Wasted Spend, Criteo

Card class: Hero • Category: Ad Platform

​

At a glance

Spend on Criteo line items, audience segments, and catalogue product-sets that produced zero attributed conversions in the trailing 30 days despite receiving meaningful spend (>3 days live, >€100 spend threshold). On Criteo this surfaces (a) audience pools that match poorly to converting users (often iOS-attribution-stress fallout), (b) catalogue product-sets that render dynamic ads but never convert (low-margin or out-of-fashion SKUs), (c) Commerce Media publisher placements with no buy-intent traffic. Typical “healthy” Criteo accounts have 5 to 12% of spend in this bucket; >15% is a signal to audit the audience and catalogue layers.

​

Calculation

Calculated automatically from your Criteo 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 French electronics DTC retailer running 14 active Criteo line items. The 30-day window is 02 Apr 26 to 01 May 26. Account currency EUR. Total Criteo spend €27,000. What this tells you:

1. 15.8% of Criteo spend is in the wasted bucket, above the 5 to 12% healthy band. On a €27k monthly spend, this is €4,260 / month, €51,000 / year, that’s a real number. The first action is to pause or restructure the offenders, not to defend them.
2. The clearance-accessories product-set is the biggest single offender. Two routes: (a) pause it entirely and reallocate budget to the converting product-sets, or (b) audit catalogue freshness (out-of-stock SKUs, image quality, pricing) before pausing. Most catalogue-driven wasted spend is fixable at the feed level, not by pausing the line item.
3. Lapsed-365-days audience is structurally a poor fit for retargeting. Users who haven’t visited in 365 days are not retargetable in Criteo’s strongest sense; they’re effectively prospecting. Either move them into a Commerce Media prospecting campaign with proper prospecting bid strategy, or exclude them entirely.
4. Commerce Media tech-news vertical is a context fit issue. Tech-news readers are in research mode, not buy-intent. Try a different vertical (review sites, comparison shopping content) or move spend to lower-funnel campaigns.
5. Cart abandoner >14 days is past the recoverable window. Cart-abandoner intent decays fast: 80% of recovered carts close within 7 days. Beyond 14 days, the audience has either purchased elsewhere or abandoned the consideration entirely. Tighten the window to 7 days.
6. The refurbished-items product-set is an image-quality problem. Criteo’s image-quality filter flagged the SKUs but didn’t fully block them; the ads serve at lower priority and against poorer placements, conversion rate collapses. Re-shoot the product images or escalate to Criteo client services to lift the flag.
7. Cross-checking against cri_total_spend: account ROAS would lift from 7.8× to ~9.2× if all €4,260 was reallocated to the converting line items at average ROAS.

Decision tree for each row:

• Catalogue-driven (product-sets): audit feed first; pause as last resort. Most product-set waste is fixable.
• Audience-driven (segments): audit definition first; tighten or exclude. Pause is acceptable for clearly-cold audiences.
• Commerce Media-driven: audit context fit first; reassign to better-fit verticals. Pause if no fit found in 14 days.
• Sponsored product (retail-media): audit margin and pricing first; the SKU may need a price adjustment, not a campaign pause.

​

Sibling cards merchants should reference together

​

Reconciling against the vendor’s own dashboard

Where to look in Criteo’s own dashboard:

Criteo Management Centre → Reporting → Custom Report → Group by Line Item or Audience Segment, sort Conversions ASC.

Criteo doesn’t surface “wasted spend” as a dedicated UI; reconcile by sorting the Custom Report ascending by conversions and filtering to spend > €100. The bottom rows match this card’s wasted-spend list. For catalogue product-sets specifically, Reporting → Catalogue Performance → Sort by Conversions ASC shows under-performing product-sets. Why our number may legitimately differ from Criteo: Cross-connector reconciliation: Wasted spend is platform-specific by definition; cross-connector reconciliation is at the audience-level diagnostic.

