How the reclaim data pages are built
Every BorderAudit reclaim data page reports aggregated, anonymised findings from post-clearance customs audits. This page sets out exactly how those figures are produced.
What the data is
Each reclaim data page reports on a slice of BorderAudit's post-clearance audit work — for example a sector, a relief type or a duty category. The figures are aggregated across the audits in that slice. They describe recoverable duty and import VAT identified by the audit, and are never single-client values.
Which importers the figures describe
These figures describe the declarations we audited — not UK imports as a whole. That distinction matters, so we would rather state it than leave it implied.
Businesses come to us because they suspect something is wrong. That makes our sample self-selected rather than a random draw of UK importers, and it means the error and overpayment rates we report are very likely higher than the rate across all UK declarations. We report what we found in the work we did, and we do not extrapolate it to the wider population.
Sample size, dates & when the analysis was done
Every page states the number of underlying audits, how many distinct businesses they are drawn from, the date range of the declarations, and when we carried out the analysis. Customs data is necessarily historic — a post-clearance audit can only examine entries that have already cleared — so the declarations are older than the analysis. Both dates are always given.
- Duty Reclaim in Fashion & Retail Imports — 149 audits across 31 businesses; declarations from Jan 2022 – Dec 2024, analysed in 2025/26 (illustrative sample — not yet published)
- Duty Reclaim in UK Manufacturing Imports — 52 audits across 14 businesses; declarations from Jan 2022 – Dec 2024, analysed in 2025/26 (illustrative sample — not yet published)
Anonymisation
No client is identifiable from any figure. The floor is set on distinct businesses, not audit count — a slice drawn from a large number of audits but only a handful of clients would still be identifiable, so we do not publish it. Slices below the floor are excluded automatically rather than by review.
“Recovered” (money HMRC has paid out) and “identified” (surfaced by the audit but not yet settled) values are never blended, and each page states which it is reporting.
Rounding & disclosure
Headline currency figures are rounded to one significant figure. Percentages and counts are reported as ranges rather than exact single-client values. Where a figure comes from a specific verified case rather than an aggregate, the page says so and does not round the case amount.
Sample data notice
Some pages currently render illustrative sample data while the template is being built and reviewed. Those pages are not indexed and are excluded from the sitemap until they are backed by reviewed, anonymised data. Sample pages carry a visible banner.