How we found the UK's 100 fastest-growing SMEs in 7 million Companies House filings (and why 4 in 10 candidates were wrong)
Description and use
The project produced a list of 100 UK SMEs with the fastest turnover growth by parsing 7.35 million Companies House filings, applying automated filters, and manually vetting candidates.
Problem
Existing AI assistants cannot reliably identify the fastest-growing UK SMEs because the data is buried in millions of machine‑readable filings.
Outcome
The project produced a verified list of 100 UK SMEs with the fastest turnover growth, revealing that 40% of raw candidates were false positives and that construction firms dominated the top 100.
Workflow
- 1. Stream each 2–4.4 GB zip, parse XHTML for turnover tags, honour scale, sign, format, keep GBP, skip consolidated contexts.
- 2. Extract current and prior periods, yielding two years per filing and three years of turnover per company.
- 3. Store facts in Parquet, query with DuckDB to build one row per company per year, preferring current over comparative data.
- 4. Apply rules: 12‑month years, consistent year‑end, rising turnover, SME thresholds, no >2.5× jump, exclude 2025 declines >30%.
- 5. Manual review of 260 candidates to remove subsidiaries, insolvencies, name changes, etc.; 97 removed.
- 6. Use free Companies House company data to filter active, non‑micro, non‑dormant, no name change, and PSC checks to exclude foreign or high‑turnover corporate controllers.
- 7. Publish data as CLI that outputs JSON; create a skill for Claude to invoke the CLI and format results.
- 8. Provide GitHub repo, CSV, and ranking URL for users.
Evidence
Companies House publishes a monthly zip of every electronically filed set of accounts. We processed 24 months (January 2024 to December 2025): 7,352,908 filings
Only 162,700 filings (2.2%) report turnover at all. Small and micro companies are allowed to file without a profit and loss account, so this is a ranking of medium-sized and larger SMEs by construction.
260 checked, 97 thrown out.
The useful lesson: most of this can be caught automatically once you know to look. Three rules from the free Companies House company data product (status exactly "Active", account category not micro, dormant or audit-exempt subsidiary, and no name change since 2023) caught all 10 of the insolvency cases we had found by hand, without losing a single company we had confirmed.
- Record date: 2026-10-14T14:56:10.304Z