Prioritising Suppliers: Carbon, Cost and Risk

A useful supplier-prioritisation model answers two different questions: where is action most urgent, and where can the buyer achieve progress now? Start with a carbon screen of the whole supply base, then score emissions, commercial exposure and due-diligence risk separately from leverage, data readiness and timing. The result should not be one league table. It should be a set of supplier segments linked to specific actions: validate data, launch reduction projects, build capability, change contract terms or monitor. This directs scarce procurement and sustainability resources to decisions that can alter Scope 3 emissions and supply risk.

Why spend alone produces the wrong priority list

Spend is easy to extract from an enterprise resource planning system, which makes it a practical starting point. It can indicate purchasing influence and help teams cover a large share of their supplier base quickly. It does not reveal how much a supplier emits, whether a material is energy-intensive or whether the buyer faces a concentrated operational risk.

The GHG Protocol Scope 3 Standard 1 is designed to identify material value-chain emissions and reduction opportunities, not to rank companies from accounting totals alone. Its calculation guidance 2 recognises supplier-specific, hybrid, average-data and spend-based methods for purchased goods and services. These methods serve different purposes. Spend-based factors can locate likely hotspots when primary data is unavailable; they are too coarse for judging which of two suppliers making the same component performs better.

Cost also extends beyond unit price. A low-cost supplier may expose the buyer to volatile energy surcharges, carbon-border costs, scarce materials, single-site dependency or an expensive redesign if switching becomes necessary. Supplier prioritisation should therefore consider total commercial exposure and the cost of inaction, not merely annual purchase value.

Build the minimum viable supplier baseline

Create one working dataset before sending another questionnaire. For each supplier, capture:

• legal entity, parent group, manufacturing site and country;
• purchasing category, product or material, annual spend and physical quantity;
• tier, contract owner, renewal date, switching difficulty and approved alternatives;
• estimated emissions, calculation method, factor source and data year;
• known process and energy hotspots, such as smelting, heat treatment, dyeing or clean-room electricity;
• climate target, reported Scope 1 and 2 emissions, product carbon footprint availability and assurance status;
• exposure to carbon regulation, physical climate hazards, energy-price volatility and other material due-diligence risks.

Do not wait for perfect master data. Reconcile the largest entities and highest-emitting categories first, record uncertainty explicitly and assign an owner to each gap. The objective of the first pass is to decide where better evidence is worth the effort.

If the company has a science-based target, translate target coverage into the dataset. The current SBTi supplier-engagement guidance 3 says supplier-engagement and/or reduction targets should collectively cover at least 67% of Scope 3 emissions when Scope 3 represents more than 40% of total Scope 1, 2 and 3 emissions. That threshold is target coverage, not an instruction to select the top 67% of suppliers by spend.

Score impact and urgency first

Use a 100-point impact-and-urgency score. The weights should reflect the business, but a workable starting model is:

Dimension	Weight	Evidence to use
Estimated carbon contribution	35	Supplier-specific data where credible; otherwise quantities, material/process factors and spend-based screening
Reduction potential	20	Energy mix, technology options, material substitution, design changes and realistic project pipeline
Supply and cost exposure	20	Switching difficulty, concentration, energy-price pass-through, lead times and margin sensitivity
Regulatory and customer exposure	15	Product rules, carbon costs, reporting needs and tender requirements
Environmental and social severity	10	Severity and likelihood of adverse impacts, including issues beyond climate

Normalise each dimension to a consistent scale and retain the raw evidence behind it. Never award zero carbon impact simply because data is missing. Use a conservative estimate and lower the confidence rating.

The US EPA supply-chain guidance 4 recommends choosing suppliers strategically, combining spend coverage with risks to financial performance or reputation and attention to energy-intensive operations. The principle remains useful: screen broadly, then concentrate deeper work where impact is material.

Due-diligence severity should remain visible rather than being averaged away. The OECD Responsible Business Outlook 2026 5 describes a high-level risk scoping exercise followed by deeper assessment of prioritised relationships, based on likely and severe impacts. It also reports that 63% of companies conducting supplier sustainability assessments did so without prior screening. A severe risk is not made less severe because the buyer has little leverage or a low spend.

