How to measure Scope 3 when supplier data is missing

When supplier data is unavailable, measuring Scope 3 emissions requires a structured, tiered approach to estimate emissions accurately and transparently. Here's the core strategy:

1. Start with Spend-Based Estimates: Use procurement spend data and multiply it by economy-wide emission factors. This is a practical starting point but less precise.
2. Progress to Activity-Based Calculations: Use measurable data like weight, distance, or energy usage combined with emission factors for better accuracy.
3. Aim for Supplier-Specific Data: Collaborate with suppliers to obtain verified emissions data for the most accurate results.

Key Tips:

• Focus on high-impact suppliers (top 20% by spend often account for 60–80% of emissions).
• Label every data point by quality (e.g., spend-based, activity-based, or supplier-specific).
• Use AI tools to automate data extraction, improve accuracy, and maintain traceability.

Even with limited supplier data, this method ensures complete value chain coverage while improving data quality over time.

The Scope 3 Data Hierarchy: How to Work Through Supplier Gaps

Building on the three-tier approach mentioned earlier, each tier represents a different level of data quality. Choosing the right tier for each inventory line is crucial to ensure your estimates are reliable and defensible.

Supplier-Specific Data: The Ideal Option

Supplier-specific data is the preferred choice. It relies on verified supplier emissions data from a Product Carbon Footprint (PCF) that adheres to ISO 14067 or the GHG Protocol Product Standard. This method is considered the most precise because it reflects the supplier's actual operations, including their energy sources, production methods, and efficiencies, rather than relying on general industry averages.

For this data to hold up during due diligence, the PCF must clearly outline its system boundary (what is included and excluded), specify how emissions are allocated across multiple products, and comply with ISO 14064-1 verification standards. Labelling each inventory line with its data quality tier is also critical. This transparency ensures that estimates remain traceable and comparable, making it easier to aggregate data across suppliers or portfolios.

Activity-Based Calculations: A Reliable Alternative

If supplier-specific data isn’t available, the next best method involves activity-based calculations. This approach uses measurable physical data (e.g., kilograms, kilowatt-hours, or kilometres) and multiplies it by an emission factor specific to the material or activity. By linking emissions to tangible quantities, this method eliminates price distortions and provides results that are consistent and reproducible.

As Jeremiah Say, Lead Systems Architect at GreenCalculus, explains:

"An activity-based emission calculation is a closed physical statement... the result is reproducible to the decimal." [4]

The Partnership for Carbon Accounting Financials (PCAF) Standard assigns activity-based methods a data-quality score of 2–3, which is significantly better than the 4–5 score given to spend-based methods [4]. If you use activity-based data for a specific category, ensure you remove the corresponding financial spend from any spend-based calculations to avoid double-counting emissions.

When physical data isn’t accessible, spend-based methods can act as a fallback.

Spend-Based and Proxy Methods: The Last Resort

Spend-based estimation involves using financial procurement data and multiplying it by an economy-wide emission intensity factor from databases like EXIOBASE 3 or the US EPA's USEEIO. This method is quick and scalable, making it an efficient starting point, especially when dealing with thousands of line items. However, it comes with greater uncertainty compared to activity-based methods. For example, in a case study involving £4.2 million spent on steel, results from four major databases varied by a factor of 1.48x [4].

If you use spend-based factors, it’s essential to adjust nominal spend to the factor’s base price year. Failing to do so could lead to inflated emissions estimates [4].

Summary of Methods

Here’s a quick comparison of the three tiers to help you decide which approach fits your needs:

This tiered system can be combined into a hybrid model, where you use spend-based data for initial screening and refine it over time with activity-based or supplier-specific data. The next section will explore how to integrate these methods into a comprehensive Scope 3 inventory.

How to Build a Scope 3 Inventory Without Supplier Data

If you're missing supplier data, you can still build a detailed Scope 3 inventory by following a structured approach. Here's how.

Screening Categories for Materiality and Data Gaps

Begin by mapping your value chain across all 15 Scope 3 categories outlined in the GHG Protocol. The goal here is to cover as much ground as possible. Use 12–24 months of spend data to identify where most emissions are likely concentrated. Focus on areas contributing more than 1% of your estimated emissions and apply the Pareto principle: typically, the top 20% of suppliers or spend categories account for 60–80% of procurement-related emissions [1]. This will help you zero in on areas where better data collection efforts are needed.

Once you've pinpointed the material categories, assign the best available data tier to each inventory line item.

