Geospatial Work Observatory.
A six-week research project turning the Geospatial Careers evidence base into a first State of Geospatial Work release for the UK and Europe: roles, requested skills, salary evidence and geographic demand, each reported with the limits of the data attached. This page is the project brief, published before the work, so the scope can be tested in the open.
One question, held to a quality gate.
Which role families, skills and geographic patterns can the current UK and European evidence support confidently, and which common career-market claims remain too weak to publish?
The second half of the question matters as much as the first. The release reports what the evidence cannot support with the same care as what it can.
Five measurements, each with its limits attached.
- Role families, and the language employers actually use to advertise them
- Requested skills, normalised against a canonical skill register
- Salary evidence, only where benchmark coverage is strong enough to compare
- Geographic demand, with a UK and European focus
- Coverage, missingness and confidence, reported alongside every finding
The research inputs, as recorded.
- 42,875job records returned by the snapshot
- 19active sources
- 38countries
- 228canonical skills
- 2,050salary benchmark rows from 13 benchmark sources
- 19,680quality-filtered, spatially publishable rows
Figures read from the Geospatial Careers public read-only snapshot captured on 13 July 2026. They are research inputs, not a claim that every record is equally complete, current, representative or licensed for redistribution.
Findings a partner can cite, on written terms.
- Reproducible findings, each published with its denominator, geography and confidence limits
- A methods and limitations note covering sources, normalisation, spatial completeness and known blind spots
- An aggregated, publication-safe evidence surface; never a raw job-record export
- Co-branded briefings under written attribution, source-link and analytics terms
- A named author, a publication date and a corrections route
What this is not.
A release you can trust has to be explicit about where it stops. These boundaries are part of the brief, not small print.
- Not a jobs board, a training platform or a community product, and not a pitch for one.
- No raw job records, source registry, models, code or subscriber data are transferred to any partner.
- No claim of market representativeness beyond the stated sources, geographies and dates.
- No forecasting of future demand; the release describes observed evidence.
- Findings that fail the quality gate are named as rejected, not quietly dropped.
A brief, not a live product.
The release, the public evidence surface and the methods note are scoped for a six-week build and do not exist yet. Publishing the brief first is deliberate: it lets practitioners, employers and educators correct the scope before the analysis starts, rather than after publication.
Hire, teach or plan in this market? Book a 30-minute research interview to test the scope against your questions. The synthesis comes back to you, and nothing is attributed without your explicit approval.