Sustainability data quality affects every calculation, comparison and decision built on environmental information. Yet waiting for a perfect dataset can prevent organisations from beginning at all. A more realistic strategy is to identify material weaknesses, document limitations and improve the information through a controlled cycle.
Having no estimations does not mean that the data is high quality. A combination of supplier-specific, activity-based, spend-based, and proxy data is often needed for scope 3 inventories. The key questions are whether the selected data is appropriate for the intended use, whether its limits are clear, and whether there is a strategy to enhance priority areas.
The GHG Protocol's Scope 3 guidance describes data quality through technological, temporal and geographical representativeness, completeness and reliability. These criteria provide a useful structure beyond Scope 3 because they test whether information reflects the activity being reported.
Environmental data quality consequently depends on context, not just whether a field has been completed. For instance, an emission factor may originate from a reliable source but relate to a different region or technology, or a supplier figure may be specific to the organization but cover only a portion of the applicable period.
Good information should also be traceable. A reviewer should be able to see the source, unit, period, method, factor and relevant assumptions. Clear records make it possible to correct errors without losing the history of what was previously reported.
An organisation cannot judge carbon data accuracy without knowing how the information will be used. A high-level screening exercise may reasonably use secondary data to identify likely hotspots. A reduction programme aimed at a specific supplier or facility usually needs more representative primary information.
The GHG Protocol advises companies to prioritise high-quality primary data for high-priority Scope 3 activities, while recognising that secondary data may sometimes be more suitable or more available. This supports an iterative approach: estimate the full picture, identify material categories and then improve the information where it will influence decisions.
Applying equal effort to every record is rarely efficient. A minor source with a dependable estimate may present less risk than a major category based on an old or poorly matched factor.
Incomplete sustainability data often reflects an ownership problem. A central sustainability manager may be responsible for the report but not control the systems that produce energy, purchasing, travel or waste information.
The required metric or activity
The responsible data owner
The source system or document
The reporting frequency and deadline
The expected unit and format
The reviewer or approver
Known limitations and improvement actions
This makes responsibility more specific than sending a general request to a department. It also helps the organisation plan for staff changes and avoid relying on undocumented knowledge.
Inconsistency can be quickly replicated by automation, organisations should therefore agree on definitions, units, reporting periods, and naming conventions before developing integrations or bulk-upload templates.
A litre of fuel, a kilowatt hour of electricity and a pound of expenditure require different calculation routes. Records should preserve the original activity unit and show any conversion applied. Dates also matter because the appropriate emission factor may change between reporting years.
Data validation rules can then flag missing units, duplicate records, unusual values or incomplete evidence. A flag should prompt review rather than automatically conclude that the record is wrong.
The UK Government publishes company-reporting conversion factors each year. Its 2026 release was updated in July 2026 to correct blank values in the flat file, illustrating why source and version information should be retained.
Organisations should record the factor publisher, dataset, publication year, factor identifier where available and date applied. Previously approved inventories should not be silently recalculated whenever a library changes. Instead, the reporting policy should explain how factor updates and base-year recalculations are handled.
This distinction helps users understand whether an emissions movement arose from changed activity, a changed factor or a methodological decision.
False accuracy should not be used to cover up missing information. If an estimate is required, note the approach, presumptions, source, and rationale behind the choice. So that they can be prioritised for improvement, indicate which records are estimated.
The IFRS Foundation notes that Scope 3 measurement is likely to include estimation rather than direct measurement alone. Estimates can still be useful for making decisions when they are reasonable and accurately described.
A structured improvement hierarchy might move from spend-based data to physical activity data, then to supplier-specific activity or emissions information where relevant. Progress should be assessed against the purpose of the inventory, not the number of supplier questionnaires sent.
Quality checks are more effective throughout the year than immediately before publication. Monthly or quarterly reviews can identify missing sites, implausible movements and new activities while people can still resolve them.
Are all expected contributors represented?
Does the period match the reporting boundary?
Are units and conversions visible?
Are material estimates clearly marked?
Can supporting evidence be retrieved?
Have methodology changes been approved and documented?
Do movements make sense in the context of operational activity?
Dashboards may help with focus, but judgement is still essential. Unsuitable definitions or factors may nonetheless be used in a dataset that appears full.
Data-quality work is easier to sustain when limitations become assigned actions. Record the affected metric, current method, materiality, intended improvement, owner and target date. Close an action only when the replacement information has been reviewed.
This log also provides context for leadership and reviewers. It demonstrates that the organisation understands where estimation remains and is directing effort proportionately. Avoid presenting the existence of an improvement plan as proof that the underlying data is already robust.
Sustainability data quality improves through clear ownership, consistent definitions, controlled reference data and proportionate review. Organisations do not need to wait for perfect information, but they should be honest about limitations and direct improvement effort towards the data that matters most.
Technology can make gaps, evidence and changes easier to see. It cannot determine materiality or methodological suitability without informed human input.
Tunley Environmental helps organisations measure and review environmental performance using science-based methods. VerdaMetric supports this work by bringing data, calculations and evidence into a structured system where quality issues can be identified and improved over time.
GHG Protocol, Scope 3 Frequently Asked Questions: https://ghgprotocol.org/scope-3-frequently-asked-questions-0
GHG Protocol, Corporate Standard Frequently Asked Questions: https://ghgprotocol.org/corporate-standard-frequently-asked-questions
UK Government, Greenhouse gas reporting conversion factors 2026: https://www.gov.uk/government/publications/greenhouse-gas-reporting-conversion-factors-2026
IFRS Foundation, Greenhouse Gas Emissions Disclosure requirements applying IFRS S2: https://www.ifrs.org/content/dam/ifrs/supporting-implementation/ifrs-s2/ghg-ifrs-s2-educational-material.pdf