> ## Documentation Index > Fetch the complete documentation index at: https://docs.vortexiq.ai/llms.txt > Use this file to discover all available pages before exploring further. # Wasted Spend, MediaMath > Wasted Spend for MediaMath stores. Tracked live in Vortex IQ Nerve Centre. How to read it, why it matters, and how to act on it. **Metrics type:** [Key Metrics](/nerve-centre/overview#metrics-types-explained) • **Category:** [Ad Platform](/nerve-centre/connectors#connectors-by-type) ## At a glance > Spend on MediaMath placements (publisher domains, app bundles, audience segments) that drove zero attributed conversions in the window. **MediaMath caveat (May 26)**: post-restructure pixel instability can artificially classify good placements as "wasted" when the issue is platform-side. **Cross-check pixel health before acting on this card on MediaMath specifically.** | | | | -------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **What it counts** | Spend (account currency) summed across publisher domains, app bundles, and audience segments with zero `conversions` and non-zero impressions/clicks. | | **Cost basis** | Whatever the campaign was bidding on (CPM, CPC, Brain CPA target). | | **Currency** | Account currency. | | **Conversion attribution** | Click-through + view-through within the configured window. | | **Attribution window** | 30/1 default; 30/7 for CTV-heavy. | | **Bot / invalid traffic** | IVT-filtered. | | **Time window** | `30D`. | | **Alert trigger** | `>$0 (any zero-conv placement)`. Practical: act when wasted spend > 6% of total AND placements have >7 days activity AND pixel health is confirmed healthy (the third condition is MediaMath-specific). | | **Roles** | owner, marketing, finance | ## Calculation Calculated automatically from your MediaMath 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 enterprise homeware brand. Window 02 Apr 26 to 01 May 26. Total spend \$72,200. | Placement | Channel | Spend (\$) | Impressions | Clicks | Conversions | Notes | | ------------------------------------- | ------- | ----------- | ----------- | ------- | ------------------ | ------------------------------- | | Premium publisher rotation | Display | 14,200 | 3.8M | 5,800 | 168 | Convertible | | nytimes.com | Display | 4,800 | 1.4M | 2,100 | 84 | Convertible | | NBC CTV (PMP) | CTV | 28,400 | 0.9M | n/a | 142 (view-through) | Convertible | | **buzzfeed.com display** | Display | **1,200** | **620k** | **820** | **0** | **Wasted, 18 days** | | **app: weather widget** | In-App | **820** | **1.2M** | **84** | **0** | **Wasted, click-fraud suspect** | | **regional sports streaming audio** | Audio | **640** | **240k** | **42** | **0** | **Wasted, audience mismatch** | | Other (52 placements, all converting) | mixed | 22,140 | 2.6M | 3,756 | 312 | | | **Total wasted (this card)** | | **\$2,660** | **2.1M** | **946** | **0** | **3.7% of total spend** | What's interesting: 1. **3.7% wasted spend is healthy band** for an enterprise DSP account. No structural issue but the named placements are still actionable. 2. **MediaMath-specific check before acting**: confirm pixel health is OK on each "wasted" placement. Open T1 → Pixels → diagnostics, ensure conversion events fired correctly during the 30-day window. **If pixel was unstable, some "wasted" placements may be false positives**, no action needed beyond pixel-stability resolution. 3. **The weather widget app (\$820)** is a click-fraud suspect on a 0.007% effective CTR. Block the bundle. Standard DSP issue, not MediaMath-specific. 4. **buzzfeed.com display (\$1,200)** is borderline; 820 clicks no conversion suggests audience scroll-mode or display creative not native-fit. Test creative change before blocking. 5. **Regional sports streaming audio (\$640)** at 21+ days = block. Audience mismatch. 6. **Total potential savings \$2,660** but expect net 60, 80% of that after Brain reallocates. Quick sanity tests: * Wasted % rising over time = inventory degrading OR pixel issues. **Differentiate by checking pixel diagnostics first on this connector.** * Wasted % stable at 3, 7% = normal exploration cost. * Wasted % spiking on a single placement = audience mismatch or click-fraud; investigate. * Account-wide "wasted" suddenly with no inventory change = pixel outage on MediaMath; check T1 status. * Zero wasted spend = over-allowlisted, missed scaling. ## Sibling cards merchants should reference together | Card | Why pair it with Wasted Spend | | ------------------------------------------------------------------------------------------ | ---------------------------------------------------------------- | | [MediaMath Zero-Conversion Spend](/nerve-centre/kpi-cards/mediamath/zero-conversion-spend) | Campaign-level twin. | | [MediaMath Total Spend](/nerve-centre/kpi-cards/mediamath/total-spend) | Context. | | [MediaMath Spend by Campaign](/nerve-centre/kpi-cards/mediamath/spend-by-campaign) | Where wasted concentrates. | | [MediaMath Clicks vs Conversions](/nerve-centre/kpi-cards/mediamath/clicks-vs-conversions) | **Pixel-health canary, especially important on this connector.