© Courtesy of Art Basel
ARTHUR ANALYTICS: THE APP FOR THE ROOM
YOU WERE NEVER IN
In dialogue with James Crichton
FVTVRIST Magazine // Text by Anna S.
15 June 2026
The art market's information problem is not that data doesn't exist. It is that data has always existed in one room, and everyone else has been standing outside it. Auction results have been searchable for decades; that was never the hard part. The hard part is everything that happens before the hammer falls and after the handshake: the museum hunting a loan, the gallery about to move an artist, the underbidder who came second on a lot three times in a row. That information circulates, in group chats and phone calls and bilateral previews, but it circulates selectively. Knowing it has always meant being in the right room.
James Crichton wants to hand everyone the same map. A former Wall Street investment manager and a member of a multigenerational collecting family, Crichton founded Arthur Analytics in 2023 on the premise that the art market's information asymmetry is a structural problem with a structural solution. Arthur tracks auction history, primary market activity, museum demand, representation changes, and fair previews across 500-plus fairs, integrating them into a single platform any collector can open. Arthur Exchange, launched this year, lets collectors list works for sale, loan, or rent at no commission. Agentic search is in development. A paid tier is coming. By Crichton's account, the app is approaching 5,000 active users, growing almost entirely by word of mouth at over 400% a year.
None of it is complete, and Crichton is the first to say so. Private sales, the handshake deals that never hit a database, remain mostly dark. That is still where the real money and the real signal live.
On the eve of Art Basel in Basel, FVTVRIST sat down with James to ask how the platform actually works, where it falls short, and what it would mean to finally bring the private market into the open.
Portrait of James Chrichton, Founder of Arthur's Analytics.
© Courtesy of Arthur's Analytics.
How does Arthur know what has actually been sold? Auctions publish their results. Private sales, gallery transactions, off-market deals do not. Where does the data come from?
Three problems, three answers. We license the auction data — our partner covers about 99% of fine art and design volume back to the 1990s, the data insurers and lenders use. Bidder and underbidder activity, who competed and who came second, comes from The Baer Faxt.
That public auction layer is what the existing platforms solved long ago. Artnet, ArtPrice, MutualArt are auction databases at heart, and the category hasn't moved in decades. No knock — they did the first useful thing. But the art world treated a searchable record of hammer prices as the finish line, when it was only the starting line. We came at it from the other end: the private sales, the primary market, the signals that never hit a database.
That end is harder, and I won't pretend otherwise. Galleries submit some directly, because being seen sells the next work. Some comes from collectors and advisors. And some off-market deals we simply don't have, because two people shook hands in a room and that was that.
Nobody has complete primary-market data, us included. We have more of it in one place than anyone else, and it improves every week. The point isn't completeness — it's that the more of this lives in one place, the less you have to be an insider to buy well.
© Courtesy of Art Basel.
Fair previews are a tightly controlled product. Galleries time them, stagger them, price the same work differently for different clients. How does Arthur get hold of them, and what's in the app the gallery did not mean for the wider room to see?
The premise needs correcting. Our fair previews are crowdsourced — from collectors, advisors, and galleries. It's not a leak. Galleries contribute because a work ten clients see sells to ten clients; when thousands of collectors see the work, it sells faster. More eyes, more buyers.
What's in there isn't secret pricing. It's the thing collectors have never had: a complete look at what's coming before they walk in. Today you fly to Basel, discover an artist, and find the piece already sold to someone on the gallery's call list. If you're newer, or just not on that list, you've always been a step behind. We cover 500-plus fairs, focusing first on Basel and Frieze, so anyone can inquire early and plan a route instead of wandering the halls.
People worry transparency takes the magic out of a work. It doesn't. It takes the mystique out of the sales pitch — and those aren't the same thing.
The Jeff Magid feature, the podcast digests, the personalised weekly summaries. There's a clear editorial layer being built alongside the data. Is Arthur becoming a media company, and how much of the team's time is now going into editorial?
We're a data analytics company that uses editorial to make the data worth checking.
It's a progression. You start with data, connect the dots into analytics, and the analytics tell a story about where an artist or a movement is heading. Price performance, sell-through rate, market share by auction house, a category's price index — on their own those are charts. Together they tell you whether collector interest is heating up or cooling. The data is the raw material; the story is the product. The art world has never responded to spreadsheets — it responds to voice and taste. And a story is how someone who isn't a specialist gets to participate; you don't need twenty years in the art market to follow one.
