A data quality framework in GHG Accounting is a systematic approach that defines how activity data is collected, classified, verified, and documented. The goal is to ensure emissions calculations are traceable, consistent, and robust enough to withstand assurance or audit.
What Does Data Quality Mean in GHG Accounting?
Data quality in GHG Accounting is not just about producing a “correct number.” The real objective is to ensure the data is:
- Source-clear (you can identify the system/document/measurement it came from)
- Consistent (the same approach across years and sites)
- Traceable and repeatable (an assurer/auditor can reproduce the same result)
A practical test for sustainability teams:
“If someone opens this dataset six months from now, can they follow the same method and reach the same result?”
Why Is Data Quality Increasingly Critical?
Carbon data is no longer only a reporting exercise—it’s a compliance and commercial-risk issue. Common drivers include:
- Growing expectations for independent assurance (e.g., evidence-based validation/verification logic aligned with standards like ISO 14064-3)
- Expanding regulatory and buyer requirements across markets and value chains
- Higher scrutiny on the evidence chain behind numbers—not just the math
Expert insight: Auditors often look at this before asking “Is the calculation right?”
“Is there an evidence chain for this data?” (invoice → ERP record → meter → methodology note → calculation file)
What Is Tier Quality (Data Quality)?
A data quality (tier) framework ensures data used in a carbon inventory is managed using a standard logic. It provides clear answers to questions like:
- Is the data direct (meter, invoice, measurement) or indirect (estimate, proxy)?
- Is it measured or modeled/estimated?
- Is the value company-specific or based on industry/regional averages?
- Is it supported by documents/evidence?
- Can it be reproduced next year using the same method?
Without this framework, you may produce a report—but you won’t build a scalable system. And without a system, audits become unpredictable.
Data Quality Dimensions in the GHG Protocol Approach
In the GHG Protocol approach (and good practice generally), data quality is multi-dimensional—not a single score.
1) Accuracy
How well the activity data reflects real operations.
- Electricity data based on invoices → higher accuracy
- Energy estimated from total spend → lower accuracy
2) Completeness
Are all sites, processes, and activities included?
A common pitfall: excluding “small” sites or activities. This often triggers boundary and completeness questions during assurance.
3) Consistency
Is the same methodology applied the same way across years, sites, and activities?
Field reality: in assurance, consistency is often treated as more critical than accuracy, because inconsistent methods break trend analysis and performance tracking.
4) Transparency
Are assumptions, estimates, gaps, and limitations clearly documented?
“We did it this way” is not a justification.
Justification = methodology note + evidence + rationale for the chosen approach.
Tier Structure: Grading Levels of Data Quality
What Is a Tier?
A tier (also seen in IPCC-style approaches) classifies data by reliability and specificity. Tier is not simply “good vs bad”—it’s about fitness for purpose.
| Tier | Description | Typical use case |
|---|---|---|
| Tier 1 | General, average data | First-year baselines / quick view |
| Tier 2 | Adapted to region/sector | Routine reporting / buyer info requests |
| Tier 3 | Company-specific, measurement-based | Audit-ready reporting / high confidence |
A Common Misunderstanding
Higher tier is not always mandatory. But auditors will ask:
“Why did you choose this tier?”
For material sources (e.g., fuels, process emissions), moving toward Tier 3 typically reduces risk
Limited access to granular data → Tier 1/2 may be reasonable
Most Common Data Quality Problems
The most frequent (and most time-consuming) issues in practice:
- Invoices stored in different units (kWh, currency, m³)
- Missing or non-responsive supplier data (especially Scope 3)
- Manual, Excel-based conversions (error risk + version confusion)
- Methodology changes between years without documentation
- Evidence (invoices/delivery notes/meter exports) not linked to the reported data
Key point: most of these are not “reporting mistakes”—they’re data governance gaps.
The Direct Link Between Data Quality and Audit Readiness
Assurers typically start with:
“Where did you get this data, and did you collect it the same way last year?”
Without a data quality framework, common outcomes include:
- More revision cycles
- Longer assurance timelines
- Lower confidence (especially when inconsistencies appear)
That’s why data quality is not “nice to have”—it’s a prerequisite for audit-ready reporting.
How to Build a Strong Data Quality Framework
Practical steps (technology-agnostic)
- Clearly define data sources (invoices, ERP, meters, production systems)
- Assign a tier level for each data type and document the rationale
- Document assumptions (e.g., how missing months are backfilled)
- Record methodology changes over time (a change log)
- Link documents to data (evidence URL/path/reference number)
- Apply a simple control: “Is it reproducible?” (can someone else reach the same result with the same files?)
Expert insight: Start with the most critical 20% of data, not “perfect data.”
In most companies, the majority of emissions sit in a few categories (electricity, fuels, process emissions, core logistics). Strengthening tier level and evidence for these categories can dramatically reduce audit risk.
Data Quality Is Foundational Infrastructure
GHG Accounting is not a one-off report—it’s an ongoing, repeatable, auditable operation.
Without a data quality framework, your setup:
- Won’t scale
- Won’t withstand assurance
- Won’t support readiness for mechanisms that require disciplined, traceable data
FAQ
Yes — but if the evidence chain is weak, assurance cycles, revisions, and risk tend to increase. A framework makes reporting sustainable and repeatable.
Not for every line item. But for high-impact categories (electricity, fuels, process emissions), higher-tier data usually reduces assurance risk.
Both matter, but assurance often checks consistency first. Without it, year-on-year comparisons and trends are unreliable.
You can, but without version control, evidence linking, and a change log, error and assurance risk rises. Excel is a tool; the framework is the discipline.
