You browse a €2,500 coat on a luxury website. The next morning, the same coat follows you across every site you visit. Same white-background product shot. Same "Buy Now" button. Same generic banner template. For three weeks.
That is what retargeting looks like when nobody thought about brand experience. And it is why luxury CMOs wince when someone suggests retargeting.
But here is the problem with avoiding it entirely. Visitors exposed to Criteo dynamic retargeting for a luxury e-commerce platform converted at 2.6x the rate of unexposed visitors. Six percent of total sales were attributed to retargeting. Cross-device retargeting accounted for 10% of attribution (Source: Criteo success story). For a platform generating €130M in revenue with 90% online, that 6% is not a rounding error. It is roughly €7M in annual revenue.
The question is not whether luxury brands should retarget. It is how to retarget without destroying the brand experience that makes the product worth €2,500 in the first place.
The Positioning Decision That Changes Everything
We built the retargeting strategy on a specific philosophy: the brand is a fashion consultant, not a fashion store. Every customer touchpoint, including the ads that follow them across the web, sells a dream. An aspiration. A version of themselves they want to become. Not a product. Not a transaction.
This is not marketing language. It is an operational constraint that shapes every execution decision.
A standard retargeting ad shows the product the customer viewed. White background. Price. "Buy Now" button. Functional. Efficient. And completely stripped of the emotional context that makes luxury purchasing work. The ad reduces a €2,500 coat to a commodity listing. It says "you left this behind" rather than "imagine wearing this."
We designed retargeting creative that continued the brand story rather than interrupting it.
| Element | Standard Retargeting | Luxury Retargeting (Our Approach) |
|---------|---------------------|----------------------------------|
| Product image | White background, e-commerce shot | Lifestyle context, styled on model, aspirational setting |
| Products shown | Only the viewed product | Viewed product + complementary items ("complete the look") |
| Call to action | "Buy Now" / "Add to Cart" | "Discover the Collection" / "Complete the Look" |
| Creative format | Static banner, single product | Dynamic multi-product, editorial layout |
| Emotional tone | Transactional reminder | Aspiration continuation |
| Placement quality | Broad network, any publisher | Curated premium inventory, luxury-adjacent contexts |
| Frequency | 15-20 exposures over 30 days | 5-8 exposures over 14 days, then suppressed |
The "complete the look" approach was the key creative innovation. Instead of showing only the coat the customer viewed, the ad presented that coat styled with a complementary scarf, bag, and shoes. Three products, one vision. The ad functioned as a digital stylist recommendation rather than a purchase reminder. It added value (styling inspiration) rather than just applying pressure (you did not buy this yet).
The Creative Quality Problem at Scale
Dynamic retargeting generates creative automatically from the product feed. That is what makes it scalable. It is also what makes it dangerous for luxury.
The product feed contains e-commerce photography. Clean, informational, shot for the product page. Not shot for brand advertising. A cropped product image in a generic banner template looks nothing like the brand's campaign photography. The visual gap between "how the brand presents itself" and "how the retargeting ad looks" creates brand dissonance that accumulates with every exposure.
We worked with Criteo to solve this at three levels.
Image library override. For hero products (the top 200-300 SKUs most likely to be retargeted), we provided brand-approved lifestyle imagery that overrode the standard product feed images. These images matched the brand's advertising standards: shot in context, properly art-directed, consistent color grading. The cost of producing this additional imagery was modest compared to the campaign spend. The impact on brand perception was significant.
Template customization. Standard Criteo templates look like standard Criteo templates. We built custom ad templates that reflected the brand's visual identity: typography, color palette, layout proportions, logo placement. The ads still used dynamic product insertion, but the frame around the product felt like it belonged to the brand rather than to an ad network.
Placement curation. A luxury retargeting ad appearing next to clickbait content on a low-quality publisher damages brand perception regardless of how beautiful the creative is. We configured strict placement controls excluding categories of publishers incompatible with luxury positioning and concentrating delivery on premium editorial environments. Lower reach. Higher quality. Better brand outcome.
Cross-Device: Connecting the Fragmented Journey
The 10% cross-device attribution number reveals a structural issue that most luxury brands underinvest in solving.
Standard cookie-based retargeting operates within a single browser on a single device. When the same customer browses on mobile Chrome and later switches to desktop Safari, the retargeting system sees two unrelated people. Two consequences follow simultaneously.
First, overspending. The system shows ads to what it thinks are two separate prospects, doubling the impression cost for a single customer. With strict frequency caps designed for one user, the actual per-user frequency is double the intended limit because each device receives its own allocation.
Second, undermeasurement. A retargeting impression on mobile that primes a desktop conversion gets zero credit because the two devices are not connected in the system's view. The mobile campaign looks ineffective. The desktop conversion looks organic. Both assessments are wrong.
Criteo's cross-device matching connected these fragmented identities. The system recognized multi-device users and treated them as single individuals. Frequency caps applied across devices, not per device. Attribution tracked the full journey, crediting the mobile impression that started the consideration even when the conversion happened on desktop.
