Microsoft Advertising Reporting & Metrics: What the Numbers Mean and Which Ones Are the Scorecard

Microsoft Advertising reporting organizes into three metric families: efficiency (CTR, CPC, CPA, ROAS), quality and competitive (Quality Score and its three components, the impression-share family), and the conversion layer (Conversions, All Conversions, view-through conversions, assists). Before any KPI conversation is meaningful, that conversion layer must exist — and it exists only when Universal Event Tracking is firing and, for EEA/UK/Swiss traffic, consent signals are passing.

Most Microsoft Advertising accounts plateau because the default column set sits at the top of the funnel. Impressions, clicks, and CTR are always populated — they require no tracking. Conversions, cost per conversion, revenue, and ROAS are the columns that decide whether spend pays back, and they are blank without a working UET tag. An impressions-and-clicks report is not a performance report. MB Adv Agency finds this to be the single most common reporting gap across new Microsoft Advertising accounts: the conversion layer was never wired up, so the account has been optimizing for clicks against no outcome signal.

This reference covers the complete reporting vocabulary: thirteen core KPIs and when each is the primary scorecard, the three Quality Score components and how to read them per-component, the six impression-share metrics and the diagnostic IS-lost-to-budget vs. IS-lost-to-rank pair, the four conversion columns and how they differ, and Microsoft’s three attribution models. On that last point: Microsoft Advertising offers three attribution models — LastClick, LastTouch, and Data-Driven. The Google-style first-click, linear, and time-decay models do not exist in Microsoft Advertising; the bingads-13 API specification (updated February 2026) confirms this. Any guide claiming five models for this platform is wrong.

The reporting interface itself is covered in the second half: 41 report types across 9 categories, the in-platform Report Builder, scheduled and emailed exports (CSV, TSV, XML), and the 2026 diagnostic additions — data-driven attribution rollout, Copilot Performance Shift Root Cause Analysis, and Conversion Tracking Diagnostics. Industry context for metric prioritization links out to Legal PPC, Financial services PPC, SaaS software PPC, Dental PPC, and Real estate PPC — each vertical has a different primary KPI, and the goal-to-metric mapping is in the KPI selection table below.

Key Takeaways

• Conversion layer first. Conversions, conversion rate, CPA, revenue, and ROAS are blank without a working UET tag. Confirm that layer before reading any other column. Microsoft does not model un-consented conversions — EEA/UK/CH traffic with a missing consent signal is simply absent from the report.
• Quality Score is a three-part diagnostic. The 1–10 score combines expected CTR, ad relevance, and landing page experience (each 1–3, updated daily). A score dragged down by landing-page experience is a different fix than one dragged down by expected CTR.
• IS Lost to Budget vs. IS Lost to Rank is the most actionable pair in the account. The first means raise budget; the second means raise bid or Quality Score. On January 15, 2026, Microsoft extended these Share-of-Voice metrics to Performance Max campaigns.
• Conversions ≠ All Conversions. "Conversions" counts click-based goals flagged “include in conversions.” "All Conversions" adds cross-device and view-through (impression-based). Reporting one as the other over- or under-states results.
• Microsoft Advertising has three attribution models, not five. LastClick (default, all campaigns), LastTouch (Audience Workflow campaigns only), Data-Driven (eligible auto-bidding strategies, rolled out to all advertisers by end of May 2026). First-click, linear, and time-decay do not exist here.
• 41 report types across 9 categories. The Report Builder supports custom column sets; scheduled reports run automatically and email to recipients; exports are CSV (default), TSV, or XML.

Key Metrics to Monitor in Microsoft Ads: The Complete KPI Reference

The thirteen core Microsoft Advertising KPIs form a funnel: spend buys impressions, impressions earn clicks (CTR → average CPC), clicks earn conversions (conversion rate → cost per conversion), and conversions earn revenue (ROAS). The metric that matters is the one tied to the campaign goal — and the first prerequisite is confirming the conversion columns are populated, which requires UET tracking to be in place.

