Methodology & Data Transparency

Last refreshed 2026-08-31 · Source: GOV.UK OZEV authorised-installer tool.

This page documents — for journalists, sponsors, partner installers and any researcher reusing our derived statistics — exactly how the 1139 listings and the figures on the UK EV installer landscape page are produced. If anything below is wrong or unclear, please tell us and we will fix it on the next rebuild.

1. Where the installer list comes from

The single upstream source is the public GOV.UK / Office for Zero Emission Vehicles (OZEV) authorised-installer tool: https://www.gov.uk/electric-vehicle-chargepoint-installers. Every business on this site is an OZEV-authorised installer that has self-declared (to OZEV) a willingness to carry out commercial installations. We do not include any installer that is not on that register.

2. Last refresh date

The data underpinning this build was last refreshed on 2026-08-31. A refresh date is also stamped into the footer of every page and into the generated_at field of installers.json.

3. Collection

The OZEV tool is queried across a UK-wide postcode grid (covering all 12 nations / English regions). Requests are rate-limited and respectful (the collection runs at a fraction of the throughput a human user would generate) and use a UA string identifying this project so OZEV / GDS can contact us if needed.

4. Normalisation

Raw OZEV records are normalised at build time:

• Trading name is collapsed to a single canonical string (whitespace, casing, trailing "Ltd" / "Limited" variations).
• Postcodes are uppercased and re-spaced to the standard UK format.
• Towns are mapped to one of 12 UK regions using ONS region boundaries; an installer whose postcode does not resolve is flagged N/A rather than guessed.
• Phone numbers are E.164-normalised and the primary contact number is preferred over fax / personal mobile where multiple are present.
• A stable, collision-safe URL slug is assigned per installer and reused across rebuilds.
• Records are de-duplicated on a (normalised name + postcode) key.

5. Accuracy rule — no fabrication

If a field is not present in the OZEV source, it is shown as "Not listed" or omitted entirely. We never infer, guess or fill in a missing website, phone number or service category. Stats and counts shown on the data landscape page are computed directly from the normalised dataset and nowhere else.

6. How updates happen

The pipeline (pipeline/extract.py → site/generate.py) runs automatically on a weekly cadence. Each run regenerates data/installers.json and rebuilds every page from scratch — no manual editing of listings is possible after publish. The refresh date in the footer is set programmatically from the build clock.

7. Known limitations

• OZEV does not sub-categorise commercial installers. The register flags whether an installer offers commercial work, but does not split that into "fleet depot" vs "workplace" vs "public destination". Our service-pages (e.g. fleet charging installers) all draw from the same single commercial-flagged pool. Buyers should still confirm suitability with each installer directly.
• OZEV authorisation reflects what an installer is permitted to do, not their current workload, project portfolio or geographic operating area. A small installer on the register may not actually serve every postcode in their nominal region.
• The OZEV register is itself self-declared (with light spot-checking). Where a listing looks inaccurate we accept correction requests via the contact page; we do not overwrite OZEV's data, but we can suppress an entry.
• Cost figures and grant rates quoted in our guides come from GOV.UK / OZEV / Ofgem / IET publications and are cited inline; they are not derived from this dataset.

8. Reporting an error

Spot a mistake — wrong town, dead website, listing that shouldn't be here? Email commercialevinstallers@gmail.com or use the contact page. Corrections and removals are actioned on the next rebuild, with no questions asked.

9. Data licence

The upstream OZEV data is reused under the Open Government Licence v3.0. Our derived statistics, normalisations and counts (everything on this site that is not the raw OZEV record) are released under the same OGL v3.0 — free to reuse with attribution. Suggested citation: "Commercial EV Charger Installers UK, derived from GOV.UK OZEV authorised-installer data under OGL v3.0, refreshed 2026-08-31".

10. Privacy stance

We deliberately do not publish scraped personal (firstname.lastname@) or free-webmail email addresses as clickable links — a privacy choice taken under UK GDPR (lawful basis: legitimate interests, balanced against the limited and already-public nature of the OZEV business data) and PECR. Where only a personal email exists in the source, we route enquiries via a generic "request a quote" flow instead. Full detail is on the privacy & data page.

11. Data schema

The public dataset is published as JSON at /data/installers.json. The per-record shape is:

{
  "name":      "Example EV Installations Ltd",
  "town":      "Manchester",
  "region":    "North West",
  "postcode":  "M1 1AA",
  "website":   "https://example.com",
  "phone":     "+441611234567",
  "email":     "info@example.com",          // generic mailbox only
  "services":  ["Commercial", "Residential"],
  "featured":  false,
  "_slug":     "example-ev-installations-m1"
}

Top-level payload also includes count, generated_at (ISO 8601) and licence.

12. Downloads

• installers.json — full public dataset (UTF-8, ~1139 records, OGL v3.0).
• sitemap.xml — every public URL on this site.

Methodology version: 2026-08-31. Pipeline source: pipeline/extract.py, site/generate.py. This page is rebuilt automatically on each refresh — any inaccuracy is fixable; please flag it.