Illustrative person reviewing industrial-equipment drawings beside an equipment model and laptop
A synthetic equipment-review scene illustrates the records behind a capital acquisition, not a measured footprint.
Diagram text description

Hands and a pen point to equipment blueprints; a printed equipment image, a small industrial model, documents and a laptop sit on a bright desk. This illustration is not a real supplier assessment, certification or carbon result.

Scope 3 Category 2 covers the production emissions of capital goods purchased or acquired during the reporting year. For procurement and sustainability teams, the task is to identify those assets, assign their production to one category, recognise the amount in the acquisition year and choose a calculation method supported by the available evidence.

The most consequential mistake is often not arithmetic. A machine can be counted in both purchasing spend and asset additions, a whole-life footprint can be mistaken for a production factor, or a financial depreciation schedule can be used to spread emissions over time. Each can produce an orderly spreadsheet with the wrong meaning.

The route below connects the asset register to a reviewable estimate. It distinguishes asset production from operation, compares the four calculation families and shows what supplier evidence must establish. The aim is not to make every capital purchase look equally certain. It is to make the category, year, boundary and remaining uncertainty understandable for each material acquisition.

What belongs in Scope 3 Category 2?

A capital good is a long-lived product the company uses to make goods, provide services, or sell, store and deliver merchandise. The relevant emissions are those associated with producing that acquired asset, from its upstream inputs to the supplier's gate. They are not the energy the asset will consume throughout its working life.

Common purchases illustrate the distinction:

  • A production machine has a manufacturing footprint even before the buyer starts operating it.
  • A building or facility brings production-related inputs into the capital-goods assessment; its later electricity use is a different inventory question.
  • A purchased vehicle has upstream production emissions distinct from the fuel or electricity used on its journeys.
  • Servers or other equipment may be capital assets under the company's accounting procedures; their purchase records and operating electricity records answer different questions.

These examples identify possible assets, not an automatic classification rule. An expensive purchase is not necessarily a capital good, and a low purchase price does not establish an emissions exemption. The company's treatment of the purchase matters when the distinction is ambiguous. Start with what was acquired and how it is accounted for, rather than assigning a category from the supplier name.

For a bundled purchase, describe the asset and the components represented by its emissions estimate. If the evidence covers a complete machine, adding its already-included motor or steel again would overstate production. Conversely, a material-only estimate may leave manufacturing stages unrepresented. The boundary needs to describe the complete result, not just the asset's label.

Use the asset register to separate Category 2 from Category 1

Use the company's financial-accounting procedures to resolve an ambiguous purchase between Category 1 and Category 2. Category 1 covers purchased goods and services; Category 2 covers capital goods. The GHG Protocol does not ask the sustainability team to create a separate capitalisation test based only on price, expected lifespan or an emissions estimate.

An asset register is a useful starting point because it gives a purchase an identity and an accounting treatment. It is not a second emissions population to add wholesale to procurement spend. The same machine may appear in a purchase order, an invoice and a fixed-asset addition. Those records describe the transaction from different viewpoints; they do not represent three separate machines.

For an ambiguous equipment purchase, have finance explain the accounting treatment and document the resulting category decision. Carry that decision back into the purchased-goods dataset. A qualifying capital asset should not remain inside a broad Category 1 machinery-spend line while also receiving a Category 2 estimate. One category assignment must govern the same underlying production emissions.

Reclassifying a purchase from Category 1 to Category 2 changes the category totals while leaving their combined amount unchanged. Removing a duplicate entry is different: it lowers a previously overstated combined total. Neither correction shows that manufacturing became cleaner; explain it separately from a change in supplier production emissions.

Where an invoice includes different goods and services, identify the part represented by each calculation and the rationale for any split. Do not assume that one invoice must always equal one emissions line. The objective is a traceable assignment of the underlying activities, with no overlap and no unexplained remainder.

Why the whole amount is reported in the acquisition year

Financial depreciation allocates an asset's recorded value across accounting periods. Category 2 uses a different timing rule: recognise its total cradle-to-gate production emissions in the acquisition year. Do not spread that emissions amount over service years, discount it because the asset will last a long time, or copy a depreciation schedule into the greenhouse-gas inventory.

