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

​
At a glance

Spend on StackAdapt placements (publisher domains, app bundles, audience segments) that drove zero attributed conversions in the window. The “obvious savings” number; this is money that produced impressions/clicks but no purchases. DSP is multi-channel and multi-touch, so wait at least 14 days before acting; CTV and display view-through can take 3, 7 days to surface and BidCore is exploring constantly.
What it countsSpend (account currency) summed across publisher domains, app bundles, and audience segments that recorded zero conversions in the window but non-zero impressions/clicks. StackAdapt has no concept of “keywords”; the dimension is publisher/segment.
Cost basisWhatever the campaign was bidding on (CPM, CPC, BidCore CPA target).
CurrencyAccount currency.
Conversion attributionClick-through + view-through within the configured window. A placement whose impression later converted via Search is zero-conv here unless StackAdapt’s view-through window claimed it.
Attribution window30/1 default. CTV-heavy accounts should use 30/7 view-through; otherwise CTV placements look “wasted” when they’re not.
Bot / invalid trafficIVT-filtered; minor leakage <2%.
Time window30D.
Alert trigger>$0 (any zero-conv placement). Practical: worth acting on when wasted spend > 6% of total spend AND offending placements have >7 days of activity.
Rolesowner, marketing, finance

​
Calculation

Calculated automatically from your StackAdapt data. See the At a glance summary above for what the metric tracks and the worked example below for a typical reading.

​
Worked example

The same US homeware brand. Window 02 Apr 26 to 01 May 26. Total spend $44,000.
PlacementChannelSpend ($)ImpressionsClicksConversionsNotes
premium publisher rotationDisplay8,2002.4M4,200142Convertible
nytimes.comDisplay3,4001.1M1,80068Convertible
espn.com (CTV pre-roll)CTV4,200320kn/a22 (view-through)Convertible
buzzfeed.com (display)Display840480k9200Wasted, 18 days
app: candy crush (in-app display)In-App620620k1400Wasted, audience mismatch
streaming audio: regional sportsAudio480180k220Wasted, 21 days
app: weather widgetIn-App380920k840Wasted, click-fraud suspect
Other (60 placements, all converting)mixed25,8804.2M11,234562
Total wasted (this card)$2,3202.2M1,16605.3% of total spend
What’s interesting:
  1. 5.3% wasted spend is healthy band for a multi-channel DSP account; healthy DSP runs 4, 8% wasted. No structural issue here, but the candidates flagged are still actionable.
  2. The weather widget app ($380) is suspicious for click-fraud. 84 clicks on 920k impressions = 0.009% CTR; legitimate display CTR is 0.1, 0.4%. This is bot traffic that StackAdapt’s IVT filter missed; in-app weather widgets are a known low-quality bundle. Action: add to the bundle block list across all campaigns.
  3. Candy Crush in-app ($620) is genuine audience mismatch. Mobile gaming audiences don’t convert on homeware; the demographics overlap is poor. Add to block list and revisit if you launch a more impulse-purchase product line.
  4. buzzfeed.com display ($840) is borderline. The 920 clicks suggest engagement but no conversions despite 18 days of activity. Likely cause: the display creative is on entertainment articles where the audience is in scroll-mode, not purchase-mode. Either change creative (more native-like format), block the publisher, or accept as awareness spend.
  5. Regional sports streaming audio ($480) at 21 days zero-conv = block. Audio is hard to attribute and this audience is wrong; not worth the noise.
  6. Total potential savings $2,320 but in practice net savings are usually 60, 80% of headline because BidCore reallocates to the next-cheapest inventory which often costs more per click.
Quick sanity tests:
  • Wasted % rising over time = inventory degrading; BidCore finding more low-quality bundles.
  • Wasted % stable at 4, 8% = normal exploration cost.
  • Wasted % spiking on a single placement = audience mismatch or click-fraud; investigate.
  • High wasted % on in-app placements specifically = the in-app bundle list needs curating; mobile gaming and weather widget bundles are usual suspects.
  • Zero wasted spend = you’ve over-allowlisted and missed scaling opportunities.

​
Sibling cards merchants should reference together

CardWhy pair it with Wasted Spend
StackAdapt Zero-Conversion SpendThe campaign-level twin. This card is publisher/segment-level.
StackAdapt Total SpendContext. Wasted % of total is the meaningful read.
StackAdapt Spend by CampaignIf wasted spend concentrates on one campaign, the campaign-targeting needs work.
StackAdapt CTR by CampaignWasted spend correlates with low CTR (click-fraud bundles) or high CTR with no conversion (audience mismatch on engaged readers).
StackAdapt ROAS by CampaignThe campaigns hosting wasted-spend placements show poor ROAS.
The Trade Desk Wasted SpendSame structural pattern on enterprise DSP. Compare bundle overlap; some low-quality in-app bundles hit both DSPs.

