7

Module 7  ·  Track 2: Operational Practice

Procurement, Supply Chains, and Vendor Evidence

Procurement is the bridge between your sustainability commitments and your actual Scope 3 footprint. Every purchasing decision is a vote.

Duration28–32 minutes
TrackOperational Practice

What you will take from this module

Why procurement is a strategic lever

The first thing to understand is the difference between a supplier's corporate ESG commitments and product-level evidence. Neither substitutes for the other.

A supplier's annual sustainability report may state ambitious net-zero targets, renewable energy commitments, and industry certifications. None of that tells you the footprint of the specific product you are buying, the specific service you are consuming, or the specific manufacturing supply chain behind it.

Corporate sustainability commitments describe the supplier's own journey. Product sustainability evidence is about the specific transaction between you and them. Both matter. When a supplier says it is committed to net zero by 2040, that is information about their trajectory. When you ask for a product carbon footprint for the specific server model you are procuring, you are asking for information about this purchase.

Conflating the two is one of the most common ways that Scope 3 data quality erodes.

What good procurement evidence looks like

When you receive a lifecycle assessment or product carbon footprint, ask seven questions:

  • Methodology: ISO 14040, ISO 14044, or another recognised standard?
  • Lifecycle stages: Cradle to gate, cradle to grave, or cradle to cradle?
  • Emission factors: Which ones, and for which geographies?
  • Use-phase assumptions: What energy estimates underpin the calculations?
  • Data sources: How complete is upstream supply chain data?
  • Functional unit: What is the reference flow?
  • Assurance status: Self-declared, reviewed, or independently assured?

Coverage statements matter

A coverage statement is a short, explicit description of what a vendor's footprint figure includes and excludes. Without it, a footprint number is almost meaningless for comparison purposes.

A defensible coverage statement should include: what emission sources are covered and which are excluded (with materiality rationale), what lifecycle stages are in scope, what geographic scope applies, what data sources were used and with what confidence, and what the refresh cadence of the figure is.

Assurance tiers comparison

Vendors frequently use the word "verified" loosely. There are three meaningful tiers. Treating these as equivalent is a significant quality error. This matrix shows what coverage each tier provides.

Coverage Area
Self-Declared
Peer-Reviewed
Independently Assured
Vendor Controls Methodology
Methodology Verified
Source Data Audited
Cost to Obtain
Low
Medium
High

Self-Declared

Vendor-produced, no external review. Speed to deliver, minimal scrutiny. Useful as a starting point, but label it explicitly as low confidence. Early-stage engagement only.

Peer-Reviewed

Methodology verified, calculations spot-checked, but source data not audited. Better than self-declared. Good for significant suppliers where you need meaningful verification without full audit cost.

Independently Assured

Full third-party verification: data to source, methodology audited, standards compliance verified. Full verification to stated assurance standard. For strategic suppliers and mandatory reporting.

The cost-confidence trade-off

Not all vendor data needs to be independently assured. The assurance tier you request should match the materiality of the spend and the weight of the data in your reporting. Requiring third-party assurance on 0.1% of Scope 3 footprint is over-engineering. Treating self-declared data as assured when it represents 20% of your footprint is a material error.

Vendor evidence checklist explorer

The evidence you request varies by what you are buying. Click a domain tab to see the specific checklist for that category.

Product carbon footprint with lifecycle breakdowns

Request the PCF for the specific model and configuration you are procuring. Must show manufacturing, transport, use phase, and end-of-life stages separately.

Energy consumption figures (idle, typical, peak)

Required for use-phase modelling. Demand specific model data, not category averages. Include sleep and standby modes.

Repairability, spare parts, and take-back programme

Reparability score, battery replacement options, spare parts availability, warranty period, and evidence of actual reuse or responsible recycling outcomes.

Conflict minerals declaration

Transparency on sourcing of minerals from conflict-affected regions. Conformance to RBA standards or equivalent.

Comparable functional units

If comparing models, ensure functional units are equivalent. Comparing a 4-core CPU with an 8-core CPU using the same PCF is misleading.

