Skip to content

Targeting and audience audit on Microsoft Advertising

Microsoft Advertising: Read configured inclusions, exclusions and audience references separately from reported demographics.

Recipe, Microsoft Advertising

Intent Read configured inclusions, exclusions and audience references separately from reported demographics.

Steps
2
Tools
4
Rules
2
Failure modes
2
skills/microsoft-targeting-audit.md

When to use it

Use it when

  • The user requests targeting and audience audit for a selected Microsoft Advertising account.

Do not use it when

  • Inferring delivery or account eligibility from a successful configuration alone.

What it needs first

Preconditions

  • Connect Microsoft Advertising and explicitly select an authorized account. Writes require an eligible paid plan and confirmation.
  • Read bing_get_setup_guide and use only the capabilities enabled for this account.
  • The account is among the chosen accounts, confirmed via list_accounts.

Evidence every conclusion must carry

  • Every figure carries its account, metric and window.
  • Anything below a readable sample size is reported as unjudged, not as zero.
Inputs
InputRequiredWhat it is
windowAnalysis window.Default: last 30 days.

Procedure

In order. Every tool named is one the gateway ships, and links to its reference.

Steps

  1. List campaigns and their ad groups or ad squads.bing_get_campaignsbing_get_ad_groups
  2. Read the actual targeting configuration and resolve its audience references.bing_get_targetingbing_get_audiences

Decision rules

The observable condition, what it lets you conclude, and what takes the conclusion back.

  • When An audience is empty or an identifier cannot be resolvedConclude Report the missing evidence; an audience object does not prove matched members or delivery.
  • When Targeting differs from demographic performanceConclude Keep configuration and observed reach separate; expansion and privacy thresholds can explain differences.

Where the agent stops

The readout changes nothing. Any change it motivates goes through Safe Writes: preview, human confirmation, then apply.

How it goes wrong quietly

The cases where the analysis is wrong and still looks right. Read them before trusting a number.

  • Read native criterion types and all pages; one targeting response may not cover every criterion.
  • Do not upload customer identifiers as part of a configuration audit.

What the answer contains

What Targeting and audience audit returns
SectionWhat it holds
ReadoutThe figures, with windows and accounts named.
FlagsWhat deserves a deeper skill or a human decision.

Manage your ads from your AI assistant

Start for free

Free with 1 ad account, no card. Pro is free for 7 days.