The RealReal raised its prices on the Saint Laurent Mombasa before Saint Laurent brought the bag back. Glossy documented the sequence: Bottega Veneta launched its curved-handle Sardine, the platform registered renewed interest in the similarly shaped Mombasa, prices on the older bag rose as demand built, and Saint Laurent reintroduced the Tom Ford-era design for Spring 2026. A consignment marketplace is doing product research on other houses’ archives and giving the findings away, because The RealReal confirmed to Glossy that brands can request its demand data free of charge.

Noelle Sciacca, the company’s associate director of fashion merchandising and content strategy, describes what the platform sees before a brand does. Shoppers chasing a new collection’s references, she told Glossy, “are finding the original version of this new version that’s being put out there,” which means the desk “can almost see before the brand itself whether that new creative direction is resonating.” The claim is narrower than prediction and more useful than it sounds. A house measures a new collection by what clears at full price in its own stores, a number that arrives late and has the marketing spend baked into it. The platform measures the same aesthetic by what someone will pay for a twenty-year-old version of it, with nobody advertising.

The granularity is what brands cannot reproduce. The RealReal’s 2026 Resale Report, published on 25 August, counts saved searches by designer tenure rather than by house: “Phoebe Philo Celine” rose 650 percent. That query names a person, a house and a window of years, and it describes garments sold under a label the house has retired. The same report puts vintage demand up 432 percent since 2020, which sets the size of the pool those queries are drawn from. No brand’s customer database holds any of it, because the shoppers generating the signal are not transacting with the brand, and the objects they are pricing were sold once, years ago, to someone else.

The archive tells a house what it made; only a market tells it what any of that is worth with no campaign behind it.

Price is the part of the signal that does work. A Chloé Paddington that averaged $275 on the platform three years ago now averages $1,290, per figures the company gave Glossy, and Chemena Kamali put the bag back on the Chloé runway for autumn 2025. That climb brackets the reissue rather than cleanly preceding it, which is the honest weakness in reading resale as a forecast. A price is still a commitment in a way a search is not. Someone paid it, on an object with no waitlist, no campaign and no boutique behind it.

The strongest objection is that all of this is an echo rather than a signal. Searches for Matthieu Blazy rose 10,967 percent after his move from Bottega Veneta to Chanel, and the thing that moved them was a press release. If every documented spike traces back to an announcement a brand made first, the desk is reading the trade press with extra steps, and the leading indicator is a lagging one wearing better clothes. Glossy is careful here and says so outright: the data “establishes chronology rather than direct causation,” and Chloé, Saint Laurent and McQueen did not respond to requests for comment on whether resale demand informed their revivals. Concede the Blazy number in full. The Mombasa is a different case, because what moved that market was another house’s bag, and no quantity of Saint Laurent’s own data would ever have contained it.

The research costs nothing because The RealReal is not paid for it, and its income statement shows what it is paid for instead. The company reported $617 million of GMV for the June quarter, monetised through commission on goods that pass through its hands. Its take rate was 35.9 percent, which is the price of moving an object; nothing in it is charged for knowing which object to move. Rati Sahi Levesque told analysts the company has “built the largest authenticated luxury data set in the world,” and the uses she named were mostly internal: pricing, search, authentication, plus the tools members use to manage what they already own. Publishing the demand read is a supply instrument. It recruits consignors, flatters the houses named in it, and costs one press release in August.

Neither side has a mechanism to charge, and that mechanism is now being built for unrelated reasons. The RealReal joined EON’s digital ID network in January 2024, an integration EON describes as two-way: brand product data travels into resale, and post-sale events travel back. The EU’s Ecodesign for Sustainable Products Regulation, in force since July 2024, attaches a digital passport to products; the regulation itself says nothing about resellers or demand data. EON is the one arguing that this passport will standardise the exchange it already runs with The RealReal, which is the position of a vendor with a stake in that exchange existing, not a description of what Brussels wrote. If passports become the default carrier of provenance regardless, the demand read stops being a courtesy and turns into a negotiated term. Leverage in that negotiation is not where it looks: resale’s binding constraint is supply, not buyers, and that supply arrives from closets rather than from brands. What a house can actually withhold is legitimacy.

The choice is not whether to take the data, since taking it costs nothing and refusing it costs a season. It is whether to keep taking it on the seller’s calendar. A report published in August describes what moved through July, filtered by what the platform held in stock and wanted more of, and presented as a trend release. A brand that wants the read on its own schedule and scoped to its own silhouettes has to ask for it as a customer — which means conceding that a channel most houses have never authorised is running research on their archives. The alternative is to keep calling it free, and to keep learning what the house is worth from the people selling it secondhand.