PUE and WUE with metering coverage

Operational Power Usage Effectiveness and Water Usage Effectiveness. State metering coverage and calculation method. Site-level data, not portfolio averages.

Carbon intensity of energy supply

Distinguish between location-based and market-based figures. Ask what instruments support renewable claims: certificates, power purchase agreements, or on-site generation?

Utilisation figures and decommissioning commitments

How is spare capacity accounted for? What is the commitment to responsible end-of-life handling of equipment?

Water sourcing and seasonality (if applicable)

For water-stressed locations, request water sourcing, WUE seasonality, and recycled water use. Critical for supply chain resilience.

Third-party certifications

ISO 50001 (energy management), LEED, BREEAM, or equivalent. Verify currency and scope of certifications.

Region-level carbon disclosures with methodology

Request region-level carbon intensity and the methodology used for allocating emissions to your consumption. Not global portfolio averages.

Location-based and market-based figures

Ask whether figures are location-based or market-based. Request both if available. Understand the implications for your Scope 2 allocation.

Allocation methodology and basis

Is allocation by server-hours, vCPU-hours, revenue, or proprietary method? Does it reflect physical consumption or is it divorced from actual resource use?

Real-time carbon intensity data

Can the provider supply real-time or near-real-time carbon intensity by region? Critical for workload scheduling and demand-response decisions.

Renewable energy commitments and evidence

What instruments support renewable claims? Power purchase agreements, certificates, or on-site generation? What is the scope and commitment timeline?

Hosting location transparency and energy sourcing

Where are the data centres located? What is the energy sourcing at those facilities? Which cloud providers are used upstream?

Service-level footprint figures

Request service-level footprint if available. Ask about the methodology for allocating shared infrastructure emissions to your tenancy.

Per-tenant intensity metrics

Are per-tenant intensity metrics published? Can they be provided on request? Essential for comparing efficiency across users.

Data centre certifications and standards

ISO 50001 (energy management), LEED, BREEAM, or equivalent. Verify currency and scope. Certifications reduce audit burden.

Regional hosting commitments (contractual)

Can the provider commit to hosting in regions that meet your carbon intensity requirements? Are those commitments contractually enforceable?

Allocation pitfalls: double counting and misleading denominators

Once you have product and service evidence, the next challenge is allocation: how a supplier attributes shared emissions to your specific consumption. Two things regularly go wrong.

How the same emissions can be counted in multiple places

Your Organisation

Counting cloud footprint

25 tCO₂e

Cloud Provider

Counting facility emissions

Same 25 tCO₂e

Result: The same emissions appear twice in the inventory

Your numbers look more complete while becoming less accurate.


The remedy: boundary documentation and single-count principle

Count each source once, in the most specific and accurate place you can.

  • Data centre emissions: Count facility emissions at the data centre owner level, not at the cloud provider level.
  • Cloud workloads: Count cloud provider efficiency and operational emissions, but reference to the data centre's underlying carbon intensity.
  • Hardware embodied emissions: Count at the device level, not again inside a broader supply chain category.
  • Clear boundaries: Document which entity is responsible for each source. Make allocation assumptions explicit.

Misleading allocation denominators

Suppliers can allocate shared emissions using different bases: server hours, VM-hours, vCPU-hours, energy consumption, revenue, or proprietary methods. Those choices can materially change the output.

✓ Allocation by consumption

Server-hours, vCPU-hours, energy use. Reflects actual resource consumption. Rewards efficiency.

✗ Allocation by revenue

Detached from physical consumption. Will not reward efficient use. Can obscure operational levers.

✗ Allocation by contracted capacity

Penalises efficient tenants. Rewards wasteful ones. Creates perverse incentives.

The question to ask any supplier is simple: What is the allocation basis, and does it reflect physical consumption? Your goal is not perfection. It is avoiding false precision and ensuring the numbers can actually influence decisions.

Circular procurement and hardware social responsibility

Sustainable hardware procurement goes beyond energy efficiency and carbon footprints. It includes circular economy criteria and social responsibility throughout the supply chain.

