Selected finding
Consent choice architecture review signals
Benchmark frequency is directional market context only. It is not a compliance benchmark, legal conclusion, or severity score. Rare findings may be top-ranked only when retained evidence is strong; common findings may remain medium when evidence is automated or context-dependent. Rarity is not severity, and prevalence is not compliance risk.
Observed
Retained consent-surface evidence showed choice-architecture signals, such as control availability, path depth, visual hierarchy, forced interaction, or repeated prompting, that may require consent UX or consumer-protection review.
Why this matters
Consent interfaces can shape how users understand and exercise privacy choices. For review teams, this signal can help identify consent UX patterns that may warrant deeper review across choice availability, effort, clarity, repetition, accessibility, and consistency with public statements.
Detection methodology
CertScore retains representative consent-surface evidence for visible controls, button labels, path depth, first-layer availability, hierarchy cues, overlays, repeated prompts, preference paths, public preference-management explanation, and scan coverage context where available. The finding is surfaced when retained evidence indicates a cluster of consent choice-architecture signals that may affect how users encounter, compare, accept, reject, or revisit privacy choices in the observed scan scope. Supporting consent-governance disclosure context may note whether retained public materials clearly explain how choices can be changed, withdrawn, retained, renewed, or managed when runtime consent relevance is present. CertScore treats consent choice-architecture results as review signals. The scanner does not determine that dark-pattern status, deception, unfairness, consent validity, legal status, or compliance status occurred. Reviewers should consider jurisdiction, region, CMP configuration, prior consent state, user intent, accessibility, localization, repeated prompts, equivalent choice paths, public claims, and whether the retained evidence reflects the relevant user-facing consent surface.
Confidence semantics: Good when retained consent-surface evidence includes multiple choice-architecture signals, such as control availability, labels, path depth, hierarchy, overlay behavior, repeated prompts, or preference-path context; stronger when retained evidence includes repeated observations across regions, viewports, pages, or states. Manual review is still needed for user impact, legal interpretation, deception or unfairness assessment, accessibility, and remediation quality.
Top-finding calibrationWhat must be retained to surface, top-rank, demote, or suppress this finding.
Minimum to surface
- Concrete consent-surface choice architecture signal, or retained consent-control lifecycle evidence showing no obvious preference-revisit control in sufficient scan coverage.
High confidence requires
- Two or more retained choice-architecture signals, or retained pages checked, footer/preferences surfaces inspected, and consent/tracking context for the revisit-control subtype.
- Consent preference-management explanation gaps are supporting disclosure context, not standalone top-card evidence.
Top ranking requires
- Forced interaction plus missing/nested reject or repeated prompt. The revisit-control subtype remains Medium by default unless existing severity calibration supports escalation.
Demote or suppress when
- CMP name only.
- Banner presence only.
- Unrelated modal.
- Prior consent state may have hidden controls.
- Blocked or shallow preference-control coverage.
These rules describe ranking calibration for already-projected findings. They do not create findings from raw signals.
Example evidence
Choice architecture review signal
artifact=consent_ui_004role=finding_supporting_artifacturl=https://example.com/component=cookie_bannersignals=reject_path_nested, accept_primary, repeated_prompt_after_dismissaccept_control_text=Accept allpreferences_control_text=Manage choicesreject_control_location=preferences_layer [manual_review_recommended]prompt_reappeared_after_dismiss=true [manual_review_recommended]review_caveat=manual review should confirm choice equivalence, repetition, accessibility, region, CMP configuration, user impact, and legal interpretation
Preference revisitability review signal
artifact=consent_control_lifecycle_001role=finding_supporting_artifacturl=https://example.com/subtype=privacy_settings_control_not_observedinitial_consent_layer_observed=trueconsent_dependent_tracking_observed=truecoverage_status=usablepages_checked=[https://example.com/]controls_searched=[cookie preferences, privacy settings, manage consent]footer_links_inspected=retained_bounded_labels_and_hrefsprivacy_settings_control_observed=falsecookie_preferences_link_observed=falsecmp_reopen_control_observed=falsewithdrawal_text_observed=falsereview_caveat=automated public-page observation; manual review should confirm regional variants, returning-user state, CMP configuration, and legal interpretation
