Benchmarking in Quartyl
The screening pipeline, AI screening, web research, the comparables grid, overrides and qualitative data.
The Screening Pipeline: Quantitative to Qualitative (8 Steps)
The eight-step async analysis pipeline in Quartyl: quantitative, qualitative and AI screening, gated web research and deep screening, final analysis, per-step progress.
Read docQuantitative Screening in Quartyl: Filters and Results
Quantitative screening in Quartyl: the exact filter order from multi-year completeness to RPT and revenue thresholds, Golden Rule aggregation, and the recorded rejection reason per company.
Read docAI Screening: How It Scores Companies and Cites Evidence
AI screening in Quartyl: LLM FAR analysis of each comparable against the tested party, verbatim evidence quotes, the outcome categories, human override and the gated deep pass.
Read docWebsite Enrichment: Where Comparable Descriptions Actually Come From
By default Quartyl does not crawl comparable websites; descriptions come from the dump you upload. The optional Web Research step reads the open web.
Read docThe Comparables Grid: Review, Accept, Reject and Override
The Quartyl comparables grid: every screened company with PLI, status and disposition, the details panel, reviewer overrides with a recorded reason, and how the grid feeds the range.
Read docOverrides: Recording Rationale and Why the Audit Ledger Cares
Overrides in Quartyl: what counts as an override, the mandatory rationale with actor identity and timestamp, the per-study micro-ledger, and aggregation into the cross-study ground-truth registry.
Read docQualitative Data: The Per-Comparable Record and How You Act On It
What Quartyl stores in a comparable’s qualitative record — score, confidence and rationale — what is deliberately not in it, and how reviewer judgment enters as a disposition.
Read docRaw Dump Data: Inspection and Reuse After Screening
Raw dump data in Quartyl: the parquet persisted after quantitative screening, inspecting the numbers behind every filter decision, reuse across re-runs, and the per-tenant retention window.
Read docBrowse other sections
Read the manual, then run the study
Every page in this manual maps to a real feature in your workspace. Sign in and follow along with your own study.