Scope 3 Category 4 freight emissions — a CDP-ready reporting guide for 2026

Scope 3 Category 4 is upstream transportation and distribution — the freight emissions from goods you bought. The CDP climate questionnaire accepts GLEC-aligned numbers reported under ISO 14083 data-quality tiers. The mistake I see most often, across maybe two dozen submissions in the last 18 months, is the same one: companies report tonne-kilometres without disclosing tier, or they report tank-to-wheel where well-to-wake is required. Both get flagged in assurance.

What follows is the briefing I give clients before they hit submit. If you only have time for one section, read calculation approaches — that is where most of the avoidable damage lives.

What Category 4 actually covers

The GHG Protocol Corporate Value Chain (Scope 3) Standard defines Cat 4 as transportation and distribution of products purchased by the reporting company "in vehicles and facilities not owned or operated by the reporting company." The phrase that does all the work is "not owned or operated." If you operate the truck, it is Scope 1. If your supplier puts a pallet on a third-party carrier and bills you for freight, that movement is Cat 4. If the carrier subcontracts to another carrier, the emissions are still yours under Cat 4.

The boundary argued most often is Cat 4 versus Cat 1 (purchased goods). The cleanest rule: if freight is itemized separately — FOB Incoterm, separate carrier invoice, 3PL service fee — it goes in Cat 4. If freight is embedded in the good's price (CIF, DDP), report it in Cat 1 or break it out into Cat 4. I prefer breaking it out; it forces the per-shipment view you will eventually need.

Then there is Cat 9 (downstream T&D): freight on goods you sold, paid by someone else. A retailer shipping an online order on their own carrier contract puts that in Cat 4 if the customer pays freight separately, in Cat 9 if it is bundled into the sale price.

The five fields CDP wants filled in

CDP's questionnaire was restructured in 2024 and again for the 2026 cycle. Numbering shifts year to year; the substance for freight is stable. For 2026:

• C6.5 / Module 7.30 — Scope 3 emissions data. One row per category. For Cat 4: total CO₂e in tonnes, calculation methodology, and the share calculated from primary versus secondary data. The headline number, and the first thing assurance reconciles against your inventory.
• C6.5a / Module 7.30.2 — methodology disclosure. Free-text plus dropdowns. Calculation approach (spend-based, average-data, distance-based, fuel-based, hybrid), GHG Protocol Quality Score 1-5, and emission-factor source (GLEC, IEA, DEFRA, supplier-specific). This is where ISO 14083 tier and GLEC v3.2 version belong — in writing, every cycle.
• C6.7 / Module 7.45 — verification status. Whether the Cat 4 number is independently assured, by whom, to what standard (ISAE 3000, ISAE 3410, AA1000), at what level (limited or reasonable). Limited assurance is normal for the first three cycles; reasonable shows up around year four for mature programs.
• C7 — emissions breakdown. Optional but expected: disaggregation by business unit, country, or activity. For Cat 4 the useful cut is by mode. CDP scoring weights this heavily for Leadership-band scores.
• C8 — energy and SBTi alignment. If you have an SBTi-validated target including Scope 3, your Cat 4 trajectory has to roll up into it. The SBTi Corporate Manual requires Scope 3 in the target above the 40%-of-total threshold — nearly always true for asset-light shippers and retailers.

Codes differ slightly in the SME questionnaire and in supply-chain disclosure flows your customers may have routed you into; the substance is the same. If your auditor cannot trace the relationship between Cat 4 in C6.5, methodology in C6.5a, and verification scope in C6.7, that is a finding.

Spend-based vs distance-based vs fuel-based

Three calculation approaches are acceptable under the GHG Protocol. The honest version of each:

Spend-based multiplies dollars spent on freight by an industry-average factor per dollar — typically from an MRIO database like EXIOBASE or the EPA's USEEIO. Fast. Auditable. Nearly useless for managing emissions: the factor does not care whether the dollar was spent on ocean LCL or air express. A 10% spend cut on either registers as a 10% emission cut. Exactly backwards from how transport emissions work.

Distance-based multiplies tonne-kilometres by a mode-specific emission factor — GLEC defaults if you have nothing better, supplier-specific factors when carriers report them. Standard for serious freight reporting and where most clients land by cycle two or three. The tradeoff: you need shipment-level data — origin, destination, weight, mode. If your TMS does not track all four, that is a data-collection project before it is a methodology problem.

Fuel-based works backward from actual fuel burn — litres of diesel, tonnes of bunker, kilograms of jet-A — multiplied by a WTW combustion factor. Gold standard. Operationally hard: primary carrier data, in auditable format, on a matching cadence. Maersk and DHL will give it to you. The long tail of regional trucking subcontractors will not.

The sequence: spend-based in cycle one, distance-based with GLEC defaults by cycle two, fuel-based selectively on your top three carriers from cycle three onward. Distance-based for the long tail plus fuel-based for the top 60-70% of activity gets you a Quality Score 4 without forcing procurement to chase invoices for eighteen months.

The data quality tier auditors actually accept

GHG Protocol scores data quality 1-5: technological, temporal, and geographical representativeness, plus completeness and reliability. ISO 14083:2023 defines three default tiers that map onto this scale.

