The limits of the annual survey
Most companies still collect Scope 3 Category 1 (Purchased Goods & Services) data the same way they did five years ago: a questionnaire goes out once a year, procurement or sustainability teams chase responses, and whatever comes back gets blended with spend-based estimates to fill the gaps. This works well enough to produce a number for the inventory, but it rarely produces a number that improves over time. Response rates vary year to year, suppliers answer differently depending on who fills out the form, and because there's no obligation attached, the easiest path for a supplier is to give a rough answer or skip the question entirely.
The result is that most purchased goods and services emissions stay anchored at the weakest end of the GHG Protocol's data quality hierarchy — spend-based — long after a company has set targets that assume better data is coming. A voluntary survey doesn't change supplier incentives. A contract does.
Why procurement language is the next step
The shift from asking to requiring is being pulled forward by disclosure regimes that put a time limit on how long "we're still collecting data" is an acceptable answer. CSRD under ESRS E1 expects Scope 3 disclosure with double materiality, phased in by company size, and it can reach non-EU companies with a large enough EU subsidiary footprint — which means their suppliers, EU or not, feel the pull. California's SB 253 includes a Scope 3 requirement with phased third-party assurance, which means spend-based estimates eventually have to survive an auditor's questions, not just an internal review. India's BRSR Core tightens the KPI set for the country's largest listed companies and phases in assurance as well. In each case, the reporting company is the one on the hook, but the pressure runs backward through the supply chain because better disclosure requires better supplier data.
An annual survey can't deliver that reliably. A standing contractual requirement can, because it changes the relationship from a request to an obligation with consequences.
What contractual language actually looks like
In practice, companies are adding a handful of recurring elements to procurement contracts and master service agreements:
- Data provision clauses that specify what the supplier must report (product- or activity-level emissions data, not just a company-wide total), how often, and in what format, so the data can actually be used in a Scope 3 calculation rather than just filed away.
- Data quality escalation language that references the GHG Protocol hierarchy directly — starting with a baseline obligation to provide activity data sufficient for average-data or hybrid methods, with a defined path toward supplier-specific primary data for higher-spend or higher-risk categories.
- Audit rights, giving the buying company the ability to verify reported figures, request supporting documentation, or conduct site-level checks, even if those rights aren't exercised on every supplier every year.
- Phased requirements, where new contracts or renewals introduce the obligation gradually — often starting with the largest suppliers by spend or emissions contribution, then extending down the supplier base as internal systems and supplier capability catch up.
The phasing matters more than it might seem. A company that tries to impose full primary-data requirements on every supplier at once usually gets inconsistent compliance and strained relationships. Sequencing by spend concentration — since a small number of suppliers typically account for most of Category 1 emissions — gets more usable data with less friction.
The trade-off: contractual versus relationship-based
Making disclosure contractual solves the enforcement problem but introduces new ones. A clause gives you leverage: if a supplier wants the business, they provide the data, and the audit rights give you a mechanism to check accuracy rather than take reported numbers on faith. It also scales — the same language can sit in hundreds of contracts without requiring a bespoke conversation with each supplier.
But leverage isn't the same as capability. A supplier can sign a clause and still not have the metering, systems, or expertise to produce anything better than a rough estimate. Treating the requirement purely as a legal obligation risks turning it into a compliance exercise — suppliers produce a number that satisfies the contract language without it reflecting real operational improvement, and the buying company ends up with data that looks more precise than it is.
Relationship-based engagement — training, shared tools, joint pilot projects, capacity-building support — tends to produce more accurate data and more actual emissions reduction, but it doesn't scale across a supplier base of thousands, and it has no enforcement mechanism if a supplier simply doesn't prioritize the work.
The approach that tends to hold up is a hybrid: contract language sets the floor — what must be disclosed, by when, with what audit rights — while supplier engagement programs handle the harder work of getting suppliers from a spend-based estimate to something closer to primary data. The clause creates the obligation; the relationship is what makes the obligation deliverable rather than just enforceable.
Designing it so it doesn't backfire
A few practical considerations determine whether this works:
- Tie data quality expectations explicitly to the hierarchy, so suppliers know what "better" means in concrete terms rather than vague requests for "more detail."
- Reserve audit rights for a risk-based subset of suppliers rather than contractual boilerplate everyone ignores — unused audit rights signal that the clause isn't really enforced.
- Phase requirements by spend and emissions contribution, not alphabetically or by contract renewal date, so the first wave of compliance actually moves the inventory.
- Pair the contract clause with support, not just a deadline, for suppliers who lack the systems to comply on day one.
Done well, contractual disclosure requirements turn Category 1 data from an annual guess into something that improves on a predictable schedule — which is what the newer disclosure regimes are increasingly going to expect buying companies to demonstrate.