Framework / 05

How the data gets built.

A prototype pipeline: ingest published membership sources, enrich from the firms' own websites, geocode, then tag capability. Every field keeps its provenance.

Firms mapped

345

100% coverage

Websites known

309

90% coverage

Geolocated

249

72% coverage

Email captured

142

41% coverage

Phone captured

174

50% coverage

01

Source ingest

Membership listings from Professional Liverpool, Liverpool Chamber, Liverpool Law Society and ICAEW are consolidated into one firm record, keeping every source it appeared in.

02

Website enrichment

Each firm website is fetched and parsed for a UK postcode, contact email, phone number and meta description. This is the layer that becomes a scheduled scraper.

03

Geocoding

Extracted postcodes are resolved to coordinates and local authority via the Postcodes.io API, filtered to Liverpool City Region postcode districts.

04

Capability tagging

Firms are classified into a three-level taxonomy — sector, core service, specialist service — and mapped onto plain-language customer needs, with each link marked confirmed or likely so gaps stay visible.

05

Next: deep scrape

Crawl service, team and sector pages to confirm capability, headcount, accreditations and sector focus; re-run on a schedule and diff against the previous snapshot.

Known gaps in this prototype

  • · 96 firms have no verified postcode yet, so they are listed but not plotted.
  • · Capability marked "Potential" is inferred from sector, not confirmed on the firm's website.
  • · Enrichment ran once as a snapshot; scheduled re-crawls and change tracking come next.
  • · Membership lists change; source records should be re-ingested each quarter.