Most of it is automated based on your own preferences. The digest writes itself from what you follow; the podcast is built from your collection, not a newsroom. That's the media company I want to build, and it doesn't contradict being a technology company.
The next move is crowdsourcing primary market data from collectors — often a dealer's best clients. Have any dealers pushed back, and what's your answer to a gallery that argues opacity is how they protect a young artist from speculation?
Some have, and the protecting-a-young-artist argument deserves to be taken seriously. A gallery placing an emerging artist’s work into collections that hold, not flip, is doing something real, and transparency can feed speculation that wrecks a young career. I won't wave that away.
But opacity doesn't stop the flippers — they're insiders, they already have the information. It mostly keeps out the serious collector who'd have held the work for twenty years. The wall built to keep speculators out keeps many serious collectors out instead.
And data cuts both ways: repeat-sale patterns let you spot a work being churned. Transparency is how you catch manipulation, not how you cause it — and done right, it brings an artist more serious collectors, not fewer. The threat to a young artist isn't us. It's the anonymous flipper.
1/ Auction DB Arthur Analytics app: Auction Database entry, Leonardo da Vinci, Salvator Mundi, sold Christie's, November 2017.
2/ Gallery Analytics Arthur Analytics app: Top Gallery Performance, Galerie Perrotin, total auction sales by represented artist, all time.
3/ Analytics / Auction Sales Arthur Analytics app: Auction Sales rankings by artist, 2025 and year-to-date.
4/ Artist Profile / Stats Arthur Analytics app: Adrian Ghenie artist profile, showing global rank, verified shows, and exhibition history by type.
The platform surfaces signals that are not sales — a museum hunting a loan, a gallery about to change an artist's representation, an underbidder active on a specific lot. None of that sits in the public record. Walk me through how it gets into the app.
Most of it comes from firms that already track it, which we license. Museum demand — which institutions want which works — comes through Museum Exchange. Representation changes ride on our exhibition data from Artfacts in Berlin, the deepest exhibition record there is. When an artist's show pattern shifts, the gallery switch usually appears there before it's announced anywhere. Underbidders are The Baer Faxt. We also map where works already live — the collections of 450-plus museums, and growing. Where a piece sits is its own signal: a work held by a major institution tells you how curators rate an artist, and that shapes collector opinion.
The signal always existed. It was scattered across newsletters and group chats, and only insiders knew where to look. We put it in one app and handed everyone the same map.
TRANSPARENCY IS HOW YOU CATCH MANIPULATION, NOT HOW YOU CAUSE IT — AND DONE RIGHT, IT BRINGS AN ARTIST MORE SERIOUS COLLECTORS, NOT FEWER. THE THREAT TO A YOUNG ARTIST ISN'T US. IT'S THE ANONYMOUS FLIPPER.
The roadmap: agentic search on the full dataset, the gamified auction features, the Exchange, the move to a paid tier. Which is closest to shipping, and which one keeps you up at night?
The Exchange is already live. Collectors anonymously list a work to buy, sell, borrow, or lend, and anyone following that artist gets a daily notification. Unlike Artsy, it's free — no commission.
Agentic search is in progress, and it's the clearest example of how we think about AI. Instead of clicking filters, you ask in plain language and get something more nuanced than any dropdown — which means a newer collector can get a real answer without knowing the jargon first. The model was never the hard part; anyone can wire one to a search box. What makes ours useful is what it searches: proprietary sales records, twenty years of exhibitions, primary-market sales submitted to us and nowhere else. So AI is a complement, not a threat. The two things an AI model can't conjure are our private data and our community — the Exchange, where collectors trade directly, and the messaging that lets them find each other. A better AI model just helps more people get more out of both.
We're also launching a fantasy auction game, on user feedback — a way to have fun and test your eye.
The paid tier is what keeps me up. Not because I doubt people will pay; collectors spending six figures a year on art will pay for insights and artist discovery. It's timing. Free is how Arthur spreads right now. Charge too early and you slow what's bringing people in; charge too late and you've trained everyone to expect it for free.
How big is Arthur's user base going into Basel, and what does this fair week make possible that earlier ones did not?
