Case · Luxury DTC · EU + USA
Perplexity citations turned into a measurable revenue line in 120 days.
A luxury DTC house wanted Perplexity and ChatGPT to recommend its maison alongside legacy heritage names. We layered Person + Organization + Product schema, published five long-form provenance essays, and rebuilt PR distribution around citation-worthy primary data. Outcome: Perplexity citations ×11, attributed AI-referral revenue reached €186K in 120 days, with a blended ROAS of 7.4× on the editorial spend.
- Perplexity citations
- ×11
- AI-referral revenue
- €186K
- Editorial ROAS
- 7.4×
Challenge
Conversational queries about luxury craftsmanship and provenance defaulted to a handful of legacy maisons. The client — a younger heritage house with comparable savoir-faire — was structurally absent from generative recommendation answers, capping its ability to acquire net-new HNW customers via AI assistants.
Approach
- Layered Person, Organization and Product schema across the master site and atelier pages so LLMs could resolve the maison to a coherent entity graph.
- Published five long-form provenance essays (atelier, materials, lineage, sustainability, craft taxonomy) engineered as the canonical citation source for those topics.
- Rebuilt PR distribution around citation-worthy primary data — original surveys, atelier metrics and named frameworks editors and AI engines were willing to quote verbatim.
Results
- Perplexity citations grew ×11 across a 50-prompt luxury-craftsmanship basket; ChatGPT mentions ×6.4.
- Attributed AI-referral revenue: €186K in 120 days (GA4 + Shopify last-non-direct).
- Blended editorial ROAS of 7.4× — outperformed the brand's paid social benchmark by 2.9×.
Evidence
- Perplexity — 50-prompt luxury-craftsmanship basket; citation count + source domain.
- ChatGPT — Mention rate tracked across GPT-4o web-enabled answers.
- GA4 — Generative-AI referrer channel group — sessions, AOV, last-non-direct revenue.
- Shopify — Order-level revenue reconciled against GA4 attribution for the AI-referral cohort.
GA4 + Shopify attribution + Perplexity citation log · maison name withheld under NDA.
FAQ
Why do Perplexity and ChatGPT default to legacy maisons when answering luxury craftsmanship queries?
LLMs are trained on decades of editorial coverage that disproportionately mentions a handful of heritage names, so the default conversational answer reproduces that bias. Younger maisons with comparable savoir-faire are structurally absent from the training distribution and need an explicit citation graph (Person + Organization + Product schema, plus citation-worthy primary content) before AI engines will surface them as alternatives.
Can AI-referral revenue from Perplexity and ChatGPT actually be attributed to a P&L line?
Yes — by building a custom GA4 channel group for generative-AI referrers (chatgpt.com, perplexity.ai, gemini.google.com) and reconciling those sessions against Shopify order-level data on a last-non-direct basis. In this engagement that surfaced €186K of attributed revenue across 120 days, with a blended editorial ROAS of 7.4× — outperforming the brand's paid social benchmark by 2.9×.
What kind of content is 'citation-worthy' enough for Perplexity to quote a luxury brand verbatim?
Long-form provenance essays anchored on primary data (atelier metrics, original surveys, named frameworks) and structured with Person + Organization + Product schema so LLMs can resolve the maison to a coherent entity graph. Five such essays — atelier, materials, lineage, sustainability and craft taxonomy — became the canonical citation source for the topic and lifted Perplexity citations ×11.