Consider a business acquiring a production line that it expects to use for several years. Finance can retain its applicable depreciation treatment for the asset value. The emissions inventory still records the line's production footprint in the year it is acquired. These are two ledgers with different purposes, not competing estimates of the same annual quantity.

Financial value can be allocated over asset life; Category 2 production emissions are recognised in the acquisition year.
Financial value can be allocated over asset life; Category 2 production emissions are recognised in the acquisition year.
Diagram text description

Two separate paths contrast financial value across asset life with production emissions in the acquisition year. The diagram assigns no numerical footprint and does not suggest that later asset operation is emission-free.

The timing illustration shows that distinction without assigning a numerical footprint. It does not say that the line has no emissions in subsequent years: fuel, electricity, maintenance-related purchases and other activities must still be assessed under their applicable boundaries. It says that the original asset-production amount is not repeatedly charged or amortised as Category 2 emissions.

A major acquisition can consequently make one year's Category 2 total much larger than the next. That pattern is not, by itself, evidence of deteriorating operational efficiency followed by a sudden improvement. Explain the investment and the acquisition-year convention when interpreting the trend. Keeping the rule consistent allows the reader to distinguish a non-recurring purchase from a change in manufacturing emissions intensity.

Keep asset production apart from operation and disposal

Define the Category 2 boundary around production of the acquired asset. Materials and manufacturing belong inside that cradle-to-gate assessment when represented by the selected method. Operation by the reporting company is outside this production amount. Fuel used in operating its equipment is generally a Scope 1 question; purchased electricity used for that equipment is a Scope 2 question, as applicable to the company's reporting boundary.

This matters when a supplier provides a whole-life study. A report covering production, decades of use and end-of-life treatment cannot automatically be entered as a Category 2 production factor. Identify the reported stages and extract or otherwise establish the relevant production result. If the report does not separate them sufficiently, retain the limitation rather than treating the whole-life total as cradle-to-gate.

The diagram encloses materials and manufacture while placing use and disposal outside. Its separation is an accounting boundary, not a claim that excluded activities have no environmental impact. A later scrappage event requires its own assessment of the relevant waste and inventory treatment. Do not place all future disposal emissions into Category 2 merely because the asset was originally purchased there.

Materials and manufacture sit inside the Category 2 production boundary; use and disposal require separate treatment.
Materials and manufacture sit inside the Category 2 production boundary; use and disposal require separate treatment.
Diagram text description

A Category 2 production boundary encloses materials and manufacture. Use, labelled Scope 1 or 2, and disposal, labelled assess separately, remain outside. The included stages connect; excluded activities are not shown as zero emissions or assigned automatically to a disposal category.

Leased equipment also needs a separate boundary decision. A lease payment is not automatically a purchase of the underlying capital asset, and the word “leased” does not by itself settle which operational scope or leased-asset category applies. Establish the arrangement and the reporting company's organisational boundary before using the appropriate guidance. Keep that unresolved decision visible instead of forcing every equipment-related invoice into the same capital-goods line.

Build a complete capital-purchase population

Before choosing an emission factor, establish the acquired-asset population. A precise factor applied to an incomplete or duplicated list is still the wrong total. Finance additions, procurement transactions and project schedules are complementary sources: each can reveal a missing identity or explain why apparent duplicates are actually parts of one acquisition.

  1. Start with the reporting-period fixed-asset additions and relevant accounting treatment. Retain identifiers, asset class and acquisition information, rather than exporting only a monetary total.
  2. Match those additions to purchase orders and invoices. Identify the physical asset or asset group, supplier, quantity and configuration that the transaction represents.
  3. Reconcile capital-project records. A building or production-line project may group several packages that appear separately in procurement; explain how packages connect to the final asset record.
  4. Resolve unmatched and repeated entries with the responsible finance or project owner. Distinguish a missing asset from a duplicate invoice, an excluded activity or a timing question; preserve the reason for the decision.
  5. Connect the resulting list to Category 1 screening so the same production is not retained in both populations. Keep the reconciliation between starting records and the final Category 2 list.

Finance, purchase and project records are reconciled into one asset population, not added as three independent totals.
Finance, purchase and project records are reconciled into one asset population, not added as three independent totals.
Diagram text description

Asset additions, purchase records and project schedules converge on one reconciled asset population. Several records may describe the same acquisition; the arrows mean matching identities and resolving differences, not summing all source rows.

