The category most companies guess their way through
Employee commuting is category 7 in the climate accounts. It is also the category most often filled in with an assumption, because the numbers do not sit in any system you already run. Here is how it is actually calculated, what it takes for the number to survive a follow-up question, and when it applies to you at all.
When it applies, really
Employee commuting is explicitly Scope 3 category 7 in ESRS E1, which follows the GHG Protocol's division into fifteen categories. The Scope 3 breakdown is required under datapoint DR E1-6.
But the requirement is doubly conditional, and that is worth knowing before anyone sells you something on it. First, the climate standard as a whole applies only where climate is material for the company. ESRS expects a detailed explanation from a company that concludes climate is not material, so the bar is high in practice, but the assessment is yours. Second, separate reporting is required only for the Scope 3 categories that screen as material. Commuting has to pass both tests.
That means commuting emissions are never a freestanding legal requirement you cannot escape. For an office-based company where commuting is a significant share of the footprint, it will often be material. But you decide that through the materiality assessment, and a vendor who claims otherwise is overselling.
After the Omnibus simplification, the reporting duty itself applies to companies with more than 1,000 employees and more than 450 million euro in turnover, with the new thresholds applying from financial year 2027. Smaller companies meet the requirements indirectly, as suppliers to the large ones who ask for numbers.
How to calculate employee commuting emissions, step by step
The GHG Protocol allows several methods. The one that is practically feasible for commuting is the distance-based method: you ask employees how far and how they travel, and calculate the emissions from the answers. The calculation per employee is simple enough to stand in a methodology chapter.
Car days per week, times two, because a commute goes both ways. Times the one-way distance, taken from the interval the employee chose. Times the number of working weeks in a year. Times the emission factor for the vehicle.
The factor depends on what your country publishes. The Netherlands maintains a national emission factor list with a fleet average for cars; Sweden publishes no fleet average, so there the factor has to come from the respondent's own answer about fuel. A report should name its factor source and the date it was checked, whichever country it is for.
The sum across all respondents, scaled to the whole workforce, is category 7. What separates a number that holds from one that does not is not the formula. The formula is the same everywhere. It is the data underneath it.
Where the estimate fails
The most common approach today is a generic estimate: headcount, times an assumed average commute, times a national emission factor. That produces a number, and the number is not illegal. But it has two problems.
The first is that nothing you do can change it. Put up bicycle parking, or move the office to a transit hub, and the estimate stands still, because none of its inputs measure anything that changed. You cannot show the effect of a measure with a number that does not know about the measure.
The second is that it does not survive a follow-up question. Asked how the number was produced, the answer is that it was assumed. An auditor may accept that. A large customer asking for supplier data may not, and CSRD tightens the requirements for Scope 3 data quality from financial year 2027 for the largest companies.
A measurement solves both at once, and that is the whole reason we build what we build.
What makes the number hold
The response rate is the single most important factor, and also the one most often ruined by execution. A survey answered by twenty percent measures, in practice, the twenty percent who had something to say, not the company.
Anonymity is not a courtesy promise here, it is a precondition for the response rate. Employees who believe the employer can see who drives either do not answer, or answer prettier than the truth. Both ruin the number. In our surveys anonymity is a technical property: no IP address, no timestamps finer than a date, and no free text are stored with an answer, and groups below five people are never reported.
And the method has to travel with the document. An emissions number without a response rate, without the date the factors were checked, and without a note on what was left out, is a claim. With those three it is a calculation someone else can verify.
VSME: when a bank or a large customer asks a smaller company
Most companies that get asked about commuting emissions in the next year will not be in scope of CSRD themselves. They will be asked by a bank, an insurer or a large customer, and the format of the ask is increasingly the voluntary SME standard VSME (renamed VSRS in the July 2026 revision; the market still says VSME). Its basic module requires only Scope 1 and 2; Scope 3, including commuting, is voluntary to report. But when a services company does report Scope 3, commuting is one of the few categories it actually has.
The good news for a smaller company is that the standard is voluntary in form: nothing forces a commuting figure into the report. But if a figure goes in, it has to hold. The distance-based method above is the same regardless of which framework asks, so one measured commuting dataset answers the bank, the customer questionnaire and your own footprint at once. We deliver the category figure, never the whole inventory: whether commuting matters for you depends on your other categories, which only you can weigh.
One dataset, two duties: the Dutch example
Some countries add a national reporting duty on top of the climate accounts, and the Netherlands is the clearest example: employers above the WPM threshold must report commuting and business travel kilometres per mode of transport every year.
The useful part is that a properly designed survey serves both documents at once. The same answers that produce category 7 also produce the commuting part of a WPM report: kilometres per mode, car kilometres split per fuel, and lease kilometres separated out, with the method choices documented in the annex. Measure once, use twice. Business travel is not included in that: it lives in your own travel and lease systems.
The same logic holds anywhere in the EU and EEA: the framework is shared, the factors are national, and a baseline measured properly in one country is comparable across borders.
Estimate your commuting footprint first
See what you get before you ask for anything: the example report shows the methodology chapter and the numbers on real pages. A quote takes one business day and commits you to nothing.
Sources and status
| Claim | Source | Last checked |
|---|---|---|
| Commuting is Scope 3 category 7, required under DR E1-6 | ESRS E1 (EFRAG); GHG Protocol Scope 3 Standard | 12 Aug 2026 |
| The requirement is conditional on materiality, and on the category screening as material; a company concluding climate is not material is expected to explain that conclusion in detail | ESRS E1, materiality provisions | 12 Aug 2026 |
| CSRD thresholds and first year after Omnibus; tightened Scope 3 data quality from FY2027 | CSRD as amended by Omnibus (Directive (EU) 2026/470) | 10 Aug 2026 |
| Dutch national emission factor list with a fleet average for cars | co2emissiefactoren.nl, maintained by Rijkswaterstaat and Stichting Stimular | 11 Aug 2026 |
| Sweden publishes no fleet average; emissions are calculated per fuel | Swedish EPA (Naturvårdsverket) climate calculation tool v9 | 9 Aug 2026 |
| The Dutch WPM duty requires commuting kilometres per mode of transport annually | Besluit activiteiten leefomgeving articles 18.11 and 18.15; RVO | 18 Aug 2026 |
| VSME, the voluntary SME sustainability reporting standard (upgraded to VSRS in the revised ESRS package, July 2026) | EFRAG | 21 Aug 2026 |