The silo problem
Most supplier emissions scorecards die the same way. Sustainability builds a spreadsheet, scores suppliers on carbon data, circulates it to procurement, and procurement keeps sourcing the way it always has — on price, lead time, and quality. The scorecard becomes a reporting artifact instead of a decision input. This happens because the scorecard was designed to satisfy a disclosure need, not a sourcing workflow.
For most companies, Category 1 (Purchased Goods & Services) and Category 4 (Upstream Transportation & Distribution) are where the bulk of Scope 3 emissions sit, and suppliers are the lever for both. If the scorecard doesn't influence which suppliers get business, it doesn't influence emissions. The fix isn't a better spreadsheet — it's designing the scorecard around procurement's existing decision logic from the start.
Start with the data quality problem, not the emissions number
Before you can compare suppliers meaningfully, you need to be honest about what kind of number you're comparing. The GHG Protocol's data quality hierarchy runs from spend-based estimates (weakest, derived from an emission factor applied to dollars spent), through average-data methods, to hybrid approaches, up to supplier-specific primary data (strongest).
Most supplier bases will have a mix. A scorecard that treats a spend-based estimate and a verified primary-data figure as equivalent will produce rankings that reward whichever suppliers happen to have poor data rather than whichever suppliers have lower actual emissions. A supplier using an industry-average emission factor can look cleaner than a supplier who disclosed detailed, slightly higher, verified numbers — simply because averages flatten outliers.
The practical fix is to score two things separately, not one blended number:
- Data quality tier — where the supplier's figure sits on the hierarchy (spend-based, average-data, hybrid, primary).
- Emissions performance — intensity or absolute figures, only compared within the same tier or category, never across wildly different product types.
This separation does double duty: it rewards suppliers for improving their measurement (moving up the hierarchy), which is often the fastest lever available to a supplier that can't yet decarbonize its operations, and it protects procurement from penalizing transparency.
What actually belongs on the scorecard
Keep the emissions side of the scorecard to a small number of fields procurement can act on:
- Data quality tier (as above).
- Emissions intensity or trend — year-over-year direction matters more than a single absolute number, especially early on, since it shows whether a supplier is managing the issue at all.
- Category relevance — whether the supplier falls into a high-impact purchasing category (materials-intensive goods, energy-intensive processing, freight-heavy logistics) versus a low-impact service category. Weighting emissions criteria differently by category prevents a low-spend, low-impact supplier from distorting the scorecard.
- Engagement responsiveness — whether the supplier responds to data requests and has a stated improvement plan. This is a leading indicator; a supplier with poor current data but an active primary-data transition plan is a better long-term partner than one that simply refuses to engage.
Resist the temptation to add more. A scorecard with a dozen carbon-related fields competing with cost, quality, and reliability fields will simply get ignored at the point of decision — procurement teams work fast, and complexity loses.
Blending with existing procurement criteria
The scorecard only changes behavior if emissions sit next to cost and quality in the same decision, not in a separate sustainability review that happens after sourcing is effectively settled. Two approaches work in practice:
- Weighted composite score: emissions criteria get an explicit weight (for example, treated as one of four or five scored dimensions alongside cost, quality, delivery reliability, and risk). The weight should be set deliberately and revisited — a 5 percent weight is a token gesture, a 50 percent weight will produce strange sourcing outcomes if the underlying data quality is still weak.
- Gating criteria: emissions data isn't scored but used as a gate — suppliers below a minimum data quality tier or with no improvement trend can't advance past a certain spend threshold or contract renewal without an engagement plan. This works well in the early years of a program when data quality across the supplier base is too inconsistent to score fairly, but it's useful as a forcing function to get suppliers reporting at all.
Most organizations start with gating and move to weighted scoring as supplier data quality matures — trying to weight a criterion you can't yet measure reliably just introduces noise into sourcing decisions.
Governance that keeps it alive
A scorecard needs an owner who sits in or works directly with procurement, not solely in sustainability. It needs a review cadence tied to existing sourcing cycles (quarterly business reviews, contract renewals) rather than a separate annual sustainability check-in. And it needs a feedback loop back to suppliers — a scorecard that scores suppliers but never tells them why, or what would improve their score, won't change supplier behavior, which is the entire point.
The suppliers most resistant to engagement are often also the ones with the least mature emissions data. That's not a reason to exclude them from the scorecard — it's the reason a supplier engagement program, run in parallel with scorecard design, matters: it gives procurement an escalation path other than simply dropping a supplier, and gives the supplier a route from spend-based estimate to primary data over time rather than a one-time audit.
Done well, the scorecard stops being a sustainability reporting tool procurement tolerates and becomes one more input procurement already trusts — which is the only way Category 1 and Category 4 emissions actually move.