Skip to main content
Card class: Sensitive • Category: Ad Platform

​
At a glance

Real-time alert for a sudden spike of zero-conversion spend on the Microsoft Audience Network. The Audience Network is on by default and uses auto-extension to find new placements, which means it can drift onto low-quality publishers and apps very fast. Left unchecked, that drift can eat a meaningful slice of the week’s budget on traffic that never converts. This card catches the burst early, before the budget is gone, so you can exclude the placements or dial back the network rather than discover the waste in next week’s report.
What it watchesSpend that produces no conversions on Microsoft Audience Network placements, evaluated as a short-window burst rather than a slow monthly total.
Why Audience Network specificallySearch spend is intent-driven and self-limiting. The Audience Network is impression-driven and auto-extends to new placements, so it is the surface most prone to sudden low-quality drift.
What “wasted spend” means hereSpend on Audience Network placements with zero attributed primary conversions inside the recent window.
Cost basisGross media cost, post Click Quality filter. The figure Microsoft will bill you for.
CurrencyAccount currency.
Conversion attributionUET-based primary conversions, default last-click cross-device. View-through excluded by convention. A placement counts as wasted only if it produced no primary conversions in the window.
Bot / invalid trafficSpend is post-Click-Quality-filter, but the Audience Network sees more low-engagement and accidental clicks than Search, which is part of why drift here is costly.
Time windowReal-time burst detection over a recent multi-day window so a single quiet placement does not trigger it.
Alert triggerZero-conversion spend on any Audience Network placement crossing an account-tuned amount inside the recent window. The threshold scales with your normal Audience Network spend.
Rolesowner, marketing, finance

​
Calculation

Calculated automatically from your Microsoft Ads (Bing) data. See the At a glance summary above for what the metric tracks and the worked example below for a typical reading.

​
Worked example

A US DTC supplement brand runs Search plus a modest Audience Network presence for retargeting. The alert evaluates over the trailing 72 hours on 18 Mar 26.
PlacementNetworkSpend (72h)ConversionsIn burst set?
Cart-abandon retargetAudience Network$2106No
Lookalike prospectingAudience Network$6400Yes
”News aggregator app cluster”Audience Network$4800Yes
Brand SearchSearch$39041No (Search)
Non-brand SearchSearch$72022No (Search)
The burst set is **1,120∗∗ofzero−conversionAudienceNetworkspendinthreedays,againstanormalAudienceNetworkrun−rateofroughly1,120** of zero-conversion Audience Network spend in three days, against a normal Audience Network run-rate of roughly 250 per week. The alert fires. What it means:
  1. This is drift, not a strategy change. The cart-abandon retargeting placement is still converting normally. The damage is concentrated in newly auto-extended placements, the lookalike prospecting and an app cluster the network found on its own.
  2. The app cluster is the textbook drift symptom. Auto-extension reaches into the Microsoft Audience Network’s app and publisher inventory, which can include low-intent placements that generate clicks and zero purchases. These are the first to exclude.
  3. The pace is the alarm, not the absolute number. 1,120overaquietmonthwouldbetolerable;1,120 over a quiet month would be tolerable; 1,120 in 72 hours against a $250 weekly norm means the whole week’s budget could be consumed before anyone notices.
  4. Hold the budget, do not just pause and walk away. Exclude the offending placements, tighten the Audience Network bid or split it into its own campaign with a hard cap, and recheck. The retargeting that still works should keep running.
The decision flow when this fires:
  • Drift concentrated in new placements with proven retargeting still converting = exclude placements, keep the network.
  • Zero conversions across all Audience Network placements at once = check the UET tag first; this may be a tracking break, not drift.
  • Burst coincides with a budget increase = the network spent the new headroom on cheap inventory; cap it and force the budget back to Search.

​
Sibling cards merchants should reference together

CardWhy pair it with Wasted-Spend Burst
Microsoft Ads Wasted SpendThe keyword-level cousin over the full window. This card is the fast Audience Network burst; that one is the slow account-wide total.
Microsoft Ads Zero-Conversion SpendThe whole-campaign view of the same idea. A drifting Audience Network campaign often surfaces here too.
Microsoft Ads Spend by DeviceAudience Network skews differently by device than Search; helps confirm where the burst landed.
ROAS Dropped Below ThresholdDrift drags blended ROAS down. If both fire, the Audience Network is the likely ROAS leak.
UET Tag Tracking BrokenRule this out first. A tag break can make healthy Audience Network spend look like waste.
Microsoft Ads Overspending CampaignsIf the burst came with a budget increase, this card shows which campaign absorbed it.

