England School & Academy Trust Change Signals

Detect what changed in official Department for Education Get Information about Schools (GIAS) public data since your previous successful run.

This Actor covers England only. GIAS is the official register for educational establishments in England; Scotland, Wales, and Northern Ireland use separate services. The Actor is a persistent change detector, not a generic school-directory scraper.

What this Actor does

On a live run it acquires the current public GIAS establishment, group, and SAT/MAT membership downloads, normalizes the fields needed for change detection, compares them with the previous successful snapshot, emits one Dataset record per material signal, and advances the baseline only after output succeeds.

Examples:

• Capacity increased from 600 to 750
• School joined Example Multi Academy Trust
• Community school converted to Academy converter
• School closed (close date 31/08/2026): Example School

Who it is for

• education technology and software suppliers
• facilities, construction, catering, and estates suppliers
• recruitment and staffing providers
• insurance and consulting teams
• academy trust market-intelligence teams
• education-sector sales and CRM operations

Official data source

Source: DfE Get Information about Schools public downloads.

The Actor uses the official public download flow for:

• all establishment records
• all group records
• academy SAT/MAT membership

GIAS public downloads are updated daily. GIAS content is available under the Open Government Licence v3.0, except where otherwise stated.

No proxy or browser automation is required. The Actor does not collect governors, trustees, personal email addresses, or login-protected information.

Signal types

Implemented signals:

• NEW_SCHOOL
• SCHOOL_CLOSED
• SCHOOL_OPENED
• STATUS_CHANGED
• TRUST_CHANGED
• ESTABLISHMENT_TYPE_CHANGED
• CAPACITY_INCREASED
• CAPACITY_DECREASED
• ACADEMY_CONVERSION
• SCHOOL_NAME_CHANGED
• PHASE_CHANGED
• ADDRESS_CHANGED

URN is the primary school key. School names are never used for entity matching. Trust affiliation uses the official trust identifier/name exposed by GIAS and, when needed, unambiguous active SAT/MAT membership rows. The Actor does not guess a trust relationship when the official data is ambiguous.

First-run baseline behavior

The first live run stores the normalized GIAS snapshot and emits zero signals. Existing schools are not returned as NEW_SCHOOL.

Expected sequence:

1. first run: baseline, 0 signals
2. same source snapshot: 0 signals
3. later GIAS snapshot with changes: relevant signals
4. rerun against that same snapshot: 0 signals

baselineOnly: true deliberately refreshes the complete snapshot without emitting differences.

Example input

{
  "signalTypes": ["TRUST_CHANGED", "CAPACITY_INCREASED", "SCHOOL_CLOSED"],
  "postcodePrefixes": ["LS", "BD"],
  "minCapacityIncrease": 50,
  "maxItems": 500
}

Other filters include school status, establishment type, URN, and trust ID. Filters and maxItems affect Dataset output only; the baseline still represents the complete official source.

Example output

{
  "signalType": "CAPACITY_INCREASED",
  "severity": "HIGH",
  "detectedAt": "2026-09-13T00:00:00.000Z",
  "urn": "100001",
  "schoolName": "Example Academy",
  "schoolStatus": "Open",
  "establishmentType": "Academy converter",
  "phase": "Secondary",
  "trustId": "1234",
  "trustName": "Example Multi Academy Trust",
  "localAuthority": "Leeds",
  "postcode": "LS1 1AA",
  "previousValue": 600,
  "currentValue": 750,
  "previousCapacity": 600,
  "currentCapacity": 750,
  "openDate": "01/09/2020",
  "closeDate": null,
  "signalSummary": "Capacity increased from 600 to 750",
  "sourceUrl": "https://get-information-schools.service.gov.uk/Downloads"
}

Each Dataset item represents exactly one signal. A school can therefore produce more than one record on the same source date, for example STATUS_CHANGED plus SCHOOL_CLOSED.

Severity

HIGH is used for school closure, academy/trust changes, academy conversion, and major capacity changes. Smaller capacity changes, new/opened schools, status changes, and establishment-type changes are generally MEDIUM. Name, phase, and address changes are LOW.

Severity is deterministic prioritization, not predictive risk scoring.

Scheduling recommendation

GIAS public downloads are updated daily, so a daily schedule is appropriate. More frequent runs normally see the same source snapshot and emit nothing.

Data freshness

Freshness is limited by the official GIAS publication cycle. This Actor does not claim real-time monitoring or instant alerts.

Snapshot safety and idempotency

Processing order is:

official download → parse/normalize → schema/count validation → diff → Dataset output → summary → baseline commit

The previous baseline remains authoritative when acquisition, parsing, required-field validation, source-date validation, source-count validation, Dataset output, or summary output fails. A large implausible collapse in school or academy-trust count is rejected instead of being committed.

maxItems never truncates the snapshot. Re-running an unchanged official snapshot emits no duplicate signals.

The baseline format remains compatible with earlier versions of this Actor. Capacity comparisons intentionally require a numeric value on both sides, so the first run after adding capacity tracking does not create false capacity signals from an older baseline that did not store capacity.

Sample mode

mode: "sample" runs deterministic in-memory before/after data through the real change logic. It demonstrates trust change, academy conversion, capacity increase/decrease, closure, and a new school without reading or modifying the production baseline.

Known limitations

• GIAS covers England, not the whole United Kingdom.
• The Actor can only detect changes represented in the official public downloads.
• SCHOOL_CLOSED requires explicit official status/close-date evidence. A row disappearing from a snapshot is not automatically labelled as a closure.
• SCHOOL_OPENED requires an explicit transition from a closed status or removal of a close date alongside an open status.
• A material correction in GIAS can look the same as a real-world change.
• Cosmetic text normalization reduces, but cannot eliminate, every possible formatting-only metadata change.
• Contact-detail changes are not emitted in this version; the Actor deliberately keeps its signal set focused on organisation-level public fields.

Compute / storage behavior

Live runs download the three required current GIAS files. A named Apify Key-value Store holds the persistent normalized baseline. The default Dataset contains only change signals. The default Key-value Store receives a SUMMARY record with source diagnostics, entity counts, signal counts, source dates, and snapshot status.