Payroll data extractionthat understands meaning.
Every employer formats payslips differently, and the same value hides behind different words: Bruttolohn, Lohn, salario, gross pay. MiruIQ extracts by meaning against semantic schemas, so one field definition covers every issuer.
Normalize salary statements and HR documents from any issuer into one schema, extracting by meaning instead of by position.
Same data, thousand layouts.
Payroll document processing breaks template-based tools by design: every issuer has its own layout, and layouts change without notice. Each variation means template maintenance, and each maintenance gap means manual fixes.
The deeper problem is vocabulary. Gross salary appears under a dozen names across issuers and languages. Position-based extraction cannot know they are the same field; extraction by meaning can.
Normalization, step by step.
From mixed HR inbox to one consistent schema.
1. Ingest
Salary statements and HR documents arrive by upload, email or API, from every employer format at once.
2. Classify
Optional: recognize issuer and document type first, useful when different types need different schemas.
3. Normalize
Semantic schemas extract by meaning: gross pay is one field whether the document says Bruttolohn, Lohn or salario.
4. Validate
Key fields are checked: figures that must add up are recalculated, and low-confidence extractions park for human review.
5. Deliver
Normalized records flow to your HR and payroll systems in one consistent structure.
What changes for HR ops.
No template upkeep
New issuer, new layout, same schema: nothing to build or maintain per employer.
Faster verification
Onboarding and salary verification stop waiting on manual reading and retyping.
Coverage that grows
Automation covers every issuer from day one instead of the top ten you built templates for.
Numbers to hold the system to.
Track normalization quality, not just throughput.
Exception rate
Share of documents with missing or low-confidence fields that need a reviewer.
Rework rate
Documents re-processed after a correction, your signal for schema gaps.
Time to first result
Intake to normalized output, per document and per new issuer.
Payroll extraction questions
What HR and payroll teams ask most.
You define the fields you need once (gross pay, net pay, employer, period) and MiruIQ extracts them by meaning rather than position. The same schema covers every issuer and language: Bruttolohn, Lohn and salario all land in the gross-pay field. There are no per-employer templates, because the extraction is not pre-trained on layouts.
No. Template maintenance is exactly what this workflow eliminates: extraction is generic and schema-driven, so a payslip from an employer you have never seen processes the same way as a known one. Complex payslip tables are handled, including footnotes that change how values must be read.
Delivery is configured per pipeline: webhook to your HR system, direct database insert, S3, or API pull. Data leaves in your schema, so the receiving system gets one consistent structure regardless of what the source documents looked like. And the whole pipeline can run on-premise or in Swiss hosting.
See it on your own documents.
The 14-day free trial covers 200 documents: enough to define your schemas, run real files and measure the results before anyone calls you.