An automated campaign can meet its target while leaving advertisers with unanswered questions about how it got there. Microsoft’s latest documented API additions expose more of that activity and give campaign managers another way to restrict where ads run.
The changes affect agencies, advertisers and software providers managing Microsoft Advertising through connected tools. They include a report connecting search terms with delivered headlines and landing pages, support for LinkedIn segments, and bulk-management support for campaign-level device exclusions.
Microsoft lists the report and LinkedIn interfaces in its August 2026 API release notes, alongside a device-exclusion mapping for its .NET software development kit. The documentation establishes an August release history, rather than a new September launch.
Performance Max Gets a Device Boundary
Advertisers can exclude computers, smartphones or tablets from Microsoft Performance Max campaigns through a campaign-level negative device criterion. Microsoft’s device-exclusion documentation specifies that at least one device category must remain eligible.
An exclusion prevents ads from displaying on the selected device type. It therefore changes eligibility directly, rather than simply expressing a preference about how much a campaign should bid for that traffic.
The scope is specific. Microsoft says these campaign-level negative device criteria are supported only for Performance Max campaigns. The record can be uploaded or downloaded through the Bulk service, allowing connected campaign-management software to handle the restriction programmatically.
For teams already reviewing Performance Max controls, device eligibility becomes another setting that can be recorded alongside the campaign’s other restrictions. It does not establish whether excluding a device will improve results. That remains a separate performance question.
The documentation describes the exclusion mechanism and its supported values; it does not prescribe a preferred device mix.
Search Term Reporting Connects the Ad to Its Destination
The reporting addition addresses a different problem: understanding which message and destination accompanied a search.
Microsoft’s Search Term Landing Page report request brings the search term, delivered headline and landing-page URL into one report. Developers can limit requests to particular accounts or campaigns and select a reporting period.
That relationship is the useful addition. A search term viewed alone does not show which headline accompanied it, while a landing-page total does not explain which searches contributed to that page’s activity.
Teams building search term reporting into their dashboards can retain those connections in the same reporting workflow. The request requires a final URL and at least one performance statistic, such as impressions. It also supports filters for ad type, delivered match type, language and search queries.
Microsoft supports time-based aggregation, including daily and weekly reporting. Advertisers can therefore examine the connections over a defined period without treating the report as a record of an identifiable visitor’s browsing session.
An AI Headline Flag Has a Precise Meaning
The report’s column definitions add detail beyond the search term and destination.
A final-URL source field distinguishes a landing page taken from an ad’s specified URL from one selected through domain expansion. A headline-source field identifies whether all headline assets in a reported combination came from the advertiser or at least one was AI-generated.
That second definition matters. A combination containing both advertiser-provided and AI-generated assets receives the AI-generated classification. The label does not say that every headline in the combination was produced by AI.
Available performance columns include clicks, spend, conversions, revenue and return on ad spend. Microsoft describes these as aggregated statistics for the requested period and aggregation level.
Taken together, the fields support a more specific reading of campaign delivery: which search appeared with which headline combination and destination, accompanied by performance totals. They do not, by themselves, establish that an AI-generated headline caused a conversion or that URL expansion caused a change in revenue.
LinkedIn Segment Support Comes With an Availability Limit
Microsoft’s August release notes also introduce the CustomLinkedInCriterion object and GetLinkedInSegments operation.
The custom LinkedIn criterion contains profile, profile ID and profile-type fields. The separate retrieval operation returns existing LinkedIn segments associated with an advertising account, including their properties and configuration.
For developers, those are distinct functions: representing a targeting criterion and retrieving segment information. Their presence in the API documentation should not be read as a statement that every account has access to every LinkedIn targeting capability.
Microsoft makes the availability limit explicit on its GetLinkedInSegments reference page, which says the feature is not yet available to everyone.
The release history also shows that LinkedIn segment support has developed across several updates. Microsoft listed operations for adding, deleting and updating LinkedIn segments in March 2026. August’s retrieval interface extends that documented work.
The Microsoft Advertising API Still Needs a Reporting Workflow
API support gives software a way to request data or manage a setting. Whether an advertiser sees those capabilities in an agency dashboard depends on how that dashboard’s integration is built.
The practical implication for paid media teams is a more detailed campaign review. Search intent, delivered messaging and destination relevance can be examined together, with conversion data providing performance context. Device exclusions remain a separate delivery decision; identifying a weak result and deciding to remove an entire device category are different steps.
Report preparation also affects comparisons. Microsoft’s request object includes an option to return only fully processed reporting periods. That option defaults to false, so a request does not automatically insist on complete data.
For agencies comparing recent campaign activity, that setting belongs alongside the selected dates, account scope and aggregation level. The API exposes the data, but the request still determines which period and which version of reporting completeness reaches the dashboard.