Calculate actionability separately

Next, create a separate actionability score. This prevents a strategically important but difficult supplier from disappearing below an easy, responsive one.

Score factors such as:

• commercial leverage: share of the supplier’s revenue, volume commitment and access to senior decision-makers;
• timing: tender, contract renewal, product redesign or capital-planning window;
• supplier readiness: named owner, inventory, target, reduction plan and ability to share site or product data;
• data confidence: boundaries, year, allocation method, verification and connection to the purchased product;
• solution feasibility: access to renewable electricity, efficiency projects, lower-carbon materials, finance and technical support;
• internal ownership: procurement lead, engineering support, budget and an agreed business case.

For product-level data, use a consistent request rather than accepting incomparable PDFs. WBCSD’s PACT Methodology Version 3 6 provides a harmonised approach to calculating and exchanging cradle-to-gate product carbon footprints, including stronger data-reliability and verification pathways. Requesting a methodology, boundary, primary-data share and calculation period makes the response useful for procurement decisions, not only disclosure.

Low readiness does not always mean low priority. A high-impact supplier with weak data is an immediate validation and capability-building priority. A high-impact supplier with mature data and a feasible project is an immediate reduction priority.

Turn the two scores into four engagement tracks

Plot impact/urgency against actionability and assign a standard response:

1. High impact, high actionability — deliver. Agree a quantified reduction plan, milestones, data rules, commercial support and senior governance. Connect progress to sourcing decisions and business reviews.
2. High impact, low actionability — de-risk and build leverage. Validate the hotspot, engage the parent company, coordinate with other customers, explore finance or technical support, and prepare alternatives if progress remains blocked.
3. Lower impact, high actionability — standardise. Use tender criteria, renewable-electricity expectations, standard product-footprint templates and preferred-supplier recognition without heavy bespoke support.
4. Lower impact, low actionability — monitor. Maintain category estimates and minimum contractual requirements; revisit when spend, product design, regulation or risk changes.

This segmentation should determine the data request. Do not send a plant-level energy questionnaire to every small supplier. Use a short common request for the wider base, a more detailed primary-data template for material suppliers and an evidence review or site assessment only for priority facilities. CDP’s 2026 report on turning supply-chain disclosure into action 7 similarly stresses integrating environmental data into procurement and using combined data points to manage critical risks.

Prevent common scoring errors

Do not confuse estimates with performance. A spend-based footprint can rise when a buyer pays a premium for a lower-carbon product. Track physical activity and supplier-specific evidence before attributing a reduction.

Do not let a blended score hide red flags. Keep severe due-diligence issues, single-source dependencies and regulatory blockers as override flags. A high overall score cannot cancel them.

Do not reward polished reporting over actual change. Data maturity is useful, but it is not emissions performance. Score verified reductions, energy and process changes separately from disclosure quality.

Do not penalise small suppliers with disproportionate requests. The European Commission’s explanation of the CSRD value-chain cap 8 states that companies should seek only the information they really need and explains protections for value-chain companies with 1,000 employees or fewer in the CSRD reporting context. Even where a commercial request is legally possible, proportionality improves response quality and supplier trust.

Do not freeze the list for a year. Refresh scores when volumes, facilities, energy sources, contracts or product designs change. Review high-priority suppliers quarterly and the broader screen at least annually.

Put the model into procurement decisions

Start with a 90-day pilot. In the first month, reconcile the supplier master and estimate category hotspots. In the second, score 20–30 candidate suppliers with procurement, sustainability, finance and relevant technical teams. In the third, test evidence requests with a small group and approve the four engagement tracks.

Define governance before launch: who owns the methodology, who approves overrides, how suppliers can challenge incorrect data and which sourcing decisions the score can influence. Store each score with its source, date, confidence and owner. This creates an audit trail and prevents subjective ratings from becoming permanent facts.

The scorecard should feed tender weighting, contract clauses, preferred-supplier status, joint projects and executive business reviews. It should also connect to a practical supplier carbon assessment checklist and the wider supply-chain decarbonisation programme. If you want to adapt the model to your categories and APAC supplier base, talk to our team.