diff --git a/CHANGELOG.md b/CHANGELOG.md
index 0c0e2cf..bd63013 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -25,7 +25,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - `list_media_files()` and `get_media_file()` accept `source="dynamic"` or
   `source="url_analysis"`. The parameter is omitted from the request when unset.
 - `MediaFile.kind` reports `"video"`, `"screenshot"`, or `None`.
-- `UrlAnalysisResponse` carries the scoring fields: `verdict_state`, `threat_score`,
+- `UrlAnalysisResponse` carries the scoring fields: `threat_score`,
   `risk_level`, `privacy_score`, `privacy_risk_level`, `privacy_breakdown`,
   `verdict_provenance`, `impersonation_target`, `completeness`, `intel_detections`,
   `metafields`, and `scored_at`. Each is `None` or empty until scoring finishes.
diff --git a/README.md b/README.md
index 48858a3..b2d90a5 100644
--- a/README.md
+++ b/README.md
@@ -131,7 +131,7 @@ Session control:
   concurrent-submission quota.
 
 Scoring fields (`threat_score`, `risk_level`, `privacy_score`, `privacy_risk_level`,
-`privacy_breakdown`, `verdict_state`, `verdict_provenance`, `impersonation_target`,
+`privacy_breakdown`, `verdict_provenance`, `impersonation_target`,
 `completeness`, `intel_detections`, `metafields`, `scored_at`) are `None` or empty until
 scoring finishes. Read them after the report reaches `completed`.
 
diff --git a/docs/TYPES.md b/docs/TYPES.md
index 532e904..0be6b65 100644
--- a/docs/TYPES.md
+++ b/docs/TYPES.md
@@ -134,7 +134,7 @@ peer report type to dynamic/static/cdr — it is guarded separately by
 
 | Type | Purpose |
 |------|---------|
-| `UrlAnalysisResponse` | Top-level report: `level`, `status`, `general_info`, `screenshot`, optional `ip_info`, `dns_records`, optional `whois`, optional `ssl_certificate`, optional `response_headers`, optional `extracted_file`, optional `threat_analysis`, `pages`, plus the scoring fields `verdict_state`, `threat_score`, `risk_level`, `privacy_score`, `privacy_risk_level`, `privacy_breakdown`, `verdict_provenance`, `impersonation_target`, `completeness`, `intel_detections`, `metafields`, `scored_at`. |
+| `UrlAnalysisResponse` | Top-level report: `level`, `status`, `general_info`, `screenshot`, optional `ip_info`, `dns_records`, optional `whois`, optional `ssl_certificate`, optional `response_headers`, optional `extracted_file`, optional `threat_analysis`, `pages`, plus the scoring fields `threat_score`, `risk_level`, `privacy_score`, `privacy_risk_level`, `privacy_breakdown`, `verdict_provenance`, `impersonation_target`, `completeness`, `intel_detections`, `metafields`, `scored_at`. |
 | `UrlAnalysisGeneralInfo` | `{url, domain, website_title}`. |
 | `UrlAnalysisScreenshot` | `{available: bool}`. Fetch actual bytes with `client.get_screenshot(uuid)`. |
 | `UrlAnalysisIpInfo` | `{ip, asn, city, country, isp, organization, threat_status}`. |
diff --git a/src/threatzone/testing/_responses.py b/src/threatzone/testing/_responses.py
index 9b312fa..1819bf1 100644
--- a/src/threatzone/testing/_responses.py
+++ b/src/threatzone/testing/_responses.py
@@ -781,7 +781,6 @@ def build_url_analysis_response(state: SubmissionState) -> UrlAnalysisResponse:
             "threatDetails": [],
         },
         "pages": [state.final_url] if state.final_url else [target_url],
-        "verdictState": "final" if state.current_status() == "completed" else "provisional",
         "threatScore": _THREAT_SCORE_BY_LEVEL[state.level],
         "riskLevel": _RISK_LEVEL_BY_LEVEL[state.level],
         "privacyScore": 35,
diff --git a/src/threatzone/types/url_analysis.py b/src/threatzone/types/url_analysis.py
index ab70193..b6ab0ae 100644
--- a/src/threatzone/types/url_analysis.py
+++ b/src/threatzone/types/url_analysis.py
@@ -243,7 +243,6 @@ class UrlAnalysisResponse(BaseModel):
     extracted_file: UrlAnalysisExtractedFile | None = Field(default=None, alias="extractedFile")
     threat_analysis: UrlAnalysisThreatAnalysis | None = Field(default=None, alias="threatAnalysis")
     pages: list[str]
-    verdict_state: str | None = Field(default=None, alias="verdictState")
     threat_score: int | None = Field(default=None, alias="threatScore")
     risk_level: str | None = Field(default=None, alias="riskLevel")
     privacy_score: int | None = Field(default=None, alias="privacyScore")
diff --git a/tests/conftest.py b/tests/conftest.py
index 437dfb1..d8fd656 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -874,7 +874,6 @@ def sample_url_analysis_response() -> dict[str, Any]:
             "threatDetails": [{"source": "intel-1", "details": {"family": "EvilKit"}}],
         },
         "pages": ["https://evil.com/login"],
-        "verdictState": "final",
         "threatScore": 72,
         "riskLevel": "High",
         "privacyScore": 35,
diff --git a/tests/test_types.py b/tests/test_types.py
index 9a289a8..d4272eb 100644
--- a/tests/test_types.py
+++ b/tests/test_types.py
@@ -714,7 +714,6 @@ def test_url_analysis_response_keeps_the_scoring_fields(
         report = UrlAnalysisResponse.model_validate(sample_url_analysis_response)
         assert report.threat_score == 72
         assert report.risk_level == "High"
-        assert report.verdict_state == "final"
         assert report.privacy_score == 35
         assert report.privacy_risk_level == "Medium"
         assert report.privacy_breakdown is not None