---

​

Known limitations / merchant FAQs

A line item is in the “wasted” list but I think it’s working; what’s wrong? Three usual causes for a false positive on Criteo:

1. Conversions API outage, iOS / Safari conversions silently dropped. Audit Criteo Pixel & Tag Manager for CAPI errors; restore the feed and the line item recovers in 7 days.
2. View-through lookback hasn’t settled, the line item just hit the 30-day spend threshold and view-through hasn’t credited yet. Wait 7 days.
3. Conversion-tag misfire, the storefront’s Criteo pixel has a CSP block, GDPR-consent regression, or a script-loader race. Check tag-firing rate in Criteo Tag Manager.

If none of those, the line item is genuinely wasted; pause or restructure. What’s a healthy wasted-spend percentage on Criteo? 5 to 12% on accounts running 8+ line items. Below 5% suggests you’re not pushing the auto-bidder hard enough (some experimentation cost is worth paying for new audience and product-set discovery). Above 15% is a cleanup signal; pause or restructure offenders. Above 25% indicates a structural problem (feed health, attribution tag, or audience saturation) and warrants a full account audit. Should I just pause everything in the wasted list? No. The right order is (1) audit the feed, (2) audit the tag, (3) wait 7 days for view-through, (4) restructure the line item (tighter audience, narrower product-set), (5) pause as last resort. Pausing is irreversible in the sense that the line item loses learning data; restoring later means starting from cold. If a line item has been wasted for 60+ days, pause is fine; if 30 days, restructure first. Why does my catalogue product-set show wasted spend if the products are in stock? Three usual causes: (1) image-quality flag on a subset of SKUs, the product-set serves but at lower priority and against worse placements; (2) price-mismatch suppression, Criteo’s price audit found feed-vs-landing-page mismatches; (3) low-margin SKUs that get clicks but checkout drops because the landing-page experience is poor. Audit (1) and (2) in Feed Manager → Diagnostics; audit (3) by checking landing-page conversion rate for the affected SKUs. My audience segment is in the wasted list but Criteo’s auto-bidder put it there originally; whose fault is that? Criteo’s auto-bidder optimises for spend pacing first, then ROAS. If pacing pressure was high (the campaign-group budget needed to fill), the bidder may have spent into a sub-optimal audience to hit cap. The line item is wasted; the bidder is doing what it was told. The fix is to tighten the campaign-group budget cap or raise the CPA target so the bidder doesn’t spend into low-quality auctions to hit pacing. How do I distinguish iOS-attribution-stress wasted spend from real wasted spend? Cross-check the iOS share of impressions for the line item. If iOS impressions are >60% of the line item’s total impressions and the rest of the line items in the same campaign group with similar audience definitions are converting, the iOS-attribution stress is the likely culprit. Restore CAPI feed integrity and re-evaluate after 14 days. If iOS share is <40% and conversions are still zero, the line item is genuinely wasted. Should I clean wasted spend before or after a major flight (Black Friday, Christmas)? Before. Going into a high-stakes flight with 15% of spend wasted means 15% less budget available for the converting line items at the most competitive auction time. Clean 14 to 21 days before the flight to give the auto-bidder time to redistribute learning into the converting line items. Can I automate the cleanup? Partially. Vortex IQ’s Wasted Spend alert flags the offenders the moment they cross the threshold; the actual pause/restructure decision needs human judgment because of the false-positive sources (CAPI, view-through, feed). The right automation is alert-then-review, not alert-then-pause. My Commerce Media line items always show as “wasted” because of the longer consideration window; is that right? Commerce Media wasted-spend reads can be noisier because the consideration window for prospecting audiences is longer (14 to 30 days from first impression to conversion vs 1 to 7 days for retargeting). Use the 30-day spend threshold as a minimum but extend the conversion lookback to 30 days for Commerce Media specifically. If after 30 days a Commerce Media line item still has zero conversions, it’s genuinely wasted. Does pausing a wasted line item actually lift account ROAS? Yes, mechanically. Pausing a line item with €1,000 spend and zero conversions removes that spend from the denominator while leaving the numerator unchanged. Account ROAS lifts proportionally. Realistic ROAS lift after cleaning 10% wasted spend is 8 to 14% on the account level. Reallocating the freed budget to converting line items lifts further (the marginal ROAS of the converting line items determines the net gain).

---

​

Tracked live in Vortex IQ Nerve Centre

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