** | | [MediaMath ROAS by Campaign](/nerve-centre/kpi-cards/mediamath/roas-by-campaign) | Wasted-spend campaigns show poor ROAS. | | [The Trade Desk Wasted Spend](/nerve-centre/kpi-cards/the-trade-desk/wasted-spend) | Migration peer; bundle/publisher overlap analysis. | ## Reconciling against the vendor's own dashboard **Where to look in MediaMath T1:** [T1 → Reports → Site/App Breakdown](https://t1.mediamath.com/) for the same window. Filter to placements with zero conversions. T1 does not surface "wasted spend" as a single number; this card derives it. **Why our number may legitimately differ:** | Reason | Direction | Why | | ------------------------------------------ | -------------------------------- | ------------------------------------------------------------------------------ | | **Time zone** | Boundary days off | Org TZ vs UTC. | | **Long-tail conversion** | Card slightly high until catchup | Some placements pick up late conversions. | | **Pixel instability (MediaMath specific)** | Card high during pixel issues | Post-restructure pixel outages can falsely classify good placements as wasted. | | **View-through window changes** | Direction depends | Window changes reflow data. | **Cross-connector reconciliation:** | Card | Expected relationship | What causes legitimate divergence | | ------------------------------------------------------------------------------------------------ | --------------------------------------------------------------------------------------------- | --------------------------------------------------------------------- | | [`the_trade_desk.the_wasted_spend`](/nerve-centre/kpi-cards/the-trade-desk/wasted-spend) | Independent. Both DSPs share much SSP inventory; bundles wasted on one often wasted on other. | Different allowlists. | | [`google_analytics.ga_sessions_by_source`](/nerve-centre/google_analytics/ga_sessions_by_source) | GA4 sessions on wasted placements should not show 100% bounce. | If healthy GA4 engagement, the issue is conversion-tag, not audience. | ## Known limitations / merchant FAQs **My MediaMath wasted spend doubled overnight, what happened?** **Pixel health first, on this connector.** A doubling overnight is highly unlikely from a real change in placement quality; far more likely the conversion pixel had an outage, classifying converting placements as zero-conv. Open T1 → Pixels → diagnostics for the affected window. If pixel was healthy, then proceed to standard checks. **Should I block every zero-conversion placement on MediaMath?** No. Same DSP rule as elsewhere (wait 14+ days), plus the MediaMath-specific rule: **confirm pixel was healthy during the zero-conv window**. Blocking placements that were falsely classified doesn't recover budget; it just shrinks future reach. **What's healthy wasted-spend % on MediaMath?** 3, 7% on stable pixel weeks; up to 10, 12% during pixel-instability events (which then resolve themselves once the platform recovers). If wasted % is consistently above 10% across multiple stable months, the underlying inventory or audience targeting needs work. **Why is in-app display so often in the wasted list?** Same as on every DSP, mobile in-app inventory has higher click-fraud rates and weaker audience targeting metadata. Maintain a curated bundle allowlist; avoid open-bundle exchange purchasing on in-app. **My CTV PMP shows wasted spend, what's wrong?** Check view-through window first. PMP CTV with 1-day view-through almost always shows wasted spend; CTV converts on 3, 7 day delay. Set CTV view-through to 7 days and most CTV "waste" disappears. **My wasted spend dropped sharply after a pixel fix, was that real waste?** No. The "wasted" before was mostly false-positive (good placements with broken pixel attribution). The drop reflects pixel recovery, not actual placement quality change. **How does this card differ from Zero-Conversion Spend?** Same as on every DSP: this card is placement-level (publishers/bundles/segments); Zero-Conversion Spend is campaign-level (whole campaigns). Read both for full picture. **What's the MediaMath-specific blocking strategy?** Be conservative on this connector. False positives are more common; block-list mistakes shrink reach and the platform's reach is already compressed post-restructure. **Use bundle-level CPM caps for borderline cases**; reserve hard blocks for clear click-fraud or audience-mismatch placements. **Should I migrate the block list when migrating to TTD?** Yes, with translation. Most bundle/domain blocks translate directly (same SSP inventory). Audience-segment blocks may not translate (different segment taxonomies). Audit during migration. *** ### Tracked live in Vortex IQ Nerve Centre *Wasted Spend* is one of hundreds of KPI pulses Vortex IQ tracks across MediaMath 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](https://app.vortexiq.ai/login) or [book a demo](https://www.vortexiq.ai/contact-us) to see this metric running on your own data.