We're nearing 5,000 active users, but the number I'd watch is the rate — well over 400% a year, and accelerating. Almost all of it is word of mouth.
What Basel makes possible is density. The platform is only as good as the room contributing to it, and this is the week that room is in one place — collectors, advisors, galleries feeding the previews and the data. Each Basel we have a bigger share of it inside Arthur, so the previews are fuller and the signal is sharper.
And the room can find each other. You can search users by how they describe themselves — a collector of postwar Italian work, an advisor placing emerging photography — and message them directly, the kind of introduction you used to need a dealer to make. Earlier Basels we observed the fair. This one, a real slice of it is on Arthur.
Five years out. Where is Arthur, what does it look like, and what's the one thing it cannot do today that you most want to fix?
Five years out, Arthur is the first thing you open when something in the art world catches your eye — the artist a friend mentioned, the fair you're flying to, the work you're thinking of giving to a museum. Not just auction history, but the live state of the market: what's selling privately, what museums want, what's coming to every fair, who's bidding, where a career is heading. Global, paid, with the free layer still bringing people in, and still as good-looking as the art it covers.
But the real measure isn't the product. It's whether the market is bigger and more open because we exist — more people confident enough to buy their first serious work, more artists found earlier, fewer good decisions lost because someone wasn't in the right room. The art world has spent a long time keeping people out. I'd like Arthur to be part of why that changes.
The one thing I can't do yet is the one from your first question. Off-market and private sales are still mostly dark, and that's where the real money and the real signal live. The day Arthur shows the private market close to real time — the handshake deals that never hit a database — that's the whole game.
1/ Auction Price Charts / Elizabeth Peyton Arthur Analytics desktop: Elizabeth Peyton artist page, showing auction price over time and auction price by year created, Dec 1999–Feb 2024.
2/ Artist Analytics / Mark Grotjahn Arthur Analytics desktop: Mark Grotjahn artist page, showing sales distribution across price brackets and lots sold per year, 2017–2026.
IF YOU WERE AT ART BASEL TOMORROW, YOU WOULD HAVE ARTHUR. YOU WOULD ARRIVE KNOWING WHICH WORKS ARE COMING, WHICH GALLERIES ARE STAGGERING THEIR PREVIEWS, WHICH ARTISTS HAVE QUIETLY MOVED REPRESENTATION IN THE LAST QUARTER. THE ART MARKET HAS SPENT DECADES TREATING INFORMATION AS CURRENCY, CIRCULATING IT SELECTIVELY AND CALLING THE RESULT TASTE.
Coverage of Arthur so far has been very US. From Madrid, Paris, and Hong Kong the data gaps look different in each market. Where is the app strongest today, and what do Europe and Asia look like to you from inside the data?
Strongest in the US — auction depth, gallery coverage, the densest primary-market network, because that's where we started, and it's the largest market. Europe is stronger than people assume, on exhibitions: our data comes from Berlin, twenty-plus years deep, and Europe's institutional and biennial record is exceptional there. Where it thins is live primary-market signal, though that fills in wherever we get a cluster of users. Asia is the frontier. Auction is well covered — Hong Kong feeds the same data. But the primary market and gallery layer are thinnest, because the trade there runs on relationships and WeChat in ways the public record never touches. Asia is where the most is happening and the least is written down. That's the gap, and the opportunity.
About Arthur Analytics
Arthur Analytics is a data platform built for collectors, advisors, and art professionals. Founded in 2023 by James Crichton, it integrates auction records, primary market activity, museum demand, gallery representation changes, fair previews across 500-plus fairs, and underbidder data into a single app, making visible the signals that have always existed but rarely reached beyond the inner circle. Its data infrastructure draws on exclusive licensing agreements with partners including The Baer Faxt, Museum Exchange, and Artfacts in Berlin, aggregating a depth and breadth of information unavailable elsewhere in one place.
Arthur tracks the collections of 450-plus museums, maps gallery exhibition patterns to detect representation shifts before they are announced, and generates personalised weekly digests and AI-driven podcast summaries built from each user's own collection and interests. Arthur Exchange, live now, lets collectors anonymously list works to buy, sell, loan, or rent at no commission. Available on iOS and at arthuranalytics.com, the platform is currently free, with a paid tier in development.
_edited.jpg)
