How often must an Indian plant get its coal analysed for UK CBAM?
If your monitoring plan uses lab analysis rather than standard values, HMRC's technical guidance sets a minimum for coal, coking coal, coke and pet coke: every 20,000 tonnes and at least six times a year. HMRC's own example: a plant burning 500,000 tonnes a year needs at least 25 lab samples, each prepared under a documented sampling plan.
When the frequency rule applies
HMRC lets a plant set calculation factors (emission factor, net calorific value, carbon content and so on) in several ways, listed in rising data quality: standard values, correlations, the supplier's analysis printed on purchase documents, analysis in a non-accredited lab or with simplified sampling, and analysis in an accredited lab with best-practice sampling. The minimum frequencies apply when you analyse. Each result is used only for the batch it was sampled from.
Minimum analyses, HMRC Table 4-1
- Coal, coking coal, coke, petroleum coke, peat: every 20,000 t and at least six times a year.
- Fuel oils, including heavy fuel oil: every 20,000 t and at least six times a year.
- Natural gas: at least weekly.
- Other fuels: every 10,000 t and at least four times a year.
- Limestone and dolomite: every 50,000 t and at least four times a year.
Analysis is not sampling
One analysis needs a sample that represents the whole 20,000 t batch, so you take many more samples than you analyse. HMRC wants a sampling plan for each fuel in your monitoring plan, covering the standards followed, method, responsibilities, locations, frequencies, quantities, how samples are stored and how they reach the lab.
When you may analyse less often
- The fuel is not in the table: HMRC points to the one-third rule, a frequency that leads to a third of the uncertainty over the whole reporting period.
- Meeting the minimum would cost unreasonably, shown with HMRC's unreasonable cost method.
- The fuel is uniform, shown by past analyses including the previous calendar year, and its values vary by no more than a third of the uncertainty that applies to its activity data.
- Whatever you change, write it into the monitoring plan, or the verifier will treat it as a gap.
Example: counting analyses for one year
- 90,000 t of imported coal: 90,000 / 20,000 = 4.5 batches, but the six-a-year floor applies, so at least 6 analyses
- 150,000 t: 150,000 / 20,000 = 7.5, so at least 8 analyses, one for each batch
- HMRC's case, 500,000 t: at least 25 analyses
90,000 t and 150,000 t are example tonnages. A batch part-way through the year still needs its own analysis.
What Senseible does here
Senseible reads coal and pet coke tonnes from your purchase invoices into a monthly record, the activity data each analysis applies to. We build the figure from your records and check every figure in-house; for the UK it must then be verified by a verifier accredited for UK CBAM. Have your own verifier? Send it to them.
- CBAM report from your records
- Using a non-accredited lab
- UK CBAM standard fuel factors
- What a UK CBAM monitoring plan holds
Last reviewed 2026-11-03. HMRC is taking feedback on this guidance until 13 November 2026 and may update it before 1 January 2027.