Straight-through processing (STP)
The percentage of documents processed end-to-end with no human intervention — the headline metric for IDP economics and the gap between every vendor demo and every production deployment.
What Straight-through processing (STP) means in practice
Straight-through processing is the percentage of documents that traverse the IDP pipeline from intake to downstream-system commit with no human touch. The gap between vendor-demo STP (typically 95%+) and 9-months-in production STP (typically 75–85% on heterogeneous corpora) is where the engineering work actually lives. MindMap's DocuMage deployments sustain 94% STP on heterogeneous document types by investing disproportionately in four areas: classifier accuracy above 98% (below this, routing fails and downstream extraction operates on wrong assumptions); per-document-type extraction strategies (one strategy across heterogeneous corpora reliably under-performs); explicit confidence-scoring with field-level routing to human review; and exception-handling design that treats the 6% as a first-class workflow, not an afterthought.
Related terms
Intelligent Document Processing (IDP) →
End-to-end automation of document workflows: capture, classify, extract structured data, validate, route, and integrate into downstream systems.
LLM-augmented extraction →
Using a large language model to extract structured fields from documents — particularly effective on layout-free documents where template-based OCR fails.
Schema-driven extraction →
The pattern where the target output schema is the primary input to the extraction prompt — the LLM is told exactly which fields to find and what type each should be.
More in this category
All 62 terms, in plain language
Sovereign AI, RAG, agentic AI, IDP, MLOps and the regulations that shape enterprise AI.