How to Build a Scope 3 Emissions Inventory Without a Sustainability Team

Every mid-market company selling to enterprises hits the same wall. A customer sends an RFP with an ESG section, or a CSRD deadline lands on the CFO's calendar, and someone realizes that Scope 3 is 80% of the footprint but no one has ever counted it. The instinct is to hire a consultant, wait 8 weeks, and pay $60,000 for a report that is stale before it prints.

There is a better path. Scope 3 is an accounting problem, not a climate science problem. If you can close your books, you can build a defensible inventory. Here is how.

What is a Scope 3 emissions inventory and why do most fail?

Scope 3 covers the 15 categories of indirect emissions defined by the GHG Protocol Corporate Value Chain Standard. Everything upstream of your operations (purchased goods, transportation, business travel) and everything downstream (use of sold products, end-of-life treatment) counts. For most B2B services companies, Scope 3 is 80 to 95% of total emissions.

Most first-time Scope 3 inventories fail for three reasons.

• Modeling instead of accounting. Teams build a bottoms-up estimate from assumptions rather than mapping ledger transactions to factors. That produces a number no auditor will accept.
• Category cherry-picking. They report the three easy categories, skip the other 12, and cannot justify the exclusion when assurance reviews start.
• No audit trail. Every ton needs to trace back to a source document. If your inventory lives in a spreadsheet where numbers were typed in, you have a report, not an inventory.

The good news: fixing all three does not require a climate scientist. It requires accounting discipline.

Which Scope 3 categories actually apply to a mid-market B2B company?

The GHG Protocol lists 15 categories. Most B2B services and software companies materially report on 5 to 7 of them. The rest need documented exclusions.

The point is not to report on 15 categories. It is to explicitly evaluate 15 and document the material ones.

How do you go from ledger data to a footprint?

The mechanism has three steps.

1. Extract every spend line. Pull a full year of general ledger transactions from NetSuite, QuickBooks, or your ERP. You want vendor name, GL account, amount, and date at minimum.
2. Map each GL account to an emission factor. Published datasets like EPA USEEIO or Exiobase provide kg CO2e per dollar for hundreds of sector categories. Your marketing agency spend maps to advertising services. Your AWS bill maps to data processing. The mapping table is roughly 40 to 80 rows for a mid-market company.
3. Multiply and sum. Emissions per line equal spend multiplied by the factor. Aggregate by category. That is your spend-based baseline.

A first-pass inventory using this method takes about two weeks. It will not be your final answer, but it will be defensible and it will tell you where to invest deeper.

When should you upgrade a category from spend-based to activity-based?

Spend-based is a starting point, not an endpoint. Upgrade to activity-based data when a category meets two conditions: it represents more than 10% of your total footprint, and activity data exists in a system you already use.

• Cloud compute. AWS, GCP, and Azure now publish account-level emissions dashboards with usage-based factors. Upgrade this on day one if you are a SaaS company.
• Business travel. Travel platforms like Navan, TravelPerk, and SAP Concur expose itinerary data. Flight miles and hotel nights beat spend estimates by a wide margin.
• Purchased goods. Only upgrade the top 10 suppliers by spend. Send them a targeted questionnaire, not a generic one.
• Employee commuting. Run a one-time survey. Refresh every two years unless work patterns change materially.

Upgrading everything is not the goal. The goal is a footprint where the material categories are activity-based and the immaterial ones are spend-based with a documented method.

What does an audit trail look like in practice?

Every ton of CO2e in your final report needs to link to (1) a source document, (2) an emission factor with a citation, and (3) a calculation that a stranger can reproduce. That is the standard.

In practice, this means your inventory sits on top of a table where each row has: transaction ID, vendor, amount, GL account, factor source (e.g., USEEIO 2022 rev A, sector 541800), factor value, method (spend or activity), and calculated tCO2e. When an auditor asks "how did you get 412 tons for advertising", you filter the table and export.

Spreadsheets can do this at small scale. Past 5,000 transactions it becomes a full-time job. Which is why finance teams doing this at scale run it out of a system connected to the ledger, not out of a workbook.

Who owns the Scope 3 inventory internally?

Three roles, all named. If any is blank the inventory stalls.

• Data owner. Usually the controller or FP&A lead. Owns the extraction from the ERP, the mapping table, and the utility bill collection.
• Methodology owner. Whoever is most senior on sustainability, even if that person also runs facilities or ops. Owns category inclusion decisions and the methodology appendix.
• Assurance owner. The CFO or head of finance. Owns sign-off on the final report and the relationship with the assurance provider.

For companies without a dedicated sustainability hire, the methodology owner is often the head of ops or a strategic finance lead. That works. What does not work is treating this as a rotating side project.

The mistake to avoid

Do not treat Scope 3 as a science project. Treat it as an annual close: same data sources, same rigor, same audit trail expectations. The companies that make this cheap and repeatable are the ones that stopped waiting for perfect data, mapped the ledger to factors they can cite, and upgraded the categories that mattered. Everything else, including hiring a full sustainability team before you have a baseline, is a way to spend more money later.