From spend to emissions: a practical guide for procurement teams

Asked where a sustainability team should start looking for data, Höskuldur Arason gave a one-word answer: "I have just one word. Procurement." He integrates ERP systems for Klappir's customers, so he knows the answer from having looked in most other places first. Procurement is where the detail lives, and detail is the whole problem.

The finance side of an ERP system cannot answer the question, and it is worth being precise about this, because it is where many data projects begin and stall. Finance modules hold monetary values grouped by accounting key. You can see that a purchase was made and what it cost. You cannot see what was bought, how much, or in what unit. "You don't get the individual lines," as Höskuldur puts it. "You have no idea which products are behind this." The procurement side is a different picture: individual line items, product descriptions, and, in many cases, quantities in kilograms, litres, and kilowatt-hours. Those quantities are activity data. They are the numbers an emissions calculation actually wants, and your organisation is already producing them as a by-product of buying things.

Two questions tell you whether you have something to work with, and you can answer both before lunch. Is your organisation actively using the procurement module, rather than posting purchases straight to finance? Are you sending and receiving electronic invoices? If the answer to both is yes, the data exists in a structured form, and the rest of this guide applies. If the answer to either is no, that is your first piece of work, and it is a procurement systems question rather than a sustainability one.

A purchase line that can carry emissions has six fields, and most procurement systems already produce five of them without being asked. A supplier. A description of what was bought. A quantity. A unit. An amount. A date. Pull ten lines out of your own system and check them against that list. In most organisations, the first five are there, in varying states of tidiness, and one is missing.

The missing field is classification, and it decides whether your data can travel. A description written by whoever raised the purchase order is readable by a person and useless to a system. "Cleaning supplies", "cleaning products", "cleaning matl", and a supplier's own product code are four ways of writing one category, and no calculation can group them.

Classification is a code that means the same thing to everyone: UNSPSC, the United Nations Standard Products and Services Code, is the scheme built for exactly this and the one procurement systems are most likely to understand already. Once a line carries a code, you can attach an emissions factor to it; your supplier can recognise the same purchase throughout.

Spend is where you start and not where you stop, and the difference is worth seeing in a real example rather than in the abstract. Line Knudsen of Keepers calculated business travel for a private equity firm with 60 portfolio companies, first from spend and then from actual flight distances. On a Copenhagen to Stockholm return of 520 kilometres at DKK 3,800, the spend-based method produced roughly 275 kg CO2e, while the activity-based method produced roughly 135 kg. The spend figure was not a small error. It was close to double, and the reason is structural rather than accidental: Nordic ticket prices are high, and multiplying a high price by a factor per unit of currency inflates the result beyond what the aircraft did.

"You can use spend-based data to identify reduction potential," Line says, "but I would always recommend going activity-based where possible, because it is more precise." That is the honest position on spend-based accounting. It is a fast, low-effort first pass that tells you which categories matter.

Classification lets you upgrade one category at a time, and that's what makes the whole thing achievable. Once your purchases carry codes, you can see which categories owns the footprint, and you can target activity data for those while leaving the rest on spend.

Fuel invoices, electricity meter readings and vehicle logs are usually within reach for Scope 1 and 2. Purchased goods and services are harder, and it is where the supplier conversation begins. A single-category supplier, a fuel provider or a logistics partner, is straightforward, because the activity is well defined. A general supplier whose account mixes stationery, cleaning products and paper is hard, and spend-based is often the only practical option until the relationship matures. Pavlina Caccioni of Konges Sløjd, who has worked through this across several industries, is direct about how that goes: "It is about collaboration. Especially with smaller or local suppliers, you have to approach it as working on this together, rather than demanding data they do not yet have."

Procurement is not doing the sustainability team a favour by taking this on, because the pressure that makes this work necessary arrives at procurement first. A manufacturer loses a customer contract, not on price or quality, but because its ESG scorecard does not meet the buyer's procurement requirements. A mid-sized company approaches its bank for refinancing and finds that clean sustainability data has become a condition of favourable terms. In both cases, the requirement lands as a question a buyer has to answer about their own supply chain, and as a question their customers are now asking them. Answering it takes the same classified purchase data. The organisation that has it can respond to a tender in a week. The organisation that does not spends that week in a spreadsheet and often loses the tender to the one that answered.

Three things a procurement team can do this week, none of which need a project, a budget or a new system. Answer the two diagnostic questions, so you know whether the structured data exists. Pull your top ten spend categories and ask, for each one, whether you could say what was bought in units rather than in currency. Then take the single largest category and check how consistently it is described across suppliers, because that one category will tell you the size of the classification job across all of them.

Klappir accepts spend as an input for purchased goods and services, so a purchase record becomes an emissions figure without waiting for weight data on every line. It classifies purchase lines to UNSPSC, which turns a description into something useful. Klappir is not a reporting tool built for one team. It is the data infrastructure that connects organisations so structured data moves on its own.

What to read next. If you are the sustainability manager, the Finance Meets Sustainability guide covers the ground upstream of this one: finding the data, reading what it tells you, and turning it into a funded proposal. The three recorded sessions behind it feature Kirstine Hartung Larsen of K² Nordic, Höskuldur Arason of Insolica and DataDrive, Line Knudsen of Keepers, Pavlina Caccioni of Konges Sløjd, and Christian Boserup of ESG Implementation. If you want to do the work rather than read about it, the first chapter of the Data Readiness course at Klappir Academy is free.

Book a 30-minute call with the Klappir team.