​
Reconciling against the vendor’s own dashboard

Where to look in StackAdapt: StackAdapt → Reports → Site/App Breakdown for the same window. Filter to placements with zero conversions; spend column should match this card to within sub-percent rounding. StackAdapt does not surface “wasted spend” as a single number; this card derives it. Why our number may legitimately differ from a manual StackAdapt cut:
ReasonDirectionWhy
Time zoneBoundary days offAccount TZ vs UTC.
Long-tail conversionCard slightly high until catchupA “wasted” placement today may pick up 1, 2 conversions in days 7, 14 (especially CTV).
View-through window changesDirection dependsIf you shorten the window, more placements appear “wasted”; if you extend, fewer.
In-app vs web definitionNoneBoth this card and StackAdapt UI use the same publisher/bundle dimension.
IVT credit timingCard slightly high before credits postIVT spend stays in the wasted total briefly until refunds settle.
Cross-connector reconciliation:
CardExpected relationshipWhat causes legitimate divergence
the_trade_desk.the_wasted_spendIndependent. The two DSPs share much of the same SSP inventory; bundles wasted on one are often wasted on the other.Different publisher/bundle allowlists between accounts.
google_analytics.ga_sessions_by_sourceGA4 sessions on wasted placements should not show 100% bounceIf GA4 shows healthy engagement, the issue is conversion-tag, not audience quality.

​
Known limitations / merchant FAQs

Should I block every zero-conversion placement immediately? No. DSP has long conversion lag (especially CTV view-through); some placements take 14, 30 days for impressions to convert. Wait until a placement has 14+ days of activity with zero conversions before blocking. Acting too aggressively shrinks the inventory pool BidCore can use, often raising CPM on remaining inventory. What’s the “right” wasted-spend percentage on DSP? 4, 8% on a healthy account. Below 4% means you’ve over-allowlisted; above 8% means audience targeting or bundle curation needs work. CTV-heavy accounts run higher (6, 12%) because view-through tracking misses some legitimate conversions and falsely labels good placements as wasted. Why is in-app display so often in the wasted list? Mobile in-app inventory is a known low-quality category. Click-fraud is more prevalent (weather widgets, low-quality games), bundle metadata is often unreliable, and the audience demographics rarely match brand-target audiences. Action: maintain a curated bundle allowlist (e.g. only premium app publishers like NYT app, ESPN app) rather than relying on open-bundle exchange. My CTV placements show as wasted but I expect long view-through, what’s wrong? Check the view-through window setting. CTV with 1-day view-through almost always shows lots of “wasted” placements because TV-driven conversions take 3, 7 days. Set CTV view-through to 7 days and most CTV “waste” disappears. Why do similar bundles waste on StackAdapt and TTD but not on Google? Google’s DV360 buys through Google’s own ad exchange (AdX) which has tighter quality controls; StackAdapt and TTD buy through the broader programmatic ecosystem (Magnite, Pubmatic, OpenX, Index) which has more long-tail inventory. The fix is curation, not platform switching: build the same bundle allowlists across both DSPs. My wasted spend dropped sharply but ROAS also dropped, what happened? You over-blocked. Aggressive bundle/publisher blocks shrink the inventory pool BidCore can choose from; the algorithm bids up on remaining inventory, CPM rises, marginal conversions cost more than the wasted spend you saved. Net efficiency falls when you over-block. Use bundle-level CPM caps for borderline cases rather than hard blocks. Can I block specific placements within a publisher (e.g. nytimes.com/sports but not /news)? Limited. Section-level blocking works on some publishers via custom inclusion/exclusion lists, but most publishers route at domain level. The reliable lever is domain-level blocking; section-level is best-effort. What’s BidCore’s role in wasted spend? BidCore is exploration-driven by design; it deliberately tests new placements to find efficient ones. Some exploration cost is healthy. The card surfaces placements that have failed exploration (extended activity with no conversions); BidCore eventually de-weights them but does so probabilistically rather than blocking outright. How does this card differ from Zero-Conversion Spend? This card is publisher/bundle/segment-level (which placements within campaigns are wasting); Zero-Conversion Spend is campaign-level (which whole campaigns are wasting). A whole campaign with zero conversions is a bigger problem (likely pixel or goal misconfig) than a single bad placement.

​
Tracked live in Vortex IQ Nerve Centre

Wasted Spend is one of hundreds of KPI pulses Vortex IQ tracks across StackAdapt 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.