{
  "affected_column_count": 4,
  "affected_columns": [
    "amount",
    "priority",
    "team",
    "ticket_id"
  ],
  "amount_outlier_count": 1,
  "business_risk_area_count": 4,
  "business_rule_reference_count": 4,
  "business_scenario": "Support-operations ticket export used by internal dashboards.",
  "dataset_id": "support_tickets",
  "duplicate_primary_key_count": 1,
  "finding_count": 5,
  "generated_by": "scripts/verify_business_impact.py",
  "high_priority_action_count": 3,
  "impact_summary": {
    "amount_outliers": {
      "affected_column": "amount",
      "business_risk": "Extreme values can skew reporting and need review before publication.",
      "check": "numeric_outliers",
      "outlier_count": 1
    },
    "check_coverage": {
      "observed_check_count": 4,
      "observed_checks": [
        "duplicate_primary_key",
        "missing_values",
        "negative_amount",
        "numeric_outliers"
      ]
    },
    "duplicate_primary_keys": {
      "affected_column": "ticket_id",
      "business_risk": "Dashboards can double-count a support case or attach remediation to the wrong row.",
      "check": "duplicate_primary_key",
      "duplicate_count": 1
    },
    "missing_routing_fields": {
      "affected_columns": [
        "priority",
        "team"
      ],
      "business_risk": "Support operations cannot reliably route or prioritize every ticket.",
      "check": "missing_values",
      "missing_field_count": 2,
      "missing_rate_per_field": 0.125
    },
    "negative_amounts": {
      "affected_column": "amount",
      "business_risk": "Refund-like events are mixed into positive customer-impact facts.",
      "check": "negative_amount",
      "negative_count": 1
    }
  },
  "issue_category_count": 4,
  "missing_routing_field_count": 2,
  "negative_amount_count": 1,
  "not_claimed": [
    "No verified external users yet.",
    "No customer production deployment is claimed.",
    "This artifact measures a reproducible business-data case study, not enterprise adoption."
  ],
  "owner_handoff_count": 4,
  "quality_score": 24,
  "recommended_action_count": 5,
  "remediation_scorecard": {
    "business_risk_areas": [
      {
        "area": "Dashboard accuracy",
        "evidence": "1 duplicate ticket_id case can double-count support volume.",
        "owner": "Data Engineering",
        "priority": "HIGH",
        "recommended_handoff": "Deduplicate by the latest event timestamp, add idempotent merge logic, and enforce a uniqueness constraint in the pipeline."
      },
      {
        "area": "Support routing",
        "evidence": "2 required routing fields are missing across priority and team.",
        "owner": "Support Operations",
        "priority": "HIGH",
        "recommended_handoff": "Trace null generation for priority and team before publishing dashboard data."
      },
      {
        "area": "Customer-impact reporting",
        "evidence": "1 negative amount is mixed into positive customer-impact facts.",
        "owner": "Analytics Engineering",
        "priority": "HIGH",
        "recommended_handoff": "Validate refund handling and separate credit events from positive payment facts."
      },
      {
        "area": "Executive metric review",
        "evidence": "1 amount outlier can skew aggregate customer-impact reporting.",
        "owner": "Data Analytics",
        "priority": "MEDIUM",
        "recommended_handoff": "Inspect source records for amount and decide whether to cap, filter, or correct them."
      }
    ],
    "resume_safe_outcome": "Produced a verified remediation scorecard mapping 5 data quality findings to 4 business risk areas, 3 high-priority actions, and 4 owner handoffs.",
    "sla_style_checks": [
      "one ticket_id per support event",
      "priority and team must be present for routing",
      "customer-impact amount must be non-negative",
      "extreme amounts require review before dashboard publication"
    ],
    "summary": "The agent converts raw quality findings into a prioritized remediation handoff for support-operations analytics owners."
  },
  "resume_safe_summary": "Quantified 4 support-ticket data quality issue categories across 8 rows, including duplicate ticket IDs, missing routing fields, negative amounts, and amount outliers.",
  "root_cause_hypothesis_count": 3,
  "row_count": 8,
  "source_artifact": "docs/verified-support-ticket-result.json",
  "status": "FAIL",
  "top_root_cause_hypotheses": [
    {
      "confidence": 0.71,
      "evidence": [
        "numeric_outliers on amount: {'outlier_count': 1, 'lower_bound': -60.18, 'upper_bound': 194.29}",
        "negative_amount on amount: {'negative_count': 1}"
      ],
      "recommended_action": "Separate refunds, credits, and exceptional values from standard fact tables or annotate them explicitly.",
      "supporting_checks": [
        "negative_amount",
        "numeric_outliers"
      ],
      "title": "Business-rule validation is not separating exceptional transactions from standard facts."
    },
    {
      "confidence": 0.65,
      "evidence": [
        "duplicate_primary_key on ticket_id: {'duplicate_count': 1}"
      ],
      "recommended_action": "Add idempotent merge logic and enforce a uniqueness check on the primary key.",
      "supporting_checks": [
        "duplicate_primary_key"
      ],
      "title": "The ingestion pipeline may be replaying events without idempotent merge logic."
    },
    {
      "confidence": 0.63,
      "evidence": [
        "missing_values on team: {'missing_rate': 0.125, 'missing_count': 1}",
        "missing_values on priority: {'missing_rate': 0.125, 'missing_count': 1}"
      ],
      "recommended_action": "Trace null generation for affected fields through the source API and transform layer.",
      "supporting_checks": [
        "missing_values"
      ],
      "title": "Source API or transform logic is producing incomplete fields for required analytics columns."
    }
  ]
}