The tension is structural, not just interpersonal
Most buyer-side Scope 3 programs eventually run into the same wall: the categories that dominate the inventory — typically Purchased Goods & Services (Category 1) and Capital Goods (Category 2) — are also the categories where the buying company has the least direct control over the underlying activity. The emissions happen in someone else's factory, using someone else's energy mix and someone else's process choices. To improve the number, you need the supplier to change something or, at minimum, hand over better data. That puts procurement teams in an odd position: they're asking counterparties for cooperation on a metric that, historically, had no bearing on the commercial relationship at all.
The risk on the other side is real. Suppliers, especially smaller ones with thin margins and multiple buyers, will walk away from a customer that turns emissions reporting into an unfunded, unexplained compliance burden — particularly if a competitor down the street is asking for the same product with no such requirement. Lose enough suppliers and you've solved a reporting problem by creating a supply continuity problem, which is a worse outcome for everyone, including the climate goal.
Why the pressure is showing up now
Supplier engagement was optional in practice for a long time. It's becoming less optional because the disclosure regimes above the buying company are starting to require exactly the granular data that only suppliers can provide. CSRD's ESRS E1 requires Scope 3 disclosure from in-scope companies, phased in by size band, and it reaches non-EU companies with a large enough EU subsidiary footprint — so multinational suppliers outside Europe are already feeling this indirectly. California's SB 253 adds a Scope 3 disclosure requirement, with phased third-party assurance, for large companies doing business in the state regardless of where they're headquartered. Assurance requirements matter here: once a number has to be verified, spend-based estimates stop being good enough for the categories that matter most, and buyers have to go get supplier-specific data or accept a weaker number.
CBAM adds a parallel pressure in the other direction: EU importers of iron and steel, aluminium, cement, fertilisers, hydrogen, and electricity need embedded-emissions figures calculated on the actual production route where possible, not a default value, which means upstream suppliers in those sectors are being asked directly for process-level data rather than a company-wide average.
The point of naming these regimes isn't that every supplier is in scope of all of them. It's that the direction of travel — data quality hierarchy climbing from spend-based toward supplier-specific, verification requirements phasing in over time — is consistent across jurisdictions. Buyers who mirror that phased approach with their own suppliers are working with the grain of where reporting is headed, not against it.
A sequence that reduces attrition risk
1. Education before anything else
Before a supplier is asked to report a number, they should understand what the number is for, how it will be used, and what the buyer is and isn't going to do with it in year one. This means walking through the basics: what Scope 3 category their goods or services fall into, why spend-based estimates are the default starting point and why they're weak, and what a realistic glide path toward hybrid or supplier-specific data looks like. Suppliers who understand the mechanics are far less likely to treat the request as an arbitrary hoop, and far more likely to volunteer better data once they have it, because they can see where it's going.
2. Scoring once capability exists, not before
Scorecards and supplier rankings are useful for prioritizing engagement effort and tracking progress, but introducing them before suppliers have had a reasonable window to build measurement capability just tells suppliers they're being judged on a test they weren't told about. Scoring should follow one or more cycles of education and data collection, and the score should initially function as a diagnostic — where to focus engagement resources, which suppliers need help versus which are already capable — rather than a pass/fail gate tied to commercial consequences.
3. Contractual requirements last, and phased
Written requirements — emissions reporting clauses, targets tied to contract renewal, minimum data quality thresholds — should arrive only after suppliers have had real runway to build toward them, and they should be phased the way the regulatory examples above are phased: assurance and rigor tightening over defined periods rather than landing all at once. A supplier asked in year one to hit a supplier-specific data standard with third-party verification, with no prior notice, is a supplier who will look for another buyer if one exists. A supplier told in year one what's coming, supported through data collection in year two, and held to a contractual standard in year three is far more likely to still be a supplier in year four.
Where this fits in the broader inventory
This sequencing matters most for the categories most dependent on supplier cooperation — Purchased Goods & Services, Capital Goods, and Upstream Transportation & Distribution — because these are exactly the categories where the data quality hierarchy runs from spend-based up through average-data and hybrid to supplier-specific, and where jumping straight to the top rung without supplier buy-in tends to produce data nobody trusts and relationships nobody wants to repair.