Scope 3 Category 1, purchased goods and services, is typically the largest single line item in a manufacturing or product company's emissions inventory. It is also the category where most companies' numbers are wrong, not because they made arithmetic errors, but because the estimation method they used generates a number that does not reflect their actual supply chain emissions intensity.
This matters for CSRD compliance and increasingly for SEC filings where Scope 3 is material. A category figure that is off by a factor of two does not just affect your total emissions number. It affects reduction targets, supplier engagement priorities, and the narrative you present to auditors about your value chain.
Why Spend-Based Is a Starting Point, Not an Answer
The spend-based approach applies an emissions intensity factor (typically expressed as kgCO2e per dollar of spend) to your procurement expenditure by category. The methodology is legitimate and explicitly described in the GHG Protocol Corporate Value Chain (Scope 3) Standard as an appropriate method when supplier-specific data is not available. Used correctly, it provides a reasonable order-of-magnitude estimate for your first inventory.
The problem is that spend-based factors are industry averages. They represent the average emissions intensity per dollar across all companies producing a given category of goods or services in a given economy. If your specific suppliers operate at a meaningfully different emissions intensity than the industry average, the spend-based estimate is wrong, in a direction and magnitude you cannot determine without supplier-specific data.
Several common scenarios produce large deviations from the spend-based estimate. A manufacturer sourcing steel from a facility that uses electric arc furnace (EAF) production rather than basic oxygen furnace (BOF) will have a Category 1 figure for steel that is 60-75% lower per tonne than the industry average suggests. A company sourcing paper from suppliers certified to sustainable forestry standards operating integrated pulp and paper mills may have 20-40% lower Category 1 intensity than the sector average for paper products. Conversely, a company sourcing concrete from aging facilities in energy-intensive regions may have higher-than-average intensity per dollar.
Spend-based methods cannot capture any of this variation. They report the same tCO2e per dollar regardless of which specific supplier you buy from.
When Spend-Based Breaks Down Most Severely
There are three specific situations where the spend-based approach produces figures that will be challenged in an assurance engagement or by an investor asking about emissions intensity.
High-carbon commodities as a significant spend category. If steel, aluminum, cement, or chemicals represent more than 15% of your total procurement spend, the emissions intensity variance within those categories is large enough that the spend-based figure may be off by 30-80% from your actual supply chain emissions. These are the categories where moving to supplier-specific data or at least material/process-level activity data has the most impact on accuracy.
Significant spend on professional services or software. Spend-based factors for professional services, IT services, and software tend to be low (services are less energy-intensive than goods). But if your services spend includes high-compute cloud services or on-premise data center costs, the emissions intensity may be higher than general professional services factors suggest. The spend-based factor does not distinguish between a law firm and a data center operator in the same "professional services" category.
Multi-tier supply chain dependencies. Spend-based Category 1 captures only Tier 1 suppliers: the companies you pay directly. It does not capture Tier 2 or deeper emissions unless you use a sector-specific economic input-output factor that incorporates upstream supply chain effects. For a company whose Tier 1 suppliers are predominantly light-assembly operations, the significant emissions may be concentrated at Tier 2 or Tier 3 (raw material extraction, intermediate processing). Spend-based Category 1 applied only to Tier 1 spend will materially understate the true upstream impact.
A Practical Framework for Moving Toward Supplier-Specific Data
We are not suggesting every company needs supplier-specific data for every Category 1 supplier. That is operationally impractical for most supply chains and disproportionate to the risk for low-spend, low-intensity categories. The GHG Protocol explicitly endorses hybrid approaches: use supplier-specific data for high-materiality suppliers, spend-based for the long tail.
The framework we recommend has three tiers based on emissions materiality:
Tier A: Top suppliers by estimated spend-based emissions. Run your full spend list through spend-based factors to get an initial rank-order by estimated tCO2e. Your top 10-15 suppliers by this ranking will typically account for 60-75% of Category 1 emissions. These are the suppliers worth engaging for supplier-specific data: either their Scope 1 and 2 emissions per unit of product sold to you (product-level emissions factors), or at minimum their total Scope 1 and 2 emissions allocated to your spend share.
