# Create custom account-correlation rules only when strong identifiers are unavailable > Use Manage account correlation rules in Microsoft Defender for Identity (Preview) to review this narrow operational decision without extending the source beyond its stated scope. - Canonical URL: https://update.dsesecurity.com/updates/create-account-correlation-rules-only-without-strong-identifiers/ - Publisher: Detection Systems & Engineering (DSE Security) - Author: DSE Security Editorial Team - Published: 2026-08-27T12:12:42+00:00 - Modified: 2026-08-27T13:04:10+00:00 - Last reviewed by DSE: 2026-08-26 - Resource type: Guide - DSE priority: Advisory - Topics: Cybersecurity, IT, Microsoft 365 & Identity - Reading time: 3 minutes ## What you need to know Use Manage account correlation rules in Microsoft Defender for Identity (Preview) to review this narrow operational decision without extending the source beyond its stated scope. ## Potentially affected Teams, systems, services, or facilities within the stated scope of Manage account correlation rules in Microsoft Defender for Identity (Preview) ## DSE recommendation Compare the observed state with the cited official source, document applicability and exceptions, and test any approved change with rollback safeguards. ## Article Use this document to connect an official requirement or behavior to observable evidence: Create custom account-correlation rules only when strong identifiers are unavailable. Only the official source and traced locations below supply facts. Confirm applicability before acting. ## Source fact: The official [Manage account correlation rules in Microsoft Defender for Identity (Preview)](https://learn.microsoft.com/en-us/defender-for-identity/custom-account-correlation-rules) from Microsoft supports the following bounded statements: - Custom correlation rules target accounts that do not share strong identifiers such as account ID, SID, object ID, or UPN. The research record locates this support at Opening overview. - Microsoft describes the rules as especially useful for privileged accounts with unique naming conventions and documents creating, editing, and removing them in the Defender portal. The research record locates this support at Opening overview and scope. Do not import neighboring assumptions into the source record. The supported task is a scoped comparison involving identity detections, contributing events, entities, alert logic, tuning, exclusions, investigation, and response decisions and the conditions the source actually describes. ## What the source does not establish The feature is marked Preview; a naming rule is an inference and must not override contradictory authoritative identity evidence. The citation is not a substitute for observed state, authorization, compliance evidence, or dependency health. Examine Active Directory, Windows DNS, sensor health, time, network visibility, cloud analytics, endpoint evidence, and security operations before translating the source into an operational decision. ## Applicability questions - For source statement 1 at Opening overview, which observable configuration, record, or test can confirm applicability here? - For source statement 2 at Opening overview and scope, which observable configuration, record, or test can confirm applicability here? - Which deployed instance of identity detections, contributing events, entities, alert logic, tuning, exclusions, investigation, and response decisions will be compared with the source, and why that instance? - How will the review distinguish a source mismatch from a failure in Active Directory, Windows DNS, sensor health, time, network visibility, cloud analytics, endpoint evidence, and security operations? - Who approves the conclusion, exception, test window, and rollback threshold? ## DSE recommendation: DSE recommends using the cited source as the evidence anchor for this decision. Use a two-person review for the source interpretation and the resulting operational decision. Record the source location, examined part of identity detections, contributing events, entities, alert logic, tuning, exclusions, investigation, and response decisions, observed and expected states, owner, and reason for deviation. For an approved change, define prerequisites, a limited test path, success and stop conditions, monitoring, and rollback. Check Active Directory, Windows DNS, sensor health, time, network visibility, cloud analytics, endpoint evidence, and security operations in design order. Protect credentials, keys, recovery material, personal data, and sensitive topology in evidence. ## Verification and evidence Build a reproducible chain from Opening overview; Opening overview and scope to the observed environment. Useful domain evidence includes sanitized alerts, source events, entity timelines, tuning conditions, exclusion scope, analyst findings, and response outcomes; label every item with scope, timestamp, collector, and stable identifier. Retain the starting state, authorization, execution record, outcome, deviation, and final state as one review package. Move disruptive checks to an approved test path. Reopen the decision when versions, design, dependencies, ownership, or official guidance changes. ## Official references - [Manage account correlation rules in Microsoft Defender for Identity (Preview)](https://learn.microsoft.com/en-us/defender-for-identity/custom-account-correlation-rules) — Microsoft ## Primary reference - Name: Manage account correlation rules in Microsoft Defender for Identity (Preview) - Authority: Microsoft Learn - URL: https://learn.microsoft.com/en-us/defender-for-identity/custom-account-correlation-rules - Source publication date: 2026-07-23 ## Citation and use Preferred citation: “Create custom account-correlation rules only when strong identifiers are unavailable,” DSE Security, https://update.dsesecurity.com/updates/create-account-correlation-rules-only-without-strong-identifiers/ Publishing principles: https://update.dsesecurity.com/updates/dse-updates-editorial-methodology/ Usage and citation policy: https://update.dsesecurity.com/usage/ Copyright © 2026 Detection Systems & Engineering. All rights reserved.