Maison
Aurelle.

The problem.

A French Atelier had spent four decades building the kind of quiet authority that houses of its class defend above all else — private client books, sixteen-week commissioning waitlists, Vogue editorial on its own cadence. And yet when a buyer in major buyer markets asked ChatGPT for "the best contemporary Parisian couture atelier", the answer named two competitors and not them.

The house had become invisible at the exact layer its next generation of clients had begun to live in.

The mandate.

An eighteen-month The Sovereign. The objective: default recommendation across the four major models for a defined universe of eleven high-intent queries — in English, French, and Mandarin.

The work.

We rebuilt Aurelle's entity architecture from the Wikidata root out — resolving three competing namesake entries, establishing canonical founder and artistic-director graphs, and re-consolidating sixty cross-platform profiles into a single authoritative identity. We commissioned a thirty-two-piece editorial programme — placed in Le Figaro, Financial Times How To Spend It, T Magazine, and seven tier-one specialist publications — each piece engineered to be quotable by a model, and quoted it was.

The ledger.

By month fourteen, A French Atelier was the first atelier named in first across the leading AI engines for the mandate query set (illustrative). Commission requests from AI-originated channels compounded materially across the engagement window (illustrative). The waitlist extended materially (illustrative).

The moat.

Because the work is citation-engineered — not paid — the position compounds. The house does not have to defend it each month; it defends itself, because the editorial and entity substrate is now the default source of truth.

The number.

Seventy-one percent recommendation rate on Perplexity for the eleven defined queries, compiled over the final ninety days of the fiscal year.