The table below covers every core KPI by formula, what it measures, and when it becomes the primary scorecard. The "head term" query this section exists to answer — "what are the key metrics to monitor in microsoft ads campaigns" — maps to a cluster winner holding position ~1.94 in Google Search Console, making this the single most important anchor in the pillar.

The funnel structure within Table 1 is the order of operations for diagnosing a campaign. Start at the bottom (is ROAS or CPA at target?), then move up (is conversion rate within range?), then ask whether clicks are quality (CTR and CPC), then ask whether the account is capturing enough of the available market (impression share — covered in the competitive metrics section below). Clicks are not results; an impression-and-click report without the conversion layer attached is not a performance report.

For e-commerce advertisers, revenue and ROAS are the primary scorecards; view-through conversions and All Conversions matter for brands running display alongside search. For lead-gen businesses — legal firms, dental practices, HVAC contractors, plumbing services — conversions, conversion rate, and cost per conversion are the primary scorecards. The goal-to-KPI mapping is in the selection table later in this article.

Microsoft Advertising Quality Score: Three Components, One Diagnostic

Quality Score is a 1–10 keyword-level rating (10 best) that measures how competitive an ad is relative to other ads targeting the same traffic. It is updated daily and built from three component sub-scores, each rated on a 1–3 scale. A higher Quality Score lowers average CPC and improves ad position — making it one of the few reported metrics that is also a lever. Source: Microsoft Learn, Quality Score and Quality Impact in Depth.

The correct way to read Quality Score is per-component, not as a headline number to chase. A keyword scoring 6 due to weak landing-page experience and a keyword scoring 6 due to weak expected CTR are different problems with different fixes. Treating the headline number as the optimization target is how advertisers spend months “working on Quality Score” with nothing to show. Read the components first.

The misconception that Quality Score is a vanity badge to push from 6 to 8 understates its diagnostic value. The three sub-components are independent signals: expected CTR tells you about the keyword-ad relationship; ad relevance tells you about the ad-query relationship; landing page experience tells you about the post-click experience. Each can be high while the others are low. A keyword with 3/3 expected CTR and 1/3 landing page experience needs a landing page fix, not a copy test. Reading which component is dragging the score down is the correct diagnostic sequence — and that sequence applies equally to lead-gen accounts in financial services and SaaS as to local service businesses in Flagstaff HVAC or Missoula legal services.

MB Adv Agency’s diagnostic framework for Quality Score: sort keywords by Quality Score ascending, then segment by which sub-component is lowest. Keywords with low landing-page-experience scores go to a page-improvement sprint; keywords with low ad-relevance scores go to ad-group restructuring; keywords with low expected-CTR scores go to copy testing and ad-extension review. The Microsoft Ads keywords and ad copy pillar covers the copy side of this in depth.

Impression Share and the Competitive Metrics: Reading the Share-of-Voice Family

The impression-share family is the competitive scoreboard for Microsoft Advertising. Impression share (impressions ÷ eligible impressions) measures how much of the available auction the account is capturing. The six metrics in this family divide into gauges (IS, top IS, absolute top IS, click share) and diagnostics (IS lost to budget, IS lost to rank). The diagnostic pair is the most actionable: it tells you whether to spend more money or improve the account. Source: Microsoft Advertising Help, Share of Voice report.

On January 15, 2026, Microsoft extended Share-of-Voice metrics — impression share, click share, IS lost to budget, and IS lost to rank — to Performance Max campaigns, with historical data back to November 10, 2025. These metrics aggregate Search and Shopping placements; Audience Network placements are excluded. This closed a significant blind spot: before the update, Performance Max campaigns had no competitive visibility data. Source: Adsroid, Microsoft Advertising 2026 Updates.

