{"reviewed_at":"2026-08-07","methodologies":[{"id":"public-evidence-benchmark","version":"1.1.0","title":"Public-evidence software benchmark methodology","summary":"A reproducible method for comparing software capabilities from dated public evidence without assigning a generic winner or weighted overall score.","selectionPolicy":["Include products that address the same primary buyer problem and have enough current public evidence to assess every selected dimension.","Define the product set and decision dimensions before interpreting individual findings, then preserve that scope in the published snapshot.","Treat audience, use cases and product emphasis as narrative context rather than scored benchmark dimensions."],"exclusionPolicy":["Exclude inventory-only products when their public evidence has not yet been normalized into the current capability contract.","Document named exclusions and their reason on the report instead of silently removing products from the evaluated set.","Do not infer absent capabilities from missing public documentation; use unknown when evidence is insufficient."],"sourceHierarchy":["Official product documentation, product pages and integration directories","Official pricing, security and regional-availability documentation","Vendor changelogs and dated first-party customer or implementation material","Independent editorial, community and review-platform evidence when the report explicitly includes sentiment"],"statusDefinitions":{"supported":"Current public evidence directly documents the capability for the reviewed product or service scope.","partial":"Public evidence supports part of the capability, but material scope, regional or edition constraints remain.","not_supported":"Current public evidence explicitly states that the capability is unavailable in the reviewed scope.","unknown":"The reviewed public evidence is insufficient to determine support; this does not mean the capability is absent."},"conflictPolicy":"Prefer the most specific and current first-party source. When credible sources conflict, retain the narrower claim, lower confidence and explain the conflict or limitation.","unknownPolicy":"Unknown means insufficient public evidence at the observation date. It must never be rendered or interpreted as proof that a product lacks a capability.","testingDisclosure":"OpenCompare did not perform hands-on product testing for this benchmark. Findings describe documented public evidence, not laboratory or customer-environment validation.","paidPlacementPolicy":"Coverage, product order, statuses and findings are not paid placements. OpenCompare does not sell placement or a favorable result in this research layer.","scoringPolicy":"The benchmark publishes dimension-level evidence and conditional decision views. It does not calculate a weighted overall score, universal rank or generic winner.","versioningPolicy":"Published reports are immutable versioned snapshots. Evidence changes are reviewed through an explicit diff and released as a new semantic version with a public revision note.","citationGuidance":"Cite the report title, OpenCompare Research, edition, version, publication date, canonical URL and access date. Cite vendor sources separately for vendor-owned claims.","licenseNotes":"OpenCompare's original benchmark text and structured observations are offered under CC BY 4.0. Vendor names, marks and source materials remain the property of their respective owners.","reviewedAt":"2026-08-07"}],"contributors":[{"id":"opencompare-research","name":"OpenCompare Research","role":"author","organization":"OpenCompare","url":"https://opencompare.net/research"},{"id":"hagen-hoferichter","name":"Hagen Hoferichter","role":"reviewer","organization":"OpenCompare","url":"https://github.com/h4gen"}]}