Tier B: High-intensity categories. For categories with above-average spend-based intensity (steel, cement, chemicals, electronics), even mid-tier suppliers warrant at least activity-based estimation rather than pure spend-based. Activity-based means applying a process-level factor (tCO2e per tonne of steel, per cubic meter of concrete) to quantity data from your procurement records rather than spend data. This eliminates the price-inflation distortion in spend-based methods.
Tier C: Long-tail suppliers. Apply spend-based factors with documented uncertainty ranges. Most auditors accept this for the long tail as long as you clearly flag the estimation method and the uncertainty band.
The Supplier Data Request Problem
Getting emissions data from suppliers is an engagement challenge, not a data collection challenge. Most small and mid-size suppliers do not have a GHG inventory. Asking them to provide one is asking them to do work they have never done before. Response rates for undirected emissions surveys are typically low unless the buyer is large enough to make the request feel non-optional.
The practical approaches that work better than mass surveys:
- Target only your Tier A suppliers (top 10-15 by emissions materiality) with a specific, bounded data request. Ask for total Scope 1 and 2 emissions for the most recent reporting year and total revenue, so you can calculate an emissions intensity per dollar. This is a much smaller ask than a full GHG inventory.
- Reference CDP's supplier engagement program if your suppliers are CDP-participating. Many mid-size industrial suppliers now report to CDP, and CDP data is accessible for registered companies.
- Use your purchasing leverage: incorporate emissions disclosure into supplier qualification criteria going forward, not as a punitive measure but as a signal that you are building toward activity-based reporting.
For suppliers who do not respond, the GHG Protocol allows use of spend-based or average-data factors as a fallback, as long as you document the data gap and note the estimation approach in your disclosure. A disclosed figure with a clear data quality note ("supplier-specific data not available; spend-based estimate applied using EXIOBASE 2023 US manufacturing factor") is far more defensible than an undocumented aggregate number.
How Emitpulse Handles Category 1
In Emitpulse, the Category 1 workflow starts with a spend export from the connected ERP. Our classification engine maps each invoice line to an NAICS or ISIC activity code, applies the corresponding spend-based factor, and produces an initial emissions estimate. The ledger flags each row with a method indicator: spend-based (S), activity-based (A), or supplier-specific (SS).
When a supplier response comes in (via our structured supplier data form or a manual upload), we replace the spend-based estimate for that supplier with the supplier-specific calculation, re-flag the method indicator, and update the category total. The data quality dashboard shows the percentage of Category 1 emissions covered by supplier-specific vs. spend-based data, which is a standard disclosure metric in ESRS E1 and CDP reporting.
The audit trail records both versions: the original spend-based figure and the updated supplier-specific figure with the supplier response date and data source. This is important because your auditor may want to see the transition from estimate to primary data, not just the final number.
The Accuracy vs. Completeness Tradeoff
We want to name a tension that comes up consistently in Category 1 discussions: the tradeoff between accuracy and completeness. A company that collects supplier-specific data from its top 12 suppliers and spend-based data for the remaining 300 will have a more accurate Category 1 figure for the top 12, but may appear to have a lower-quality inventory overall because the long tail is still spend-based.
This is not a reason to avoid supplier-specific data collection. It is a reason to frame the improvement correctly in your disclosure narrative. A disclosure that says "Category 1 covers 73% of spend; supplier-specific data covers 11 suppliers representing 62% of estimated Category 1 emissions; remaining 38% estimated using EXIOBASE spend-based factors" is more credible than a disclosure that says "spend-based method applied to full Category 1 spend." The former shows that you understand your supply chain and are making deliberate data quality investments. The latter is a defensible baseline but tells auditors and investors less about the trajectory of your data program.