IS Lost to Budget and IS Lost to Rank are the most actionable pair in the account because they replace guesswork with a diagnosis. An account with 42% impression share, 35% lost to budget, and 23% lost to rank has a clear split: most of the gap is a budget problem, some is a rank problem. Budget allocation and bid strategy changes are different conversations — Microsoft Ads bidding and budget covers the bid-strategy side. IS Lost to Rank pointing at Quality Score loops back to the Quality Score section above.

The Share-of-Voice report in Microsoft Advertising’s reporting interface surfaces these metrics at campaign, ad group, and keyword level. For accounts running both standard Search campaigns and Performance Max, the January 2026 extension means the competitive visibility is now comparable across campaign types — a gap that had made it impossible to assess PMax auction penetration alongside Search on the same terms.

Conversions vs. All Conversions vs. View-Through vs. Assists: The Four Conversion Columns

Microsoft Advertising surfaces four distinct conversion columns, each counting a materially different set of interactions. Reporting one as another over- or understates campaign performance. The default “Conversions” column is the narrowest definition; “All Conversions” is the broadest. All four depend on Universal Event Tracking. Source: Search Engine Land; Adobe Advertising, Microsoft Advertising conversion data.

The distinction between Conversions and All Conversions is load-bearing for how you set up bidding and reporting. “Conversions” is the column smart bidding strategies use as their optimization signal. “All Conversions” is the broader analysis view — it includes view-through and cross-device touchpoints, which are real attribution signals for upper-funnel campaigns but are too broad to use as a bid target for lead-gen accounts. An account where “All Conversions” is 3× the “Conversions” figure is not tripling its lead count; it is capturing impression-based and cross-device paths that are analytically interesting but not the same signal as a click-based form submission.

The UET dependency is non-negotiable. Unlike Google Ads, which fills un-consented conversion gaps with modeled data for EEA/UK/Swiss traffic, Microsoft Advertising does not model the gap. A missing UET consent signal (mandatory from May 5, 2025 for EEA/UK/CH traffic) means those conversions are absent from all four columns. An account serving European traffic without a consent management platform integrated with UET has undercounted conversions with no recovery path. The Microsoft Ads conversion tracking and UET pillar covers the setup in full.

Microsoft Advertising Attribution Models: Exactly Three, Not Five

Microsoft Advertising offers three attribution models: LastClick, LastTouch, and Data-Driven. The Google-style models — first-click, linear, and time-decay — do not exist on this platform. Google itself deprecated first-click, linear, and time-decay in October 2023. Any guide claiming Microsoft Advertising offers five attribution models is citing Google’s old feature set. Source: Microsoft Learn, AttributionModelType Value Set (bingads-13, updated February 2026).

Data-Driven Attribution (DDA) rolled out to all eligible advertisers by end of May 2026, announced April 26, 2026. The rollout matters for reporting because DDA redistributes conversion credit away from the last click — which changes which keywords and campaigns “look” efficient in the platform. An account switching from LastClick to DDA will see the Conversions column redistribute credit across the conversion path, surfacing keywords that were contributing but receiving no credit under the last-click default. Source: ppcnewsfeed.com, Microsoft Advertising Introduces Data-Driven Attribution; ALM Corp, DDA explainer.

Attribution model selection is a reporting decision, not just a settings default. Switching models changes what the Conversions column reports without changing the underlying conversions — and it changes the optimization signal smart bidding uses. The practical implication: compare performance before and after a model switch using a consistent lookback window; the change in reported conversions reflects the attribution redistribution, not a change in actual lead or sale volume. MB Adv Agency treats attribution model selection as a prerequisite conversation before any performance review that involves conversion columns — because two accounts running the same campaigns report different conversion counts if one uses LastClick and the other uses DDA.

Report Types in Microsoft Advertising: The Complete Reference

Microsoft Advertising’s reporting service supports 41 distinct report types across 9 categories as of the April 2026 bingads-13 API documentation, last updated June 15, 2026. Every report is available through the in-platform Report Builder for custom column sets, as an instant one-click download, or as a scheduled report that runs automatically and emails to recipients. Export formats are CSV (default), TSV, and XML. Source: Microsoft Learn, Report Types (bingads-13); Microsoft Learn, Reports; Microsoft Advertising Help, Understanding the Report Builder.