Applying the Three-Tier Method to Each Line Item

After identifying the key categories, use the three-tier hierarchy to assign calculation methods to each line item. Always aim to use the highest-quality data available and only fall back to lower-tier methods when necessary.

For major suppliers and high-spend categories like Category 1 (Purchased Goods and Services) and Category 4 (Upstream Transportation), prioritise activity-based or supplier-specific data. Physical activity information - such as tonnes of material or tonne-kilometres of freight - is often more accurate and accessible than spend-based estimates. For example, Hitachi Rail increased its Scope 3 inventory coverage from 10–13% to over 90% by systematically working with suppliers to replace spend-based estimates with more precise data [1].

For smaller suppliers, where collecting physical data may not be feasible, spend-based calculations can fill in the gaps. However, when you replace a spend-based estimate with activity-based or supplier-specific data, make sure to remove the corresponding financial spend from your inventory to avoid double-counting.

Labelling Each Line by Data Quality

Every line in your inventory should include a label that clearly explains how the figure was calculated. This isn’t just good practice - it’s increasingly required by regulations. For instance, under CSRD (ESRS E1), companies must disclose the percentage of their Scope 3 inventory that comes from primary versus secondary data [1].

Use a transparent three-tier system to label your data:

Document the metadata for each line - such as the database used (e.g., EXIOBASE 3.8.2 or USEEIO v1.3), the factor reference year, and the currency base. As Jeremiah Say, Lead Systems Architect at GreenCalculus, recommends:

"Fix the database, the reference year, and the currency base at baseline, document all three, and treat any change to them as a Major recalculation... never as a year-on-year improvement." [4]

If any Tier C line represents more than 5% of your total inventory, flag it as a priority for improvement in the next reporting cycle. This ensures that less reliable estimates remain visible and creates a clear roadmap for enhancing data quality over time.

How AI Can Support Scope 3 Analysis in Due Diligence

Manually building a Scope 3 inventory is a slow, error-ridden process. AI-powered tools simplify this by automating repetitive tasks, making analysis faster and more thorough.

Extracting Relevant Data from Unstructured Documents

Supplier documents like invoices, purchase orders, and contracts often lack standard formats for carbon accounting, leaving valuable data buried in inconsistent layouts. AI tools can extract and standardise this data at scale, mapping internal procurement categories to established sector classifications (such as NAICS or CPA) using fuzzy matching techniques [6][2]. This is crucial because auditors view large expenses categorised under vague labels like "miscellaneous" as a warning sign.

Two common technical issues can distort spend-based calculations: price-basis mismatches, which can cause 15–35% inaccuracies, and vintage drift, leading to 20–30% errors over several years. AI tools resolve these by applying the correct deflators and price adjustments automatically. By efficiently extracting and standardising data, AI sets the foundation for audits that are both transparent and traceable.

Building a Clear Audit Trail

Every figure in a Scope 3 inventory must be traceable back to its source. Regulators and third-party verifiers require clarity on where each number originates - whether it’s a database, reference year, or currency base. AI platforms ensure this by creating cell-by-cell provenance, linking emission estimates to their specific source documents, emission factors, and calculation methods [4].

Jeremiah Say, Lead Systems Architect at GreenCalculus, explains:

"The same procurement spend, run through different MRIO databases, produces emission totals that diverge by tens of percent - and the divergence has nothing to do with anything the company did." [4]

To address this, AI tools lock key variables like the database, reference year, and currency base, while documenting every step of the normalisation process. This metadata is embedded into export files (e.g., JSON or CSV), making it easier to meet CSRD and SBTi verification requirements without having to rebuild the audit trail annually [6][2]. This level of traceability also highlights suppliers in need of data improvements, gradually increasing accuracy.

Transitioning from Proxies to Primary Data

Spend-based estimates are just the beginning. The next step is identifying which suppliers to prioritise for collecting primary data. AI tools help by flagging "upgrade candidates" - suppliers responsible for over 5% of total emissions but currently assessed using lower-quality spend-based proxies [6]. By identifying these key suppliers, AI ensures that gaps in the inventory are addressed systematically, improving data quality over time.

Axion Lab's data flywheel effect is particularly useful in private equity. Each deal enriches the system with sector benchmarks, supplier data, and emission factor mappings. Over time, spend-based estimates for common suppliers are replaced by more precise activity-based or supplier-specific data. This iterative improvement compounds across a portfolio, sharpening accuracy with each reporting cycle.