The convergent figure represents that reconciliation, not a command to concatenate three spreadsheets. A single acquired asset may have many supporting records but should have one identifiable production-emissions treatment. It can be useful to retain a package hierarchy for a large project, provided child packages and the parent total are not both counted. This population work gives later uncertainty analysis a meaningful denominator.

Choose the Category 2 calculation method

Choose the calculation method from the evidence available for the assets, not from the apparent sophistication of its name. Category 2 uses the same broad calculation families as purchased goods and services. Their inputs differ, so a method change needs an explanation of what evidence improved and which uncertainty remains.

MethodStarting evidencePrincipal matching question
Supplier-specificProduct-level cradle-to-gate inventoryDoes the result represent this asset, quantity and production boundary?
HybridSupplier activity with secondary data for gapsAre allocated operations and remaining upstream stages complete without overlap?
Average-productPhysical amount and a secondary product factorDoes the mass, unit or asset specification match the factor?
Average spend-basedPurchase value and a secondary monetary factorDo the economic category, currency and price basis fit the spend?

Table 1, shown one row at a time. Every field is retained.

Row 1

Method
Supplier-specific
Starting evidence
Product-level cradle-to-gate inventory
Principal matching question
Does the result represent this asset, quantity and production boundary?

Row 2

Method
Hybrid
Starting evidence
Supplier activity with secondary data for gaps
Principal matching question
Are allocated operations and remaining upstream stages complete without overlap?

Row 3

Method
Average-product
Starting evidence
Physical amount and a secondary product factor
Principal matching question
Does the mass, unit or asset specification match the factor?

Row 4

Method
Average spend-based
Starting evidence
Purchase value and a secondary monetary factor
Principal matching question
Do the economic category, currency and price basis fit the spend?

A supplier-specific result is not automatically reliable because it is precise. Its declared unit, manufacturing scope and product configuration must match the acquisition. For a hybrid approach, distinguish the supplier's allocated operations from upstream materials, transport and other production inputs. Secondary data should fill identified gaps, not be added on top of stages already included in a supplier result.

An average-product estimate combines a physical quantity with a compatible emissions-per-unit factor. If a factor uses mass, a count of machines is insufficient until a justified mass basis is available. If the factor describes one complete product, multiplying again by its material mass can count the quantity twice. Write down what one factor unit actually represents.

Spend-based estimates are useful where physical or product-level information is limited, but price changes can affect the result without a corresponding physical change. A secondary monetary factor must be interpreted on its documented economic and price basis. Do not invent a universal currency conversion, inflation correction or precision claim. The method record should retain those assumptions so a later improvement can be compared honestly with the earlier estimate.

Request supplier evidence that matches the purchased asset

For a material purchase, ask the supplier for a product-level production result that can be connected to the asset actually acquired. A corporate emissions total describes the supplier's organisation, not automatically one machine, building component or vehicle. Dividing that total by an arbitrary product count does not create a supported product factor.

A useful request identifies the following:

  • Asset and configuration: model or product family, included components and whether optional equipment changes what the result represents.
  • Declared unit: one complete asset, a mass unit or another stated basis, with the purchased quantity needed to apply it.
  • Boundary: the production stages included and excluded; distinguish cradle-to-gate from a whole-life total or an isolated manufacturing stage.
  • Period and geography: the production data represented and whether its location and age are relevant to the acquisition.
  • Allocation: how shared operations or multi-product processes were assigned to the reported product; retain the explanation rather than assuming equal division.
  • Assurance and limitations: the document version, any verification or review scope, qualifications and unresolved gaps. A logo alone does not establish what was checked.

A product carbon footprint or environmental product declaration may supply useful evidence, but its title is not enough. Read the unit and the reported modules or stages before accepting the number. A declaration for a related product can be a proxy rather than an exact match; record that distinction and why it is considered suitable.

Suppose the buyer purchased a configured production line but the supplier supplies evidence for one standard machine within it. That document can support the represented component, not the whole line by default. The remaining equipment and production stages still need a method. Stating that gap is more useful than disguising a partial estimate as a complete supplier-specific footprint.

