ESG data: how to collect and manage it simply and reliably
Utility bills, invoices, waste forms and payslips already contain almost all the ESG data you need. Here is how to collect it with a simple method, check it and reuse it for reports, banks and clients.
Collecting reliable ESG data does not require complex tools: you need a clear list of the data, a source document for each item, an owner and a few checks repeated every year. Almost everything that banks, clients and the VSME standard ask for is already in your utility bills, invoices, waste forms and personnel records. Here is a practical six-step method.
Which ESG data you really need
The first mistake is starting from the questionnaires, one at a time. Each request uses different wording, but the questions almost always revolve around the same data. It is better to start from a standard and build your list there. For an SME the most suitable reference is EFRAG’s VSME, whose Basic Module covers energy, emissions, pollution, water, waste, workforce and business conduct.
For each data point you always need four pieces of information: the value, the unit of measurement, the period and the document it comes from. Here is the starting map:
| Data | Unit | Source document | Who usually has it |
|---|---|---|---|
| Electricity | kWh | Bills, for each supply point (POD) | Administration |
| Natural gas | Smc | Bills, for each gas delivery point (PDR) | Administration |
| Vehicle fuel | litres | Fuel invoices, fuel card reports | Administration or fleet |
| Waste produced | kg | Waste identification forms (FIR), waste register | Environmental manager |
| Water withdrawn | m³ | Water bills, meter readings | Administration or maintenance |
| Workforce | number, hours | Payroll register, payslips, training records | Human resources or payroll consultant |
| Injuries | number, days | Internal register, reports to INAIL | Health and safety manager |
| Governance | yes or no, text | Articles of association, code of ethics, policies, minutes | Management |
Assign sources and owners
A data point without an owner tends to disappear the following year. For each row of the table decide who provides it, by when and in what form. A few rules are enough:
- one owner per data point, not per area: whoever receives the bills uploads the bills;
- a shared folder with a fixed structure, for example year, site, document type;
- a coordinator who gathers everything, checks completeness and acts as the point of contact for anyone asking for the data;
- an internal deadline ahead of the external ones: requests from banks and investors often arrive in the first quarter.
For social and governance data the work is different: it changes little from one year to the next, but it is spread across different people. The payroll consultant has headcount, contracts and hours worked, the health and safety manager has injuries, management has policies and the code of ethics. Ask each of them for an extract in the same format every year, so the comparison with the previous year is immediate.
If the company has several sites, add a site column to every data point from the start. You will need it to explain changes, for client questionnaires asking for data by plant and to assess climate risks site by site.
Collect from documents, not estimates
The most useful rule is also the simplest: every number must be traceable to a document. Estimates should only be used when the data is genuinely missing, and must be declared as such.
Some practical points, document by document:
- Bills: the billing period often does not match the calendar month. Allocate consumption over the actual days and distinguish actual readings from the supplier’s estimated ones, which will be adjusted later.
- Fuel invoices: record the litres, not just the euros. The price changes; the quantity is what you need for emissions.
- FIR: use the weight verified at destination when available and note the EWC code and the destination operation, recovery or disposal. With RENTRI and the digital FIR this data becomes easier to extract.
- Workforce: set a reference date, for example 31 December, and a single criterion, headcount or full-time equivalents.
It is also the step that takes the most time. With Envify you upload bills, invoices and FIR as they are: the AI reads quantities, periods and sites and keeps the link to the source document, so every number remains verifiable.
Quality checks you can do in an hour
Before using the data, run four checks: completeness, consistency of units, plausibility and traceability. These are the most common errors and how to catch them:
| Typical error | How to spot it | How to fix it |
|---|---|---|
| A month missing on a meter | Count the months covered for each POD and PDR: there must be twelve | Retrieve the missing bill or the data from the supplier’s portal |
| Mixed units | Look for values in MWh among the kWh, or in tonnes among the kilos | Convert everything to the reference unit before adding up |
| Double-counted consumption | Compare the total with the previous year and with spending | Remove correction bills already counted and duplicate adjustments |
| Unexplained changes | Flag large deviations from the previous year, site by site | Check whether production, opening hours, sites or supplier have changed |
| Data without a source | Check that every value has a linked document | Reconstruct the source or mark the value as an estimate |
Always record corrections and the reason for them. If one day a client, a bank or an auditor asks how a number was obtained, the answer will already be written down.
From data to indicators
Raw data becomes indicators with a few calculations, as long as the method is stated. Greenhouse gas emissions, for example, are obtained by multiplying consumption by an emission factor according to the GHG Protocol: gas and diesel go into Scope 1, purchased electricity into Scope 2. You will find a full example in how to calculate emissions from a utility bill.
Three precautions make indicators comparable over time:
- Same scope: state which sites and activities are included, and flag any change.
- Same factors: use recognised sources and note the year of the source, updating it explicitly.
- Intensity indicators: alongside absolute values, add a ratio to turnover, production or employees, so that company growth does not look like a deterioration.
Keep raw data and calculation results separate. If next year an emission factor changes or you correct a bill, you can recalculate the indicators without collecting the documents again.
Set up the process once, reuse it every year
The heaviest work is in the first year. From the second year, with the same list of data, the same owners and the same checks, collection becomes a routine activity. A simple calendar works like this:
- every quarter: upload the bills and invoices for the period and check for missing months;
- at year end: close the workforce, waste and governance data;
- in the first quarter: calculate the indicators, compare them with the previous year and prepare the report;
- all year round: answer external requests from the same base, without collecting everything again.
A single database is also the most effective answer to bank ESG questionnaires and client requests: the questions change, the numbers do not. The VSME does not require external assurance, but traceable data makes the report ready for a voluntary check too.
An SME’s ESG data is already in everyday documents: utility bills, invoices, FIR and personnel data. To make it reliable you need a list based on a standard such as the VSME, an owner for each data point, a link from every number to its source and a few checks repeated every year.