The Performance category is where the diagnostic work happens. The Search Query report shows what actual queries triggered ads — the foundation of negative keyword management and match-type refinement. The Share of Voice report surfaces impression share, top IS, absolute top IS, click share, and the IS-lost diagnostic pair. The Conversion Performance report covers conversions, assists, revenue, and cost-per-conversion columns. The Goals and Funnels report shows conversion goal paths. The Professional Demographics report gives company size, industry, and job function breakdowns — the B2B signal that has no equivalent in Google Ads and matters specifically for SaaS software and financial services advertisers targeting by professional profile.

The Geographic and User Location reports are distinct: Geographic reports where the search query came from based on search intent signals; User Location reports where the user is physically located. For local service advertisers — Austin PPC management, Chicago PPC management — running both reports together reveals whether the account is capturing searchers outside the target service area, which is a common budget-waste source.

KPI Selection by Campaign Goal: Which Metric Is the Scorecard

The right primary KPI depends on the campaign goal. Impressions are the lead KPI for brand awareness; conversions and CPA are the lead KPIs for lead generation; revenue and ROAS lead for e-commerce. The metrics in the left column of Table 1 are always visible — the correct decision is choosing which of them is the scorecard for the goal at hand, and confirming the conversion layer makes that measurement possible.

The B2B row is the Microsoft-specific differentiator: the Professional Demographics report surfaces company size, industry, and job function for Microsoft Audience Network placements — a signal that has no equivalent in Google Ads and that matters for advertisers in SaaS software, consulting, and financial services targeting by buyer persona. For e-commerce accounts in categories like fashion or furniture, ROAS is the primary scorecard but All Conversions and view-through matter for Audience Network campaign evaluation.

Reporting Misconceptions: Three Wrong Assumptions That Corrupt the Numbers

Three persistent misconceptions lead advertisers to misread their Microsoft Advertising reports. Each is worth treating as a checklist item when onboarding a new account or interpreting a monthly report for the first time.

2026 Reporting Additions: The Platform Is Shifting from “Pull a Report” to “Ask Why”

Three new diagnostic features surfaced at Microsoft Advertising Activate 2026 (May 19, 2026). Each represents the platform adding an interpretation layer on top of raw report data. Source: ALM Corp, Activate 2026 Recap; Microsoft Advertising, Activate 2026 Key Takeaways.

Data-Driven Attribution — rolled out to all eligible advertisers by end of May 2026 — changes which keywords and campaigns appear efficient by distributing credit across the conversion path rather than assigning it entirely to the last click. Its primary reporting implication: comparing performance before and after a switch requires using a consistent lookback window, because the change in reported conversions reflects the attribution redistribution, not a change in actual conversion volume.

Copilot Performance Shift Root Cause Analysis explains why a metric moved, categorizing the cause as delivery, bidding, or competitive pressure. This surfaces inside the campaign management workflow and is designed to reduce the analytical work required to diagnose a week-over-week performance swing. Conversion Tracking Diagnostics checks that UET tags and conversion goals are configured correctly, identifying tracking gaps before they become reporting gaps. Both features extend the platform’s diagnostic posture — the raw report data is the same, but the platform now adds a layer that contextualizes metric changes without requiring the analyst to cross-reference multiple report types manually.

How to Create and Schedule Reports in Microsoft Advertising

The Microsoft Advertising reporting interface has three entry points: the Report Builder for custom column sets, instant one-click campaign downloads, and scheduled reports that run automatically and email to recipients. Export formats are CSV (default), TSV, and XML. The Report Builder is the workhorse for any report that requires a non-default column set or a date range other than the past 30 days. Source: Microsoft Advertising Help, Understanding the Report Builder.