• Tier 1 (default factors). GLEC v3.2 average for the mode and vehicle class, no operator-specific information. Typically scores 3 — sometimes 2 if your geographic match is poor (European truck factor for a Southeast Asian lane). Fine for first-year reporting and the long tail of small shipments.
• Tier 2 (modelled factors). Default adjusted for operator-specific characteristics: vessel deadweight, Euro class, load factor, route circuity. Usually scores 4. Requires supplier-supplied operator data or a calculation engine that models against a known fleet baseline.
• Tier 3 (primary fuel data). Activity-specific fuel consumption from the operator, multiplied by a published WTW combustion factor. The only tier that consistently scores 5. Requires the kind of carrier cooperation only top-tier global operators reliably deliver.

A score of 3 across your Cat 4 inventory is fine for the first submission. By cycle two, assurance teams start asking pointed questions if you have not moved any meaningful share to Tier 2 or 3. By cycle three they expect a data-improvement plan with named carriers and dates. If you are still at uniform Tier 1 in year four, you will not reach Leadership band regardless of the narrative.

What the EcoFreight API returns and how it maps to CDP

The EcoFreight calculation API is shaped to make the CDP mapping mechanical. When you POST a shipment to /api/v1/calculate, a few response fields are worth knowing by name.

• emissions.wtw is the well-to-wake CO₂e in kilograms. The number that goes into CDP C6.5 Cat 4. WTW is what CDP asks for — tank-to-wheel will be flagged.
• calculation.emission_factor.source identifies the factor used: "GLEC Framework v3.2", "IMO Fourth GHG Study 2020", or a supplier-specific identifier where the carrier has reported primary data. Paste into the methodology field in C6.5a.
• calculation.emission_factor.quality is an integer 1-5 that maps directly onto the GHG Protocol Quality Score. Tier 1 returns 3, modelled operator factors 4, primary fuel data 5. Aggregate across shipments for the weighted-average score in C6.5a.
• calculation_id is stable across versions of the same shipment. Keep these in your evidence package — if assurance asks you to show that two reporting cycles used consistent methodology, the calculation_id plus the response payload is your audit trail.

The mechanical mapping is the point. Most errors I see are not arithmetic — they are field-mapping: right number, wrong cell, methodology missing the GLEC version, quality score buried in a comment instead of the dropdown.

One unhedged opinion

Spend-based emissions reporting is a vanity exercise. It satisfies the disclosure checkbox without ever telling you which lane to change. If a sustainability team is still on spend-based after their second CDP submission, the budget is being spent in the wrong place. The CDP score improvement from moving to distance-based is real but secondary; the operational improvement — seeing your worst lanes, your highest-emitting carriers, your modal-shift opportunities — is the whole point.

One acknowledged gap

I have never met a Scope 3 Category 4 disclosure that survived contact with the company's TMS data without finding at least one shipment lane the freight desk forgot to mention. Charter air, intercompany sample shipments, returns logistics, expedited ground from the back-up carrier you only use during seasonal peaks — one of these is always missing. The first reconciliation pass after primary data goes in usually moves the Cat 4 total by 5 to 15%. I budget two cycles to get a reporting boundary I trust. Anyone telling you the first-year number is precise is selling you something.

Quarterly checklist for your CDP submission

Six things I run every quarter in the lead-up to a CDP submission, in order:

1. Reconcile Cat 4 against three independent sources. Calculation engine output, procurement spend rolled up to a spend-based estimate, and top-five carriers' reported tonnage. If those three do not agree to within 15%, close the gap before you submit.
2. Sample 30 shipments and walk the audit trail end-to-end. TMS booking record → API call → emission factor source → CDP cell. If you cannot complete the walk for any sampled shipment, the methodology disclosure is not ready.
3. Refresh the factor source statement. Confirm "GLEC Framework v3.2" (not v3.1 or earlier) is cited in C6.5a, and ISO 14083:2023 is referenced as the framework. Auditors check these strings first.
4. Weight quality by activity, not shipment count. A million Tier 1 small parcels and one Tier 3 ocean container do not weight equally. Weight by tonne-kilometre or by emission-tonne — document which.
5. Compare trajectory against SBTi. If absolute Cat 4 is not on a 4.2% per year reduction pathway (the SBTi near-term default for Scope 3), the CDP narrative needs an explicit explanation, not a hopeful one.
6. Lock methodology before you re-run the numbers. Mid-cycle changes are the single largest source of restatements. Pin GLEC version, ISO tier policy, and calculation engine version at period start; freeze until the cycle closes.

Where to start

First CDP cycle? Start with a clean baseline. The EcoFreight calculator uses GLEC Framework v3.2 emission factors per shipment and returns the WTW number in a form that maps directly to CDP C6.5. The factor sources and ISO 14083 tier mapping are documented on the methodology page; the API lives at /docs. If you are sequencing this behind broader regulatory exposure — CSRD, EU ETS Maritime, IMO CII, plus CBAM and FuelEU Maritime for shipping-heavy reporters — the 2026 freight compliance checklist covers how those obligations layer on top. And if you are at the residual-emissions question, the carbon credits and offsets guide covers what comes next.

Sources

CDP technical guidance and current questionnaire structure: CDP guidance hub. Scope 3 category definitions and quality scoring: GHG Protocol Corporate Value Chain (Scope 3) Standard and Scope 3 Calculation Guidance. Tier framework: ISO 14083:2023. The Smart Freight Centre's GLEC Framework v3.2. SBTi guidance: SBTi Corporate Manual and transport sector guidance. Commentary on assurance practice reflects how I have seen Big Four and second-tier ESG teams approach Cat 4 disclosures over the last three reporting cycles.