Review context
possible_source=cmp_choice_architecturecontexts_to_review=first_layer_controls, preference_path, visual_hierarchy, repeated_prompting, preference_revisitability, keyboard_access, screen_reader_accessjurisdictional_review_needed=truemanual_review_needed=true
What should not count by itself
cmp_vendor=Example CMP [insufficient_without_consent_surface_artifact]banner_present=true [insufficient_without_choice_architecture_signal]accept_button_primary=true [audit_only_without_refusal_path_context]dark_pattern_label [not_a_finding_determination]
View redacted sample JSONHide redacted sample JSON
{
"findingId": "consent_dark_patterns_detected",
"label": "Consent choice architecture review signals",
"category": "Consumer protection",
"criticality": "medium",
"evidenceConfidence": "good",
"directVsInferred": "correlated_observation",
"evidence": {
"summary": "Retained consent-surface evidence showed choice-architecture signals, such as control availability, path depth, visual hierarchy, forced interaction, or repeated prompting, that may require consent UX or consumer-protection review.",
"examples": [
{
"title": "Choice architecture review signal",
"lines": [
"artifact=consent_ui_004",
"role=finding_supporting_artifact",
"url=https://example.com/",
"component=cookie_banner",
"signals=reject_path_nested, accept_primary, repeated_prompt_after_dismiss",
"accept_control_text=Accept all",
"preferences_control_text=Manage choices",
"reject_control_location=preferences_layer [manual_review_recommended]",
"prompt_reappeared_after_dismiss=true [manual_review_recommended]",
"review_caveat=manual review should confirm choice equivalence, repetition, accessibility, region, CMP configuration, user impact, and legal interpretation"
]
},
{
"title": "Preference revisitability review signal",
"lines": [
"artifact=consent_control_lifecycle_001",
"role=finding_supporting_artifact",
"url=https://example.com/",
"subtype=privacy_settings_control_not_observed",
"initial_consent_layer_observed=true",
"consent_dependent_tracking_observed=true",
"coverage_status=usable",
"pages_checked=[https://example.com/]",
"controls_searched=[cookie preferences, privacy settings, manage consent]",
"footer_links_inspected=retained_bounded_labels_and_hrefs",
"privacy_settings_control_observed=false",
"cookie_preferences_link_observed=false",
"cmp_reopen_control_observed=false",
"withdrawal_text_observed=false",
"review_caveat=automated public-page observation; manual review should confirm regional variants, returning-user state, CMP configuration, and legal interpretation"
]
},
{
"title": "Review context",
"lines": [
"possible_source=cmp_choice_architecture",
"contexts_to_review=first_layer_controls, preference_path, visual_hierarchy, repeated_prompting, preference_revisitability, keyboard_access, screen_reader_access",
"jurisdictional_review_needed=true",
"manual_review_needed=true"
]
},
{
"title": "What should not count by itself",
"lines": [
"cmp_vendor=Example CMP [insufficient_without_consent_surface_artifact]",
"banner_present=true [insufficient_without_choice_architecture_signal]",
"accept_button_primary=true [audit_only_without_refusal_path_context]",
"dark_pattern_label [not_a_finding_determination]"
]
}
]
}
}Regulatory review context
Consent UX choice architecture review
Retained consent-surface evidence showed choice-architecture signals that may be relevant to consent, cookie/tracker, transparency, accessibility, consumer-protection, and privacy-claims review. Applicability depends on jurisdiction, region, purpose, CMP configuration, equivalent choice paths, public statements, accessibility, user impact, and manual review.
View applicability notes
Legal and regulatory frameworks
- GDPR valid consent reviewConsent may be used for cookies, tracking, personal data processing, profiling, or advertising.
- EDPB cookie banner design reviewRetained consent-surface evidence suggests hidden reject paths, preselected choices, confusing hierarchy, or unequal interaction cost may require review.
- CCPA/CPRA choice-architecture / dark-pattern review contextThe interface may affect California privacy choices, opt-out paths, or consent.
- FTC choice architecture / dark-pattern review contextDesign choices may obscure, burden, or otherwise affect user choices involving privacy or commerce.
Jurisdictional contexts
- EU GDPR/ePrivacy consent UI reviewEU/EEA users and cookie or tracking consent UI may be in scope.
- UK PECR / ICO cookie-choice reviewUK users and non-essential cookie choices may be in scope.
- U.S. privacy choice-architecture / dark-pattern review contextRetained UI evidence suggests privacy choices, opt-outs, consent, or targeted advertising controls may be affected.
