Carbon accounting is only as good as the data behind it. A well-structured emissions inventory can still mislead if the numbers feeding into it are incomplete, inconsistent, or unverified. Before you can report credibly — whether for BRSR compliance, an export customer, or your own net-zero strategy — you need to understand both what data you need and why its quality determines whether your numbers can be trusted.
This article covers both questions together, because they are inseparable in practice.
The Five Core Data Categories
Every emissions inventory, regardless of business size or sector, draws from the same five broad categories of activity data:
1. Energy Consumption
This is typically the largest source of emissions for most businesses. It includes:
- Electricity — monthly bills from your discom, in kWh. If you have multiple meters at multiple facilities, collect separately.
- Diesel and petrol — litres consumed in generators, boilers, and owned vehicles. Pull from purchase invoices or fuel logs.
- LPG or PNG — volume (kg or cubic metres) consumed in heating, cooking, or industrial processes.
- Coal or biomass — if your facility uses solid fuel, volume in tonnes with moisture content where possible.
2. Travel and Transport
This covers emissions from movement — both your own operations and your people:
- Company fleet — kilometres driven by vehicle type (passenger car, LCV, truck). Your fleet manager or vehicle logbooks are the primary source.
- Business air travel — passenger-kilometres or flight routes, pulled from travel booking systems or expense claims.
- Employee commuting — a Scope 3 category. Collected via employee surveys, typically averaged across your headcount and location.
- Freight and logistics — tonne-kilometres shipped, sourced from logistics providers or dispatch records.
3. Fuel and Refrigerant Consumption
Refrigerants are a frequently overlooked but significant source for businesses with large air conditioning or cold chain equipment. Track refrigerant top-up quantities from your service records; HFCs commonly used in AC systems have a global warming potential hundreds of times that of CO₂. Even small leaks matter at scale.
4. Waste Generated
Waste going to landfill generates methane, a potent greenhouse gas. Data sources include:
- Waste contractor invoices or weighbridge records (tonnes by waste type)
- Internal waste tracking logs if you have them
- Estimates based on headcount and industry benchmarks if no contractor data exists
5. Procurement and Supply Chain
Scope 3 from purchased goods and services is often the largest slice of a company’s total footprint — sometimes 70–90% for consumer-facing businesses. Data here is harder to collect: it requires either supplier-specific data (direct emissions per unit purchased), industry-average emissions factors applied to spend, or physical quantity data (kg of raw material) multiplied by material-specific factors.
For most Indian SMEs in their first year of carbon accounting, Scope 3 procurement is best tackled with spend-based estimation — imperfect but a reasonable starting point before supplier data programmes mature.
Why Data Quality Matters
Collecting data is the first step. Ensuring it is good quality is the second, and often underestimated, challenge. Here is what quality means in practice:
Completeness
Are all sites, all vehicles, all fuel types accounted for? A common gap is multi-site businesses that collect data from headquarters but miss remote facilities, depots, or rented spaces. Incomplete coverage means your reported number is lower than reality — which is not just inaccurate, it is misleading if taken at face value by an external audience.
Consistency
Are you measuring the same things the same way each year? If you switch from purchase-based diesel tracking to fuel-card data mid-series, your year-on-year comparison becomes unreliable. Consistency in methodology matters as much as consistency in the raw numbers.
Accuracy
Are the numbers derived from actual measurements, or from estimates? Meter readings are more accurate than engineering calculations. Invoices are more accurate than memory. The GHG Protocol quality hierarchy runs: measured data at the top, then calculated (from engineering models), then estimated (from proxy data or averages), with each step down carrying more uncertainty.
Transparency and Traceability
Can someone else follow your calculation from raw number to final tCO₂e figure? If you are audited — by a third-party verifier, a supply-chain partner, or a stock exchange compliance team (SEBI now requires BRSR disclosures for the top 1,000 listed companies in India) — you need to show your working. Every estimated figure needs a source or method noted. Every emission factor used needs a reference.
Practical Tips for Improving Data Quality Over Time
- Start with what you have. Year 1 data will have gaps and estimates. That is expected and acceptable — flag them clearly and set a plan to fill them.
- Assign ownership. Someone in finance or operations needs to own data collection for each category. Unassigned tasks produce missing data at reporting time.
- Set up a data calendar. Map out which data arrives when (monthly electricity bills, annual fuel invoices, quarterly freight summaries) and build collection triggers around those dates.
- Prefer primary data over proxies. If you can get an actual number from a supplier or service provider, use it. A proxy or industry average should be a fallback, not a default.
- Document your assumptions. Where you estimate, record the estimation method, the source of any benchmark used, and the date. This protects you in an audit and lets you replace estimates with actuals in future years.
The Connection to Credibility
Data quality is ultimately a credibility question. A sustainability report built on well-documented, internally consistent, annually verified data carries weight with investors, customers, and regulators in a way that a loosely assembled estimate does not.
In India, this is becoming increasingly consequential. SEBI BRSR requirements, export customer due-diligence requests, and the EU CBAM framework all converge on the same expectation: show your data, show your methodology, and be prepared to have it verified. Businesses that build good data practices early will meet those demands without a scramble.
High-quality data is what separates a carbon number that can be defended from one that merely looks credible on paper. Invest the time to build your data infrastructure well, and every subsequent year of reporting becomes faster and more trustworthy.