Explain unusual CapEx years and improve estimates

Explain a capital-investment spike by separating changes in the asset population from changes in the estimation basis. An acquisition-heavy year and a better factor can both affect the total, but they mean different things for the reader interpreting performance.

  1. Describe the exceptional acquisition and the year in which it enters Category 2. Keep the full production amount in that year rather than smoothing the inventory to make the trend look regular.
  2. State how the estimate was made. Identify material proxies or incomplete supplier evidence, and describe which part of the asset population they affect.
  3. Separate new evidence from changed activity. Replacing a broad spend estimate with matched product information changes the measurement basis; it does not alone prove a real reduction in production emissions.
  4. Use the improved evidence to inform the next comparable purchase. Compare alternatives delivering an equivalent function and make their production boundaries explicit before drawing a procurement conclusion.

For example, a price increase can raise a spend-based estimate even when the buyer purchases the same physical equipment. Conversely, a cheaper contract is not proof of a lower manufacturing footprint. This is a reason to investigate the estimate's basis, not to discard every spend-based result.

Retain the explanation alongside the reported figure. A future reader should be able to distinguish a different investment cycle, a revised dataset and a changed supplier product. None supports a universal reduction percentage, and the acquisition-year rule should not be replaced by a favourable-looking average.

Capital-goods evidence checklist for procurement teams

The evidence record should let a colleague follow one asset from purchase to category total without reconstructing the entire procurement exercise. Use the checklist as a handover test, not another list of documents to collect regardless of relevance.

  • Identity and population: asset or project identifier, supplier, specification, acquired quantity and reporting-period inclusion; explain packages or grouped assets.
  • Assignment and timing: finance treatment, Category 1/2 decision, acquisition-year rationale and the cross-check that prevents duplicate production entries.
  • Calculation basis: selected method, activity amount, factor value and unit, factor source/version and the production stages represented.
  • Source trail: the actual supplier inventory, declaration or secondary dataset, together with its date, applicability and material qualifications.
  • Accountability: preparer, reviewer, outstanding question, responsible owner and the effect of the unresolved issue on the result.

The reference needs to lead to the exact source used, not simply a supplier's homepage or a folder containing many versions. If the calculation changes, retain the earlier basis and the reason for the change. Otherwise a reviewer may see a new total without knowing whether it reflects corrected quantity, different boundaries or better information.

A blank field is informative when it names an unresolved issue. An unexplained blank is not. Distinguish “not applicable”, “not available” and “awaiting a decision”, and assess the consequence rather than treating all three as zero emissions.

Keep the evidence handover separate from accounting judgement

A shared evidence record can reduce the handover problem between procurement, finance and the inventory preparer. Its value is the connection between a named asset, the source used and the decision made about it. Storing more documents does not itself establish that the accounting treatment or the calculation is correct.

One practical handover has three stages:

  1. Procurement supplies the asset identity, configuration, acquired quantity and supplier evidence. Where the supplier response is partial, the record names what it covers instead of presenting it as a complete footprint.
  2. Finance resolves the relevant accounting classification and acquisition information. The inventory preparer then documents the category, calculation method, factor boundary and treatment of any gaps.
  3. A designated reviewer examines the connections and exceptions. Decisions remain attributable to people and evidence; an unanswered question has an owner and a defined effect on what can be reported.

This is a generic workflow, not a claim that a particular platform automatically classifies capital goods, calculates every Category 2 result or certifies an inventory. A source record can make review easier without replacing the judgement it records. Keep those responsibilities explicit when selecting or using any document-management arrangement. The immediate test is whether another person can retrieve the basis of one material asset estimate and understand the remaining limitation.

Conclusion: make every capital-goods figure traceable

A traceable Category 2 figure starts with the right acquired-asset population, not an attractive factor. Resolve ambiguous Category 1 versus Category 2 treatment through the company's accounting procedures, prevent dual entry and recognise production emissions in the acquisition year. Financial depreciation does not change that emissions timing rule.

Then match the estimate to the asset and the production boundary. A product-level source, a hybrid calculation, a physical proxy and a spend estimate can support different levels of understanding; none becomes complete simply because it returns a precise number. Keep material gaps visible.

For the next reporting cycle, choose one significant acquisition and follow its identity, timing, method and source through to the total. If another person cannot explain that chain, resolve the missing connection before extending the same approach to the rest of the register.