Circular Economy Criteria

  • How repairable is this product?
  • Are spare parts available and supported?
  • Does the manufacturer offer a take-back programme?
  • Is there an upgrade path to avoid full device replacement?
  • Can you access the device's repair manual?

Social Responsibility Criteria

  • Labour standards in manufacturing supply chains
  • Conflict mineral declarations and sourcing transparency
  • Supply chain transparency and visibility
  • Alignment with RBA Code of Conduct or equivalent
  • Health and safety standards verification

IT hardware contains rare earth minerals and materials sourced from supply chains with serious social and environmental risks. The Responsible Business Alliance Code of Conduct sets out expected standards for labour, health and safety, environmental practice, and ethics in global electronics supply chains. Requiring suppliers to demonstrate alignment is increasingly a standard procurement requirement.

The vendor sustainability scorecard approach

Apply a consistent scorecard across categories with four dimensions:

  • Corporate commitments: Net-zero targets, renewable energy pledges, industry certifications
  • Product-level evidence: PCF, energy consumption, lifecycle data, assurance level
  • Circular economy criteria: Repairability, spare parts, take-back programmes, upgrade paths
  • Social responsibility: Labour standards, supply chain transparency, RBA alignment

None of these should substitute for the others, and none should be treated as sufficient on its own.

Four-step vendor management lifecycle

Sustainable procurement works best as an ongoing programme rather than a one-off assessment at contract award. Each step builds on the previous, creating a continuous cycle of engagement and improvement.

Step 1: Qualify

Establish and maintain the programme.

Define your sustainability requirements by category. Set minimum evidence thresholds by assurance tier. Build the vendor sustainability scorecard. Create a supplier engagement process that distinguishes between strategic partners (where you invest in collaborative improvement) and transactional suppliers (where you apply standard requirements and enforce consistently).

Step 2: Embed

Onboarding and divestment decisions.

New suppliers should be assessed against sustainability criteria as part of standard onboarding. Existing suppliers who cannot meet evidence thresholds should be given a defined improvement timeline. If improvement does not follow, that information should carry weight in renewal decisions. Procurement teams need the authority and support of technology leadership to act on this.

Step 3: Monitor

Sustainable IT value realisation.

Work with key suppliers to improve evidence quality over time. Use your buying power actively. Collective purchasing frameworks and industry coalitions can amplify the signal, particularly for smaller organisations whose individual volume does not create sufficient leverage. Schedule regular reviews and evidence updates.

Step 4: Escalate

Vendor lifecycle management and renewal decisions.

Sustainability performance should be a component of every contract renewal evaluation. Not the only component, but a genuine one, with the ability to affect commercial outcomes. If sustainability criteria are included in selection but carry no weight in renewal, suppliers will learn to treat them as box-ticking exercises.

Bridging the ambition-reality gap

At scale, technology organisations have real buying power. That power can be used to drive supplier behaviour, create market demand for better evidence, and make sustainability a factor in every commercial relationship, not just a criterion in a sustainability questionnaire filed once at contract award.

The difference between a procurement team with authority and without is the difference between a policy and an outcome.

The refresh trap.

Organisations default to one of two extremes with hardware lifecycle management. Both are wrong. Lifecycle discipline means getting out of the lazy middle.

Refreshing too early

Replacing hardware on a fixed cycle (three or four years) regardless of condition, performance, or actual need. This maximises embodied carbon waste. A device replaced a year early carries all of its manufacturing footprint into the waste stream prematurely, while a new device carries its full embodied carbon into the estate. The combined impact is significant.

Common driver: vendor refresh incentives, IT asset management defaults, or a culture that equates new equipment with better performance.

Refreshing too late

Running hardware until failure, regardless of energy performance degradation, security risk, or repair cost. Older devices consume more power, generate more heat, require more support, and often run outside warranty. The operational carbon cost of running an inefficient device for an extra two years can exceed the embodied carbon of replacing it.

Common driver: budget constraints, lack of lifecycle data, or the assumption that keeping hardware longer is always the sustainable choice.

