{"closest_prior_art":[{"name":"Summon Relevance Metrics Framework","overlap":"Analyzes discovery-service query and click logs at both query and session levels, computes abandonment and reformulation-effort measures, aggregates statistics, addresses bot/noise effects, and uses findings to improve the search experience.","remaining_difference":"It does not organize assurance around a valid stable library-wide indicator being overextended to local contexts, nor specify a joint macro-distribution display, predeclared operational strata, variance decomposition, sustained-excursion thresholds, or routing to authorized library owners.","source_ids":["SRC2"]},{"name":"Primo Zero Result Searches analytics","overlap":"Provides recurring reports of zero-result searches broken down by date, discovery view, tab, search scope, resource filter, search type, field, and user group, making contextual failures visible.","remaining_difference":"It is a product reporting capability limited to leading zero-result queries; it does not require comparison with a stable aggregate success series, trajectory analysis, uncertainty and minimum-cell rules, or a governance crosswalk for persistent excursions.","source_ids":["SRC3"]},{"name":"Oria query-log assessment","overlap":"Uses discovery-system transaction logs to examine popular and zero-result queries, demonstrates that ostensibly popular queries can still be unsuccessful, and connects failures to query form, system sensitivity, metadata, and curriculum integration.","remaining_difference":"It is a retrospective query study rather than a standing two-level assurance rule and does not test stable aggregate performance against predeclared session strata or route sustained local excursions to accountable owners.","source_ids":["SRC1"]}],"contrastive_claim_falsifier":"The contrastive claim fails if the declared aggregate is unstable or invalid, instrumentation harmonization and bot exclusion eliminate the apparent excursions, adequately observed predeclared strata show no consequential persistent heterogeneity, or reviewers cannot reproducibly assign supported excursions to an authorized metadata, interface, accessibility, or collection owner.","contrastive_claim_remaining":"Beyond existing discovery-log studies and vendor reports, a valid stable discovery-wide success indicator may still conceal persistent, consequential failures in predeclared non-identifying session contexts; jointly displaying the stable aggregate with uncertainty-qualified stratum trajectories and routing only sustained excursions through an explicit owner crosswalk may add reproducible diagnostic value.","experiment_id":"eoa_inverse_innovation_exp13_second_slot_policy60_20260806","gates":{"adequate_source_search":{"rationale":"The bounded search covered the proposal directly, OPAC and transaction-log terminology, vendor analytics and privacy guidance, and combinations involving zero results, abandonment, reformulation, stratification, aggregate masking, and owner routing. Four opened sources from four publishers were retained, including primary research, first-party documentation, and official guidance.","source_ids":["SRC1","SRC2","SRC3","SRC4"],"status":"PASS"},"bounded_next_test":{"rationale":"An eight-week, read-only retrospective analysis is feasible because prior work derives query-, session-, zero-result-, abandonment-, and reformulation-related measures from discovery logs. Pre-registration, instrumentation checks, bot exclusion, minimum cells, uncertainty, and a go/no-go memo keep the test bounded; privacy controls are mandatory.","source_ids":["SRC1","SRC2","SRC3","SRC4"],"status":"PASS"},"distinct_testable_claim":{"rationale":"The retained art supports contextual discovery analytics but does not directly specify the claimed conjunction: a valid stable aggregate retained alongside predeclared distributional and trajectory companions, sustained-excursion criteria, and reproducible routing to authorized local owners. That incremental diagnostic claim has explicit empirical falsifiers.","source_ids":["SRC1","SRC2","SRC3"],"status":"PASS"},"no_obvious_safety_or_authority_stop":{"rationale":"ALA guidance confirms that even nominally anonymous activity data may contain identifiable information and requires minimization, restricted access, anonymized reporting, retention limits, and privacy audits. The proposed approved, de-identified, minimum-cell, read-only first step with deletion and halt conditions is compatible with those constraints; no production or patron-level action is authorized.","source_ids":["SRC2","SRC4"],"status":"PASS"},"supported_problem":{"rationale":"Discovery-log research and product documentation make zero-result, abandonment, repeated-query, and context-specific failure patterns observable and connect them to metadata or system conditions. The sources do not directly demonstrate the proposal's precise condition of persistent heterogeneity beneath a genuinely stable library-wide success indicator, so support is partial rather than complete.","source_ids":["SRC1","SRC2","SRC3"],"status":"PASS"}},"prior_art_disposition":"ADJACENT_PRIOR_ART","problem_evidence":{"finding":"The underlying problem is visible: discovery sessions and queries exhibit measurable zero-result, abandonment, and reformulation difficulties