La Redoute
Paid social campaign fully automated by AI (audiences, placements, bids, creative combinations), measured incrementally
In 2024, La Redoute achieved +51% incremental ROAS and -26% cost per purchase by entrusting its acquisition campaigns to Meta's AI (Advantage+ Shopping), enriched with creator Reels, measured by Conversion Lift across Spain and Portugal.
Key points
- Paid social campaign fully automated by AI, measured incrementally.
- Meta Advantage+ Shopping enriched with creator Reels, measured through Conversion Lift.
- +51% incremental ROAS, +35% purchases, -26% cost per purchase.
- Evidence level B, status confirmed.
Objective
Sell more online in Spain and Portugal without adding campaign management overhead: Meta's AI drives delivery, the team focuses on the catalog and the creative.
The deployment
La Redoute, the French fashion and home retailer, handed its sales campaigns on Meta over to Advantage+ Shopping mode. The principle: the advertiser supplies the product catalog and the visuals, and Meta's algorithm decides on its own which audiences, placements, bids, and creative to serve. The team tested a variant by adding creator videos produced with the agency Ykone, served in Reels and Stories. To find out whether it really worked, it did not look at the ROAS displayed in the interface: it ran a Conversion Lift study with a control group, from June 7 to July 11, 2024, comparing Advantage+ alone against Advantage+ enriched with creator creative. The enriched variant delivered +51% incremental ROAS, +35% purchases, and -26% cost per purchase. A second test with the platform The Cirqle gives a consistent order of magnitude, with ROAS rising from 5.19 to 8.35.
Results Proof B
Official Meta case study, quantified and backed by a multi-cell Conversion Lift (a serious incremental method), but published by the platform itself; partially corroborated by a third-party test (The Cirqle) with revenue attribution.
How it works
Documented architectureThe stack in detail
- plateforme Meta Advantage+ Shopping targeting, bids, placements, and creative combinations driven by Meta's ML
- infra Conversions API (CAPI) server-side reporting of purchases, the signal that feeds the algorithm
- outil Meta Conversion Lift incremental measurement with a control group, the arbiter of the test
- integrateur Ykone production of the creator videos (Reels/Stories)
How it runs, concretely
For ops teams-
1Sync the catalog E-commerce / feed team
The product feed (price, stock, visuals) goes to Meta and becomes the ad inventory. A dirty or stale feed translates directly into an ad for an out-of-stock product.
-
2Make purchase measurement reliable Data team
Pixel plus server-side Conversions API, so that every sale is reported even when the browser blocks tracking. This is the fuel of the optimization.
-
3Load a stock of creative Creative agency
Feed in vertical videos and creator Reels. The algorithm tests the combinations itself; the team does not pick the winner for it.
-
4Let the learning happen Meta AI
The algorithm explores audiences, placements, and bids. Cutting a campaign too early or changing it every two days breaks the learning and restarts from scratch.
-
5Decide on incrementality Media team
The decision to keep or cut is made on a Conversion Lift with a control group, not on the attributed ROAS that inflates the figures.
The purchases reported by the Pixel and the Conversions API. If this signal is incomplete or delayed, the algorithm optimizes on noise and performance collapses without any visible reason.
How your customers perceive this type of use
Sourced studiesLe pricing algorithmique est le terrain le plus inflammable : 68% des consommateurs disent se sentir leses quand les marques utilisent le pricing dynamique et 80% jugent plus dignes de confiance les marques aux prix constants (Gartner, 2024). L'equite percue varie selon le secteur : le pricing dynamique n'est juge juste que par 33% a 40% des repondants selon qu'il s'agit de concerts ou de cinemas (YouGov, 17 marches). Le prix personnalise par les donnees individuelles est le plus rejete : 47% des Americains s'y opposent fermement (Consumer Reports, 2024).
Acceptance conditions
- La constance des prix comme signal de confiance : 80% jugent plus fiables les marques aux prix stables (Gartner 2024)
- Le secteur conditionne l'equite percue : le pricing dynamique est mieux tolere pour les cinemas (40% le jugent juste) que pour les concerts (33%) (YouGov 2024)
Red lines
- Le pricing dynamique percu comme abus : 68% se sentent leses (Gartner 2024)
- Le prix individualise a partir des donnees personnelles : 47% d'opposition ferme (Consumer Reports 2024)
- Les frais caches et hausses imprevues, vecus par 79% des consommateurs sur un an et associes a la perte de confiance (Gartner 2024)
Sources: Gartner 2024 · YouGov 2024 · Consumer Reports 2024
How to replicate
Inference, not sourcedData prerequisites
- Clean and up-to-date product catalog (feed)
- Pixel + Conversions API with reliable purchase events (50+ conversions/week)
- Customer lists for exclusion / capping existing customers
Org prerequisites
- Bank of vertical video creative (creators / UGC)
- Ability to let the algorithmic learning run without over-managing it
Possible stack
- Meta Advantage+ Sales Campaigns + CAPI (server-side GTM, Shopify, or partner)
- Measurement through Conversion Lift or geo-test
The plan, step by step
- Step 1Audit the product feed and the purchase measurementDeliverable: Clean feed (prices, stock, visuals) + valid Pixel/CAPI in Events Manager
- Step 2Build the creative bankDeliverable: 10-15 vertical videos (creators or UGC) + catalog ads ready
- Step 3Launch Advantage+ on 20-30% of the social budgetDeliverable: Live campaign, learning phase engaged (stop touching it)
- Step 4Let it converge, then measure incrementallyDeliverable: Conversion Lift or geo-test that decides against the manual structure
- Step 5Scale or arbitrate based on incrementalityDeliverable: Budget decision based on the real lift, not the attributed ROAS
First step: Launch a structured A/B test: an Advantage+ campaign (20-30% of the social budget) vs the existing manual structure, with Conversion Lift or geographic incrementality as the arbiter.
The same pattern, proven in other industries
Allwyn
AI-driven omnichannel programmatic buying with retail data activation
Nivea
genAI testing and optimization of creative elements + automated localization at scale
Clarins
AI-automated sales campaign plus dynamic catalog ads, evaluated on incrementality via Conversion Lift against last-click
Sources
- S1 La Redoute - Advantage+ Shopping + creator Reels (Conversion Lift multi-cellules) Interested party archive pending
- S2 How La Redoute drove a 61% ROAS lift with Creator ads - The Cirqle Interested party archive pending
- S3 Introducing New Automation Tools to Increase Sales and Drive Growth Interested party archive pending
An error, newer info, a source?
This page lives on its accuracy. If a figure has moved, if the deployment has changed, or if you have a higher-quality source, tell us. Every sourced correction is verified before publication.