For luxury e-commerce, where multi-device journeys are the norm rather than the exception, this capability is not an advanced feature. It is a baseline requirement for accurate campaign management.
Frequency: The Line Between Visible and Intrusive
In mass retail, high frequency works. Show someone a €30 product 15 times and the repetition eventually breaks through. The low price reduces the decision friction enough that persistence pays off.
In luxury, high frequency is counterproductive. Showing someone a €2,500 coat 15 times across two weeks communicates desperation. It says the brand needs this sale. Luxury should communicate the opposite: the product will sell whether this particular customer buys it or not. Exclusivity, not availability, drives desire.
We set frequency caps significantly lower than Criteo's standard e-commerce recommendations. The exact numbers were tested and refined, but the principle was consistent: enough repetition to maintain mental availability, not enough to trigger annoyance. After the cap was reached, the user was suppressed from the campaign entirely rather than receiving additional impressions at reduced bid prices (which is the default behavior in many platforms).
The frequency testing produced a counterintuitive finding. Reducing frequency from the platform default did not reduce conversion volume proportionally. It reduced it marginally. The relationship between frequency and conversion is not linear for luxury. The first 3-4 exposures do most of the work. Exposures 5 through 15 add diminishing returns that are offset by the brand perception damage they cause. Less was genuinely more.
Audience Segmentation: Who Gets Retargeted Matters More Than How
Not every site visitor should enter the retargeting pool. Indiscriminate retargeting wastes budget and contaminates campaign metrics. We built explicit inclusion and exclusion rules.
Inclusion criteria. Users who demonstrated genuine purchase intent: product page views with time on page above a meaningful threshold, add-to-cart actions, wishlist saves, size guide interactions, return visits to the same product within 7 days. These signals separate serious consideration from casual browsing.
Exclusion: recent purchasers. A customer who just spent €2,000 should not see ads for the product they already own. The post-purchase window (7-14 days) received different creative: complementary products, styling content, brand storytelling. Not purchase reminders. Showing someone an ad for what they already bought communicates that the brand does not know them. For luxury, where the customer relationship is the core asset, that is unacceptable.
Exclusion: chronic non-converters. Some visitors browse daily without ever purchasing. Industry researchers. Competitors monitoring pricing. Habitual window shoppers. Retargeting them wastes budget on impressions that will never convert and inflates engagement metrics without generating revenue. We identified these patterns through behavioral analysis (high-frequency visits with zero add-to-cart events over 60+ days) and excluded them.
Exclusion: price-sensitive behavioral signals. Users who exclusively visited during sale periods, used coupon codes on previous purchases, or exhibited comparison-shopping behavior across competitor sites (detected through referral patterns) were deprioritized. Retargeting price-sensitive users at full price rarely converts and risks conditioning them to expect promotions.
The resulting audience was smaller than the default "everyone who viewed a product" pool. It converted at dramatically higher rates because every user in the pool had demonstrated genuine luxury purchase behavior.
The Scale Story
The retargeting strategy existed within a broader digital transformation that scaled the operation from 10-12 employees to approximately 200, growing revenue to €130M with 90% from online (Source: Criteo success story).
At 10 employees, personalized follow-up was human. A sales associate who remembered a client's taste would call to mention a new arrival. That personal attention was the brand experience.
At 200 employees serving millions of annual visitors, that level of personal touch could not scale through human effort alone. Retargeting, done well, became the systematized version. Not a cold algorithm chasing customers. A digital extension of the associate's memory: "You looked at this coat. Here is how it looks styled for evening. And here is a scarf in the same cashmere that just arrived."
The philosophical frame matters. Retargeting is not pursuit. It is continuation of a conversation the customer started when they visited the site. The question is whether that continuation feels welcome (like a knowledgeable friend's suggestion) or intrusive (like a salesperson who followed them out of the store and down the street). Creative quality, frequency discipline, audience curation, and placement control determine which side of that line the campaign falls on.
Privacy and the Road Ahead
Cookie-based retargeting is contracting. Browsers are restricting third-party cookies. Privacy regulations require explicit consent. The technical infrastructure that has powered retargeting for 15 years is being dismantled.
This does not mean retargeting ends. It means the mechanism changes.
First-party data retargeting (using the brand's own customer data platform to build audience segments) is privacy-compliant and increasingly effective. Server-to-server integrations through platforms like Google Customer Match and Meta Conversions API enable retargeting without relying on browser-based cookies. Contextual targeting, which serves ads based on the content of the page rather than the identity of the user, offers a brand-safe alternative for luxury when individual-level targeting is not available.
We are already transitioning luxury retargeting strategies toward first-party data approaches. The brands that built their customer data infrastructure early (unified profiles, consent-managed data collection, first-party analytics) are best positioned. Those that relied entirely on third-party cookie infrastructure are facing a capability gap that takes 12-18 months to close.