POLKEY
EPLI underwriting risk data
UK employment tribunal signals for EPLI

Employer litigation risk, structured for underwriting workflows.

Polkey turns UK employment tribunal decisions and hearing-list appearances into employer-level frequency, severity, outcome and activity signals for EPLI underwriters, MGAs, brokers and underwriting-platform partners.

Request the EPLI sample pack See delivery options Built as a data feed first: CSV batch, portfolio refresh, API lookup, or partner integration.
132,000+employment tribunal decisions archived from the public record
130,000+structured extraction rows with respondents, outcomes and awards where grounded
2,800+parsed Courtserve hearing-list appearances in the live archive
Dailyincremental decision and hearing-list monitoring

The feed

The raw tribunal record is public. The underwriting value is the employer-level signal: resolved identities, longitudinal activity, award severity, litigated outcomes, fresh hearing activity, and confidence/audit flags.

Employer risk profiles

One row per normalized employer: decision count, award totals and ranges, outcome mix, jurisdiction mix, latest publication, and supporting evidence count.

Pre-judgment activity

Hearing-list appearances are counted separately from judgments. They are not liability findings, but they show active tribunal exposure before a decision exists.

Underwriting delivery

Designed for portfolio enrichment and referral workflows: CSV batch, scheduled portfolio refresh, API lookup by employer, or integration through a platform partner.

Sample fields

Field groupExamplesUse in underwriting
Employer identitynormalized name, Companies House number, SIC codes, resolution confidenceMatch applicant/portfolio names to a stable employer spine.
Frequencypublished decision count, first/latest decision date, latest publication dateCompare tribunal activity against peer employers and renewal history.
Severitytotal awards, award count, average award, maximum awardFlag employers with demonstrated litigated award exposure.
Outcome mixupheld, partly upheld, dismissed, withdrawn/struck out, other/unknownSeparate volume from adverse litigated outcomes.
Current activityhearing-list appearances, future appearances, next hearing dateMonitor live exposure before judgments are published.
Confidencelow-confidence row count, evidence count, methodology caveatsKeep the signal auditable and suitable for human review.

Methodology, honestly

Published judgments are a partial record: many employment disputes settle, withdraw, or resolve through Acas without a published judgment. Polkey reports litigated-outcome signals as litigated-outcome signals, not claim-population truth.

Hearing-list appearances are pre-judgment activity signals and are counted separately. Award totals are included only where a source-grounded figure is found. Low-confidence extraction rows are flagged rather than silently hidden.

The feed is employer-side only. Claimant/person search is out of scope, and raw evidence URLs are controlled because public slugs can contain claimant names.

Delivery and pricing

Underwriting data is not a £399 self-serve report. Pricing depends on coverage, refresh cadence, delivery shape, evidence requirements and whether Polkey is feeding one portfolio, one MGA, or a platform.

Pilot feed

Scoped pilot

Portfolio-matched refresh for a defined book, class or referral workflow.

  • Employer matching and coverage report
  • Monthly or weekly refresh
  • Evidence and confidence review loop

Data licence / partner

Custom annual

Ongoing feed or platform integration for carriers, MGAs, brokers and underwriting systems.

  • Batch, API or integration delivery
  • Defined SLA and audit trail
  • Commercial terms by usage and distribution