that vary with query and interface context and can implicate metadata or system design. Evidence that such consequential heterogeneity persists specifically while a valid collection-wide success indicator remains stable was not found in the retained sources.","source_ids":["SRC1","SRC2","SRC3"],"status":"PARTLY_SUPPORTED"},"research_id":"eoa_inverse_innovation_exp13_light_screen_20260806","schema_version":1,"screen_id":"E13P127","screen_survival":true,"search_lanes":{"component_combination":{"no_result_note":"No retained source combined stable aggregate monitoring, variance decomposition, privacy-thresholded strata, persistent-excursion detection, and an authorized owner crosswalk in one library-discovery assurance rule.","queries":["library discovery stable aggregate success metric subgroup zero result reformulation analytics","library discovery ecological fallacy aggregate average subgroup search success","library search analytics sustained alert subgroup excursion dashboard metadata owner"],"source_ids":["SRC1","SRC2","SRC3"]},"direct_problem_and_intervention":{"no_result_note":"No direct match was found for the full proposed two-level assurance rule or its stable-valid-aggregate framing.","queries":["Distribution-Aware Assurance for Stable Library Discovery Metrics","library discovery stable aggregate success metric subgroup zero result reformulation analytics","library discovery assessment aggregate metrics subgroup disparities search"],"source_ids":["SRC1","SRC2"]},"products_practices_and_standards":{"no_result_note":null,"queries":["site:knowledge.exlibrisgroup.com Primo Analytics zero results search session actions report","library discovery search analytics dashboard zero result reformulation sessions","library search logs privacy guidelines de-identification small cells"],"source_ids":["SRC2","SRC3","SRC4"]},"synonyms_and_historical_terms":{"no_result_note":"Older OPAC and transaction-log literature showed local failures and reformulation analysis, but the retained results did not state the full stable-macro/local-excursion governance pattern.","queries":["OPAC transaction log analysis zero hit searches reformulation abandonment","library discovery usability analytics transaction logs query reformulation abandonment","Simpson's paradox library discovery analytics search logs"],"source_ids":["SRC1","SRC2"]}},"sources":[{"claims_supported":["Discovery-system transaction logs can expose popular and zero-result query patterns.","Some popular queries were unsuccessful, while zero-result queries were associated with long or pasted-citation queries.","Observed failures can implicate discovery-system sensitivity, metadata, and curriculum-material integration."],"publisher":"Nordic Journal of Information Literacy in Higher Education","source_id":"SRC1","source_type":"PRIMARY_RESEARCH","title":"“More than Meets the Eye” — Analyzing the Success of User Queries in Oria","url":"https://noril.uib.no/article/view/2510"},{"claims_supported":["A discovery-service framework can analyze both individual queries and inferred search sessions.","Abandonment, queries to first click, and related behavioral measures can reveal likely search difficulty, although clicks are imperfect relevance proxies.","Instrumentation, session definitions, scripts, bots, and noisy traffic materially affect interpretation of aggregate log metrics."],"publisher":"Code4Lib Journal","source_id":"SRC2","source_type":"PRIMARY_RESEARCH","title":"Actions Speak Louder than Words: Analyzing Large-Scale Query Logs to Improve the Research Experience","url":"https://journal.code4lib.org/articles/8693"},{"claims_supported":["Primo Analytics supports zero-result reports with dimensions including date, view, tab, search scope, resource filter, search type, searched field, and user group.","The product aggregates and reports leading zero-result searches daily.","Available dimensions make some contextual discovery failures operationally inspectable but impose coverage and instrumentation limitations."],"publisher":"Ex Libris","source_id":"SRC3","source_type":"FIRST_PARTY_PRODUCT","title":"Primo Zero Result Searches","url":"https://knowledge.exlibrisgroup.com/Primo/Product_Documentation/Primo/Analytics/040Primo_Analytics_Subject_Areas/Primo_Zero_Result_Searches"},{"claims_supported":["Discovery-service activity data may contain personally identifiable information even for users who are not logged in.","Libraries should minimize collection, restrict access, anonymize broadly distributed analytics, limit retention, and securely destroy unneeded activity data.","Library privacy obligations and governance apply regardless of whether systems are library-managed, vendor-managed, or hybrid."],"publisher":"American Library Association","source_id":"SRC4","source_type":"OFFICIAL_GUIDANCE","title":"Library Privacy Guidelines for Library Websites, OPACs, and Discovery Services","url":"https://www.ala.org/advocacy/privacy/guidelines/OPAC"}],"world_novelty_boundary":"This bounded public-web screen found adjacent research, vendor analytics, and privacy guidance but no direct source for the complete proposed rule. It establishes only coarse researchability and a remaining contrastive test; it cannot establish world novelty, patentability, market size, expert acceptance, or realized value."}