For private equity firms managing diverse portfolios, this approach - starting broad and refining where it counts - is the most practical way to achieve reliable Scope 3 data at scale.

sbb-itb-6ca8558

Using Scope 3 Estimates to Support Investment Decisions

A well-prepared Scope 3 inventory, even if initially created using spend-based estimates, can guide investment decisions beyond mere compliance. The key lies in presenting uncertainties and data quality with transparency.

Presenting Uncertainty and Data Quality Clearly

Investment committees need to understand the broader context behind the numbers. Using the three-tier data hierarchy helps label the uncertainty tied to each estimate. For example, spend-based estimates typically have an uncertainty range of ±40–60% [10], while activity-based figures are more precise, with ranges closer to ±10–30% [10]. Adding a PCAF scoring system label to each line item allows for quick assessment of data confidence. A score of 1 represents high-quality supplier-specific data, while scores of 4–5 indicate spend-based estimates [4][2].

Interestingly, upgrading from spend-based to primary data often leads to a reported emissions increase of up to 30% [1]. This should be seen as a step toward greater transparency rather than a sign of declining performance.

Identifying Where to Improve Data and Cut Emissions

Using data quality labels, organisations can pinpoint areas for improvement. Typically, the top 20% of suppliers by spend are responsible for 60–80% of procurement-related emissions [1][8]. This group should be the primary focus for collecting higher-quality data and reducing emissions. Targeting these high-impact suppliers ensures resources are used efficiently and leads to meaningful decarbonisation opportunities.

A well-constructed inventory can uncover actionable steps to reduce emissions - such as shifting freight from road to rail, consolidating deliveries, or redesigning products. These opportunities often remain hidden when relying solely on industry averages [9][3].

Linking Scope 3 Data to SFDR, CSRD, and Value Creation

With clear data and targeted improvements, a strong Scope 3 inventory supports both compliance and business growth. The CSRD (ESRS E1) requires companies to disclose gross Scope 3 emissions by category, including the split between primary and secondary data [7]:

"A single well-structured, GHG Protocol-compliant scope 3 inventory can serve both [CSRD and SBTi] purposes." - Paul Ferreira, Associate Climate Strategy Advisor, Normative [7]

For private equity, the implications are direct. Companies with unmanaged Scope 3 risks face transition challenges that could reprice 5–15% of their enterprise value [10]. Improving data quality is, therefore, not just about meeting reporting standards - it’s a critical measure for protecting value.

The table below summarises the key requirements for using spend-based estimates across major frameworks:

Aligning the CSRD reporting year with the SBTi base year from the start is a practical move. It avoids the need for later restatements and saves significant time as regulations continue to evolve [7].

Conclusion: A Practical Path to Scope 3 Clarity

Lack of supplier data shouldn't hold organisations back from starting their Scope 3 measurements. A clear, actionable approach exists: begin with spend-based estimates, progress to activity-based calculations, and aim for supplier-specific primary data. As Josh Prigge, a Carbon Accounting Expert, wisely notes:

"The goal of Scope 3 accounting should not be perfection in year one. The goal should be continuous improvement." [11]

Starting with spend-based data is practical and ensures complete value chain coverage right away, utilising financial data that most organisations already possess [4][2]. From there, focusing on the top 20% of suppliers by spend - who typically account for 60–80% of procurement-related emissions [1] - allows for the most impactful shift to activity-based or supplier-specific data, significantly improving accuracy.

A key element of this approach is data quality labelling. Assigning a PCAF score to each data point ensures transparency, clearly distinguishing between estimates and verified figures. This not only satisfies auditors, investors, and regulators but also avoids misinterpreting methodological upgrades as real emissions reductions. For instance, when Hitachi Rail partnered with Normative to engage its suppliers, they boosted their Scope 3 coverage from just 10–13% of actual emissions to over 90% by 2025 [1]. Such progress is only credible with a well-documented data quality journey.

AI-powered tools add further practicality by automating the mapping of procurement categories to industry-average emission factors, identifying high-impact areas for improvement, and maintaining a detailed audit trail - including adjustments for inflation and currency changes [6].

With the proposed GHG Protocol Revision B1 (March 2026) mandating a 95% Scope 3 coverage threshold [5], the direction is clear. Organisations that establish structured, labelled inventories now - even if primarily spend-based - will be far better prepared for tightening disclosure requirements than those waiting for perfect data that might never materialise.