doing something real makes the number move.

The practical route is usually both, in sequence: a spend-based pass to find the material categories, then primary data where it matters, with the measured share rising season on season. What matters is not that estimation was used (it always is somewhere) but that the split is stated and the direction of travel is the right way.

The two bases for a Scope 3 inventory, and what each one can carry.

Spend-based

Supplier spend multiplied by an industry average factor. Buildable from the finance system in weeks, with no supplier engagement at all.

Use it for

A first screening inventory, to find which categories are material and where the effort should go next.

Built from product footprints

Traced orders, real materials and processes, facility-level energy where it can be obtained, with the rank of every input recorded on the result.

Use it for

A reduction target, a retailer’s data request, an assurance process, and any figure that has to move when the supply chain does.

Why it comes to a head

A target is what breaks a spend-based estimate.

An inventory on its own can survive being approximate. A target cannot, and this is where most brands discover the problem, usually a year or two after the first inventory, with a public commitment already made.

The mechanism is simple. A target requires a base year, and progress is measured against it. If your inventory is spend-based, then the arithmetic of progress is the arithmetic of your purchasing: buy the same garments more cheaply and you have hit your target; grow while decarbonising genuinely and you have missed it. Neither statement is about emissions.

It also fails the other way round. The reductions genuinely available in a textile chain are specific and physical: the fuel behind a dye house’s steam, the grid a spinning mill sits on, a wet process changed, a material substituted where it actually helps. None of those is visible in a spend-based model, so a brand working from one cannot tell which intervention is worth funding, and cannot show that one worked after paying for it.

The honest answer

What building it properly actually takes.

The uncomfortable part first: a product-based Scope 3 inventory depends on other companies. Your own effort can get you to a product list and a supplier list; everything after that is evidence arriving from facilities you do not own, which is why this work is measured in seasons rather than sprints and why the pace is set by collection rather than by calculation.

The order that works is to trace before you measure. A footprint needs to know which facility did the work, so the chain comes first, from purchase orders outward, with each partner confirming their own link. Then the product footprints, on ranked data with the measured share stated. Then the company inventory, built up from those products, reconciling with them by construction rather than by adjustment.

Where a supplier will not or cannot answer, the gap is recorded as a gap (graded critical, important or optional) and chased, rather than quietly filled with a default and presented as collected. An inventory whose weak points are marked is one you can improve deliberately. An inventory whose weak points are invisible is one you will defend by accident.

  1. First

    Trace the chain

    From purchase orders outward, each partner confirming their own link. Nothing downstream can be attributed properly until this exists.

  2. Then

    Footprint the products

    Per product, on ranked data, with the measured share on the face of the result rather than in an appendix.

  3. Then

    Build the inventory up

    Purchased goods and services assembled from the product library, so the company footprint and the product footprints cannot disagree.

  4. Then

    Set the target basis per scope

    Absolute or intensity, per scope, against a baseline built from primary data rather than from a spend estimate.

  5. Continuously

    Restate rather than overwrite

    Scope 3 is never finished. When better data replaces an estimate the figure moves, and every figure carries the basis it was computed on.

The words, gathered

Everything this page defined.

In the order you met them, in case you came back for one of them rather than for the whole page.

Scope 1
Direct emissions from sources a company owns or controls: its own boilers, vehicles and refrigerants.
Scope 2
Emissions from the energy a company buys. Reported location-based on the local grid and market-based on what was contracted for.
Scope 3
All other value-chain emissions, in fifteen categories. For an apparel brand, around 90% of the total.
Purchased goods and services
Scope 3 category 1: what you bought to make your product. In fashion it typically outweighs the other fourteen categories together.
Spend-based estimate
A footprint derived by multiplying supplier spend by an industry factor. Responds to prices rather than to production.
GHG Protocol
The accounting standard that defines the three scopes and the fifteen Scope 3 categories.
If you only wanted one answer

The shorter answers live on the FAQ.

Each of these is answered in a paragraph on the FAQ, which is where those answers live. This page is the long version of the same subject.

What an inventory can and cannot settle

A Scope 3 number is never finished and anyone who tells you otherwise is selling a certificate. Suppliers change, categories get re-scoped, and last year’s figure moves when better data replaces an estimate, so we restate rather than quietly overwrite, and every figure carries the basis it was computed on. We would rather hand you an inventory with its weak points marked than a tidier one you cannot defend, and we will tell you when a category is too thin to carry a target yet.

Bring us your own product

Watch the boundary get drawn on something you made.

Pick one garment and we will walk it end to end with you: the unit, the boundary, what we would need from your chain, and what the result would and would not let you say.