The discipline: evidence-based lifecycle decisions

The right refresh point is not a fixed number of years. It is the point where the operational carbon cost of continuing to run the device exceeds the embodied carbon cost of replacing it, adjusted for performance, security, and support. That calculation requires actual data: energy consumption profiles, repair history, utilisation, and the embodied carbon of the replacement. Without that data, the decision defaults to convention, and convention is usually wrong in one direction or the other.

⏸ Pause & Reflect

Take 10–15 minutes. Write answers down. Specificity matters more than completeness.

1
What is the largest category of technology spend in your organisation? What assurance tier does your evidence currently sit at for that category?
2
Pick one vendor of strategic importance. What evidence could you request from them tomorrow? What would it take to move from self-declared to reviewed or independently assured?
3
Look at your last major technology contract. Was sustainability included in the initial RFP? In the renewal decision? What would it take to make it a genuine factor in both?
4
Can you identify a double-counting risk in your current Scope 3 footprint? What boundary documentation would you need to resolve it?

Discussion question: Where is procurement currently showing up in your organisation's sustainability programme, and where should it show up?

Knowledge Check · Module 7 · Q1

What is the key difference between corporate ESG commitments and product-level evidence?

Select an answer to reveal the explanation.

✓ Correct: Option C

A supplier may have ambitious net-zero targets and industry certifications, but none of that tells you the footprint of the specific product you are buying. Conversely, a single product footprint does not tell you whether the supplier is on a genuine sustainability journey or just window-dressing on an otherwise extractive business model. You need both types of evidence, and conflating them is one of the most common ways Scope 3 data quality erodes.

Knowledge Check · Module 7 · Q2

A cloud provider tells you their Scope 2 emissions are allocated to customers by revenue. What should be your primary concern about this allocation basis?

Select an answer to reveal the explanation.

✓ Correct: Option B

The best allocation basis reflects actual resource consumption (server-hours, vCPU-hours, energy use). Revenue-based allocation is divorced from physical demand, which means it creates perverse incentives: an efficient, well-optimised tenant pays the same per unit revenue as a wasteful one. This obscures the operational levers where you could actually drive reduction. When you ask a supplier for their allocation basis, the question to ask is: does it reflect physical consumption?

Knowledge Check · Module 7 · Q3

In the four-step vendor management lifecycle, at which step should sustainability performance be weighted in a contract renewal decision?

Select an answer to reveal the explanation.

✓ Correct: Option C

If sustainability criteria are included in the initial RFP but carry no weight in the renewal decision, suppliers will quickly learn to treat them as box-ticking exercises. The renewal decision is where procurement has the greatest leverage. Making sustainability a genuine factor, not the only one, but a real one that affects commercial outcomes, is what separates policy from practice.

Module 7: Key Takeaways

Procurement is a strategic lever.

At scale, technology organisations have real buying power. That power can drive supplier behaviour, create market demand for evidence, and embed sustainability in every commercial relationship.

Corporate ESG commitments and product evidence are not interchangeable.

A supplier's net-zero target tells you about their journey. A product carbon footprint with assurance tells you about this transaction. You need both, and conflating them erodes data quality.

Assurance tiers matter. Treat them as meaningfully different.

Self-declared, peer-reviewed, and independently assured are not equivalent. The assurance tier you request should match the materiality of the spend and the weight of the data in your reporting.

Allocation basis can make or break the utility of the data.

Allocation that reflects physical consumption (server-hours, vCPU-hours, energy) drives decisions. Revenue or capacity-based allocation obscures operational levers and can create perverse incentives.

Procurement is where ambition meets reality.

The difference between a procurement team with authority and without is the difference between a policy and an outcome. Sustainability must carry weight in renewal decisions, not just initial selection.

Procurement sets the evidence standard. But evidence without governance is a filing exercise. Module 8 closes Track 2 by connecting the threads: who owns the programme, how maturity is assessed across dimensions, what governance artefacts make accountability real, and why most programmes stall not because they lack ambition but because they lack the operating discipline to sustain it.

← Previous M6: Software Engineering and Application Sustainability