​
Reconciling against Microsoft Advertising

Where to look in Microsoft Advertising: Microsoft Advertising → Reports → Performance → Website (Publisher) report, filtered to the Audience Network, with Spend and Conversions columns and a short date range. Sort by spend descending and read off the placements with zero conversions. Summing those over the recent window should match this alert closely. Other Microsoft Advertising views that look related but differ:
  • The Network segment on a campaign report shows Audience Network versus Search split but not the per-placement breakdown; you need the publisher or website report to find the specific drifting placements.
  • Auto-targets and dynamic placements can appear under different labels than the campaigns you built; the spend is still yours and still counts here.
Why our number may legitimately differ from the Microsoft UI (rare):
ReasonDirection of divergence
Real-time ingestion lag. Conversions can arrive hours after the click, so very recent placements may look worse than they will settle.Ours higher on the freshest window
Placement attribution. Microsoft may roll some app inventory into grouped labels.Per-placement totals may not line up one-to-one
Burst window vs UI date range. The alert uses a rolling recent window; a manual UI range rarely matches it exactly.Either
Cross-connector reconciliation:
CardExpected relationshipWhat causes legitimate divergence
google_ads.gads_wasted_spendGoogle’s Display Network drift is the direct analogue; a brand prone to one is often prone to the otherDifferent inventory and auto-extension rules mean the timing rarely matches.
shopify.total_revenueIf commerce revenue held while this burst grew, the Audience Network spend genuinely added no salesLate-converting awareness traffic can occasionally show value downstream; check assisted conversions before fully writing it off.

​
Known limitations / merchant FAQs

What is Audience Network drift and why does it happen so fast? The Microsoft Audience Network uses auto-extension to find new placements across its publisher and app inventory. When it has budget headroom it can move into fresh, untested placements within hours. Some of that inventory is low-intent and produces clicks but no purchases. Because it is impression-driven rather than intent-driven like Search, the waste accumulates quickly, which is why this is a burst alert rather than a monthly total. Should I just turn off the Audience Network? Not necessarily. Audience Network retargeting often converts well; it is the prospecting and auto-extended placements that drift. The better move is to split the Audience Network into its own campaign with a hard budget cap and placement exclusions, so a drift event cannot raid the Search budget. Turn it off only if it never converts for your account. The burst showed zero conversions everywhere on the network at once. Is that really drift? Probably not. Drift is usually concentrated in new placements while proven ones keep converting. If the entire Audience Network went to zero simultaneously, suspect a UET tag break first and check UET Tag Tracking Broken. A measurement failure can make healthy spend look wasted. How do I stop drift from recurring? Three durable controls. First, give the Audience Network its own campaign with a capped budget so it cannot borrow from Search. Second, build a placement exclusion list and add to it whenever this alert fires. Third, set a tighter Audience Network bid so the network is less able to chase cheap, low-quality impressions. Why a 72-hour-ish window instead of the full month? Because the danger is pace. A small monthly waste total is tolerable, but the same amount spent in three days means an out-of-control placement is burning budget right now. A short window catches the run-rate spike while there is still budget left to protect. Could a late-converting awareness placement be unfairly flagged? Occasionally. Some Audience Network impressions assist conversions that close days later or on another channel. Before permanently excluding a placement, check whether it shows assisted or view-through value downstream. If it never contributes anywhere, exclude it; if it consistently assists, treat it as upper funnel and judge it on a longer window.

​
Tracked live in Vortex IQ Nerve Centre

Wasted-Spend Burst (Audience Network drift) is one of hundreds of KPI pulses Vortex IQ tracks across Microsoft Ads (Bing) and 70+ other ecommerce connectors. Nerve Centre runs the detection layer; Vortex Mind investigates the cause when something moves; Ask Viq lets you interrogate any number in plain English. Start for free or book a demo to see this metric running on your own data.