Extract data from employment contracts
An employment contract sets the terms of the working relationship, and HR, payroll, and legal each read it for a different reason at a different moment. When a new hire such as Priya Raman signs with an employer like Meridian Software Ltd, HR keys the start date, the job title, and the salary into the HRIS, payroll sets up the pay run, and legal files the executed copy against the day an employment tribunal might ask for it. A company hiring 300 people a year holds 300 of these, drafted from templates that recruiters, managers, and outside counsel all edit, so no two read alike even though the operative terms are the same: the parties, the position, the start date, the salary and pay frequency, working hours, holiday entitlement, the probationary period, notice on termination, and any confidentiality or non-compete restriction.
The variation lives in the terms that get negotiated. Pay can be quoted as an annual figure of £72,000 in GBP or as a monthly amount, and a bonus tied to performance sits beside the base. A fixed-term contract carries an end date while a permanent one does not, and the probationary period sets a different notice window for the first few months. Holiday entitlement runs to 25 or 28 days depending on the jurisdiction, and a remote-work clause and an intellectual-property assignment appear in some contracts and not others. Under the Employment Rights Act 1996 in the United Kingdom, an employer has to give a written statement of the main terms, so the contract is the record that a tribunal, an auditor, and the employee all rely on. A start date of 2026-07-06 on a contract dated 2026-06-20, with probation ending 2027-01-05, sets the payroll and right-to-work deadlines that follow.
Talonic reads the contract and returns the parties, the position, the compensation, the dates, and the key clauses as structured fields. A base salary of £72,000 paid monthly comes back as £6,000 a period with a 10% bonus target, and the termination and non-compete clauses are kept as excerpts, so HR and legal work from structured terms while still reading the exact wording.
What gets extracted from employment contracts
How extraction works for employment contracts
Contracts are generated from HRIS platforms such as Workday and BambooHR, from Word templates, and from counsel for senior hires, so the clause order and the numbering drift from one contract to the next. Talonic classifies the document and runs it through the contract schema in the Field Registry, which captures the parties, the position, the compensation, the dates, and the key clauses without a per-template setup. Salary is normalized to an annual figure with its pay frequency and its ISO 4217 currency, so a monthly quote and an annual quote both resolve to a consistent number in GBP for payroll. A fixed-term end date, a probationary period, and a notice period are captured as discrete fields, while termination, confidentiality, intellectual-property, and non-compete clauses are tagged and kept verbatim because their wording is negotiated and matters in a dispute before an employment tribunal in the United Kingdom. Every value returns with a confidence score and a pixel-region pointer under DIN SPEC 91491 conformity, so HR and legal can verify the salary, the start date, or a restriction against the signed contract before opening a payroll record.
Sample extraction
A UK permanent employment contract with a compensation schedule
{
"document_number": "EC-2026-0442",
"document_date": "2026-06-20",
"candidate_name": "Priya Raman",
"employer": "Meridian Software Ltd",
"job_title": "Senior Backend Engineer",
"employment_type": "permanent",
"effective_date": "2026-07-06",
"salary": 72000,
"currency": "GBP",
"pay_frequency": "monthly",
"compensation_details": {
"salary_amount": 72000,
"pay_frequency": "monthly",
"monthly_amount": 6000,
"bonus_amount": 7200,
"total_target": 79200
},
"working_hours": "40 per week",
"vacation_days": 28,
"trial_period": "6 months",
"termination_clause": "3 months notice after probation",
"non_compete_clause": "12 months, within the United Kingdom",
"confidentiality_agreement": true,
"governing_law": "England and Wales"
}Frequently asked
Does it normalize salary quoted monthly or annually?
Yes. A salary stated as an annual figure or as a monthly amount is captured with its pay frequency and an annualized value, so payroll receives one consistent number. A base of £72,000 a year in GBP resolves to £6,000 a month regardless of how the contract phrases it.
How are fixed-term and permanent contracts told apart?
A fixed-term contract carries an end date such as 2027-07-05 and a permanent one does not, so the presence of an expiration date drives the classification. The probationary period and its shorter notice window are captured separately, since they change the terms for the opening months.
Are termination and non-compete clauses captured verbatim?
Yes. Notice on termination, confidentiality, intellectual-property assignment, and any non-compete or non-solicitation restriction are tagged and kept as excerpts, because the exact wording and the duration decide whether a clause is enforceable in a dispute.
Does it capture working hours and holiday entitlement?
Standard weekly hours, holiday entitlement, and any remote-work arrangement are captured as discrete fields, so an HR system can load leave balances and working patterns for an employee such as Priya Raman without someone rereading the contract.
What fields are extracted from employment contracts?
Talonic returns employment contracts as schema-validated, typed fields. Common fields include Employee Name, Employer, Job Title, Employment Type, and more, each normalized (dates to ISO 8601, amounts as numbers) and mapped to a stable key so the output shape stays the same across layouts.
How accurate is extraction from employment contracts, and how is confidence reported?
Every extracted cell carries a confidence score from 0.0 to 1.0 and a provenance pointer back to the source page and region, so low-confidence values can be reviewed against the original before the data is trusted downstream. There is no single accuracy number: confidence is reported per field so you can gate on it.
Can I use employment contracts extraction in production?
Yes. The same engine behind this guide is available as a production REST API and Node SDK with sync, async, and streaming modes, schema versioning, signed webhooks, and EU-resident processing. Start free with an API key, then scale on usage-based pricing.
What does it cost to extract data from employment contracts?
There is a free tier for prototyping and agent evaluation with no credit card. Paid usage is credit-based at 1,000 credits per euro: page ingestion is 100 credits per page and registry-resolved queries are free. See talonic.com/pricing for current rates.
Ready to extract from your own employment contracts?
- GDPR
- HIPAA
- ISO 27001 / 42001 aligned
- DIN SPEC 91491 co-author
- EU-resident infrastructure
Author note
Reviewed by Talonic engineering · last reviewed 2026-07-06