To run a custom report: navigate to Reports in the left-rail navigation → select Report Builder → choose the report type (Keyword Performance, Search Query, Share of Voice, Conversion Performance, or any of the 41 types in the reference table above) → select columns → choose date range and granularity → run or schedule. Scheduled reports accept a recurrence (daily, weekly, monthly) and an email address list. The format defaults to CSV; select TSV or XML in the format dropdown if downstream processing requires it.

For the absorbed intent around "how to create and interpret reports," the most common first-report sequence is: (1) Keyword Performance report with columns for Quality Score, impression share, IS lost to budget, and IS lost to rank — this surfaces the account’s competitive position and Quality Score diagnostic in one download; (2) Conversion Performance report with Conversions, All Conversions, view-through, assists, and cost per conversion — this shows whether the conversion layer is configured correctly and which goals are producing; (3) Search Query report to identify irrelevant queries and negative keyword opportunities.

Advanced Reporting: Spreadsheet Analysis and Reconciliation Workflows

Exported CSV and TSV reports from Microsoft Advertising feed directly into pivot-table and dashboard workflows. The practical patterns that generate the most insight from raw exports:

Period-over-period delta tables. Export Keyword Performance for the current period and the prior period, join on keyword ID or keyword text, and calculate deltas for CTR, CPC, impressions, conversions, and IS. A keyword where CTR dropped 30% period-over-period but impression share held flat points at a creative problem; a keyword where impression share dropped 20% alongside stable CTR points at a budget or bid problem. The IS-lost split completes the diagnosis.

Conversions vs. All Conversions reconciliation. In a pivot table, segment Conversions and All Conversions by campaign or ad group. The gap between the two columns represents cross-device and view-through interactions. For Search campaigns where the gap is large, the account is capturing intent on one device and converting on another — which informs match-type and device bid-adjustment decisions. For Audience Network campaigns, a large gap is expected and correct; the view-through column is the primary measurement signal for those placements.

GA4 reconciliation. Export Conversions from Microsoft Advertising and compare against the equivalent conversion event in GA4 — filtered to Microsoft Ads / CPC source/medium. Discrepancies above 10–15% indicate either UET tag misfires (double-firing or under-firing), attribution window mismatches (Microsoft’s default lookback vs. GA4’s session model), or cross-device conversions that GA4 attributes to a different session. The Microsoft Ads platform conversion count and the GA4 count serve different purposes — neither is wrong, but using them interchangeably produces incorrect CPA and ROAS figures. For accounts at Austin or Chicago scale, this reconciliation is a monthly reporting step, not an optional audit. The Microsoft Ads vs. Google Ads comparison pillar covers how this cross-platform attribution complexity differs from Google’s approach.

Frequently Asked Questions

Data and methodology: Keyword volume from Ahrefs, June 2026; GSC position data from Mbadv property (90-day window, March–June 2026). Quality Score component definitions from Microsoft Learn, Quality Score and Quality Impact in Depth (learn.microsoft.com/en-us/advertising/msa-help/hlp_ba_conc_aboutqualityscore). Attribution model definitions from Microsoft Learn, AttributionModelType Value Set (bingads-13, updated February 2026). Report type counts from Microsoft Learn, Report Types (bingads-13, April 2026, updated June 15, 2026). Conversion column definitions from Microsoft Help Center, Search Engine Land, and Adobe Advertising. Impression share / Share-of-Voice metrics and PMax extension (January 15, 2026) from Microsoft Advertising Help and Adsroid. Data-Driven Attribution rollout (end of May 2026) from ppcnewsfeed.com and ALM Corp. Activate 2026 announcements (May 19, 2026) from ALM Corp recap and Microsoft Advertising official blog. UET consent mode mandatory deadline (May 5, 2025) from Usercentrics and CookieInformation. No Microsoft-specific CPC/CTR/CPA-by-industry benchmark figures exist; none are cited. Reviewed by MB Adv Agency, June 2026.