This finding does not determine legal status, deception, unfairness, dark-pattern status, consent validity, or compliance status. Review the retained consent-surface evidence, labels, hierarchy, path depth, prompt behavior, region targeting, CMP configuration, accessibility, public statements, and applicable exemptions.
Evidence standard
Strong
- Retained consent-surface evidence includes multiple concrete choice-architecture signals, such as hidden refusal path, materially higher refusal effort, forced interaction, imbalanced visual hierarchy, repeated prompting, or unclear choice labels.
- Evidence includes page URL, observed consent layer, labels or path context, and scan-state context.
- Evidence distinguishes finding-supporting artifacts from review context and unrelated interruptions.
- Coverage context indicates the consent surface was not materially blocked, truncated, or replaced by unrelated overlays.
- Repeated observations across viewports, regions, pages, or consent states may strengthen confidence when retained.
Good
- Retained evidence shows one or more consent choice-architecture signals with enough context for reviewer inspection, but full user impact, repetition, visual prominence, or equivalent-choice analysis may require manual review.
- The retained example is enough for a reviewer to inspect the observed surface, available choices, and likely remediation owner.
- The evidence is likely a choice-architecture review signal, but legal interpretation, dark-pattern, deception, or unfairness analysis, accessibility, and regional configuration require manual review.
Audit-only
- Contextual signals suggest choice-architecture risk, but retained evidence lacks enough detail to identify the affected consent layer, controls, labels, or interaction path.
- CMP template, policy language, or static configuration suggests possible risk, but no retained consent-surface artifact supports the observed user-facing state.
- A single low-detail UI signal appears relevant, but the evidence is not enough to distinguish consent UX from unrelated modals, paywalls, bot challenges, or login flows.
Insufficient
- CMP vendor name alone.
- Policy text alone.
- Banner presence alone.
- Generic dark_pattern_label without retained consent-surface artifacts.
- Unrelated modal, paywall, bot challenge, age gate, login wall, or newsletter prompt without consent-surface linkage.
- Claims about deception, unfairness, dark-pattern status, legal status, compliance status, or consent validity based only on automated UI evidence.
Evidence levels explain how CertScore treats retained consent-surface artifacts. They are not legal conclusions.
Common causes
- CMP template makes acceptance more prominent or easier than refusal.
- Refusal is nested behind preference screens or unclear labels.
- Consent prompts reappear after dismissal or refusal.
- Overlays, scroll locks, or modal behavior require interaction before ordinary browsing.
- Region, localization, A/B test, or returning-user configuration changes the observed choice architecture.
Recommended review questions
- Which consent-surface signals were retained: missing reject, nested reject, forced interaction, visual hierarchy, repeated prompting, unclear labels, or path depth?
- Which layer, page, region, language, viewport, and browser state produced the observation?
- Are accept and refusal paths available, equivalent, and accessible?
- How many steps are required to accept, reject, or manage choices?
- Does the interface use labels, colors, ordering, repetition, or modal behavior that may affect user choice?
- Could the signal be caused by a paywall, bot challenge, age gate, login wall, newsletter prompt, or unrelated modal?
- Do results vary by region, localization, viewport, A/B test, prior consent state, or CMP configuration?
- Should privacy, legal, UX, and accessibility review confirm user impact, legal interpretation, and remediation quality?
Limitations and cautions
- This finding is an automated consent UX review signal, not a legal conclusion, certification, compliance determination, dark-pattern determination, deception determination, unfairness determination, or determination of consent validity.
- Automated UI checks can identify choice-architecture signals, but they may miss or misclassify regional variants, A/B tests, localization, returning-user states, mobile layouts, preference-layer behavior, repeated prompts, accessibility issues, unrelated modals, and post-login flows.
- Automated evidence may not fully determine user intent, user impact, deception, unfairness, dark-pattern status, consent validity, or whether a legal standard applies.
- Manual review is needed to confirm consent context, equivalent choices, accessibility, public claims, legal interpretation, user impact, and remediation quality.
- CertScore retains representative evidence for review and may not list every variant across regions, viewports, languages, CMP states, or user journeys.
- Findings should be evaluated with implementation context and applicable privacy, consent, accessibility, and consumer-protection requirements before operational or legal reliance.
- Automated findings may contain errors and should be reviewed with the retained evidence.
- Not detected means not observed in the scan scope; it is not proof of absence.
- Findings are runtime evidence and public-surface observations for review, not legal conclusions.