Investigate risk with Microsoft Entra ID Protection
In brief
The article now explains how to separately review verification completion, access decisions, and user or sign-in risk records. It also documents enabling verification history collection and clarifies that successful verification may not clear every risk.
What Entra admins need to know
When investigating Require Identity Verification policies, review all relevant outcomes and do not dismiss risk solely because verification succeeded.
This editorial summary was generated by AI from the documentation changes. Verify important details in the full Microsoft Learn article.
Documentation change
The comparison below shows only the changed extract. Use the full-page view for complete context.
How to investigate risk
- Depending on the results of the investigation, mark the user or sign-in as confirmed compromised, confirmed safe, or dismiss the risk. You can confirm compromise in the Microsoft Entra admin center or programmatically using Microsoft Graph.
- Set up risk-based Conditional Access policies to prevent similar attacks or to address any gaps in coverage.
Investigate identity verification outcomes
For a risk-based policy using Require Identity Verification (preview), investigate verification completion, the policy's access decision, and the relevant risk record as separate outcomes. An issued credential, successful access attempt, or registered passkey alone isn't evidence that the risk was remediated.
To view users' verification status and history, enable verification data collection as described in the Verified ID authentication method prerequisites.
- Review the applicable risk condition, policy, and sign-in result for the tested access attempt.
- Review available identity verification status and history. Distinguish credential issuance and presentation from a completed verification outcome.
If you review credential provenance, remember that during preview it applies only to passkeys. It isn't a substitute for verification history or the risk record. Collect only approved diagnostic evidence.
Investigate specific detections
Certain risk detections require specific investigation steps. The following sections outline some of the most common risk detections and recommended actions.
@@ -2,10 +2,12 @@ title: Investigate risk with Microsoft Entra ID Protection description: Learn how to investigate risky users, detections, and sign-ins in Microsoft Entra ID Protection. ms.topic: how-to-ms.date: 05/27/2026+ms.date: 10/08/2026+author: tilarso+ms.author: tilarso ms.reviewer: lvandenende ai-usage: ai-assisted-ms.custom: sfi-image-nochange+ms.custom: sfi-image-nochange, msecd-doc-authoring-108 --- # How to investigate risk @@ -71,6 +73,23 @@ Once the risk is contained, more investigation might be required to mark the ris 1. Depending on the results of the investigation, mark the user or sign-in as confirmed compromised, confirmed safe, or dismiss the risk. You can confirm compromise in the Microsoft Entra admin center or [programmatically using Microsoft Graph](howto-identity-protection-simulate-risk.md#confirm-compromise-using-microsoft-graph). 1. [Set up risk-based Conditional Access policies](howto-identity-protection-configure-risk-policies.md#enable-policies) to prevent similar attacks or to address any gaps in coverage. +## Investigate identity verification outcomes++For a risk-based policy using **Require Identity Verification (preview)**, investigate verification completion, the policy's access decision, and the relevant risk record as separate outcomes. An issued credential, successful access attempt, or registered passkey alone isn't evidence that the risk was remediated.++To view users' verification status and history, enable verification data collection as described in the [Verified ID authentication method prerequisites](../identity/authentication/concept-authentication-verified-id.md#prerequisites).++> [!NOTE]+> Creating an identity verification profile automatically enables **Collect identity verification history**; otherwise, enable it in the authentication methods policy under **Authentication Methods** > **Policies** > **Verified ID** > **Configure** > **Data collection**.++1. Review the applicable risk condition, policy, and sign-in result for the tested access attempt.+1. Review available [identity verification status and history](../identity/authentication/concept-authentication-methods-manage.md#identity-verification-status-and-history). Distinguish credential issuance and presentation from a completed verification outcome.+1. Check the relevant user-risk or sign-in-risk record for the expected remediation outcome. Successful verification can remediate risk in supported scenarios, but isn't proof that every risk was cleared.+1. If verification succeeds but risk remains, continue the investigation and record both outcomes. Don't dismiss risk solely because verification succeeded.+1. Record cancellation, failed verification, missing or unavailable credentials, and the return to the original access attempt as separate test outcomes. Use the [existing remediation process](howto-identity-protection-remediate-unblock.md#validate-identity-verification-remediation) when self-remediation doesn't complete.++If you review [credential provenance](../identity/authentication/concept-authentication-passkeys-fido2.md#credential-provenance), remember that during preview it applies only to passkeys. It isn't a substitute for verification history or the risk record. Collect only approved diagnostic evidence.+ ## Investigate specific detections Certain risk detections require specific investigation steps. The following sections outline some of the most common risk detections and recommended actions. 