In this article (6 sections)
Reading time: 6 minutes | Updated July 2026
In 1504, a Florentine merchant commissioned a portrait from a relatively unknown painter. The contract specified the dimensions, the materials to be used, and the fee. The painter was paid. The merchant got his portrait. And somewhere in that transaction, the question of what a painting was actually worth - separate from what it cost to make - went entirely unasked.
The painting was the Mona Lisa. The painter was Leonardo da Vinci. And the fee Leonardo received would not have bought a modest apartment in Florence today.
This is the oldest problem in art valuation. It has never been fully solved. But six centuries of trying have taken us from a system where wealthy patrons set prices by personal whim, through the invention of the auction room, through provenance research, through formulas and indices - to a moment, right now, where artificial intelligence is beginning to read paintings the way a trained eye does.
If you have ever wondered how the number on your painting’s price tag should be determined, the answer starts here.
When price meant materials, not meaning
For most of art history, a painting’s value was calculated the way a carpenter’s cabinet was calculated. You added up the cost of the wood, multiplied by the hours worked, and arrived at a price. In art, the equivalent was lapis lazuli, gold leaf, and the painter’s daily rate.
During the Renaissance, contracts between artists and patrons stipulated the final cost, the timescale, and the quantity of precious materials to be used - sometimes even including an illustration of the intended work. Prices for paintings remained low in the 15th century. It was not until the 16th century that notions of artists being more elevated than artisans began to gain currency.
This matters for anyone pricing their work today, because the core tension it reveals has never gone away: the cost of making a painting and the value of a painting are two entirely different things. In the medieval period, artists were effectively craftsmen. Gradually, people began to understand artists were not just painters and decorators but had a particularly gifted insight onto the world - and with that shift came the first stirring of what we now call market value.
The Medici didn’t commission Botticelli’s Primavera because they needed wall decoration. They commissioned it to signal power, cultural authority, and proximity to the divine. The painting’s value was political as much as aesthetic - and none of that showed up in the materials invoice.
The auction room invents the market
For centuries, the price of a painting was a private negotiation between an artist and a patron, or between two collectors. There was no public record, no benchmark, no way for an artist to know whether they were being paid fairly or exploited.
It was not until the 17th century that art auctions emerged, as the rise of a collector base within the mercantile middle class of Flanders and Holland engendered a flourishing open market. Newspapers began reporting on the auctioning of artworks in the coffeehouses and taverns of London and Paris. Sotheby’s was established in London in 1744, followed by Christie’s in 1766 - and between them, they built something that had never existed before: a public, transparent mechanism for determining what a painting was worth in the eyes of willing buyers.
This was genuinely revolutionary. For the first time, an artist could look at auction records and understand the market they were entering. A collector could compare what they paid against what others had paid for similar work. Dealers could build businesses on the spread between private and public valuations.
But there was a structural problem with auction-based valuation that persists today: auction houses became instrumental in establishing benchmarks for art prices, but this system heavily favoured well-known artists and often excluded emerging talents. The auction room is brilliant at valuing work by artists with a sales history. It is largely blind to work by artists without one.
If you are an emerging painter, the auction house system was not built for you. It still isn’t.
Provenance: when history becomes value
By the twentieth century, art valuation had developed a third dimension beyond materials and market: the history of the object itself.
Provenance - who owned a painting, where it was exhibited, whether it passed through significant collections - began to be understood as a value driver in its own right. A painting that had hung in a notable private collection commanded a premium over an identical painting with an undocumented past. An exhibition history with a respected gallery added credibility that no formula could replicate.
This remains true today and it matters directly for emerging artists. The paintings you submit to open exhibitions, the galleries that accept your work even for group shows, the collectors who buy early - these are not just biographical details. They are value anchors that will affect what your work is worth for the rest of your career.
Document everything from the beginning. Keep sale records. Photograph installations. Retain correspondence with galleries. Every credible touch point in your painting’s history is part of its future valuation.
The formula era: pricing by the centimetre
By the late twentieth century, as the contemporary art market opened up and thousands of artists began selling work without gallery representation, a more accessible pricing method emerged: formula-based valuation.
The square-inch and square-centimetre methods allowed artists to price consistently and transparently. Multiply the surface area by a rate that reflects your career stage, add material costs and a reasonable hourly rate, and arrive at a defensible number. Not a market number - a cost floor. But a number you could explain to a buyer without embarrassment.
These formulas are still widely used, and they remain a useful starting point. They solve the confidence problem that most emerging artists face: not knowing where to begin. They create consistency across an artist’s body of work. And they give buyers a logical framework that feels fairer than a number pulled from the air.
Their limitation is the same as it was in the Renaissance: they measure what a painting cost to make, not what someone will pay for it.
The formula is the floor, not the ceiling. Everything above it is market.
AI enters the room - and finds that reputation still wins
The most recent shift in art valuation is the one currently underway, and it is the most significant since the invention of the auction house.
Machine learning models are now being trained on decades of auction data to predict painting prices. Sotheby’s AI-driven estimate engine, launched in 2022, boasted an 83% accuracy rate within predicted ranges across 112 lots. Artprice forecasted a price band of $45M–$52M for a Gerhard Richter; it sold for $48.7M. At the high end of the market - established artists with extensive sales histories - AI valuation is becoming remarkably precise.
But when applied to emerging artists, it reveals something important. A Large Vision Model led by art economist Magnus Resch was trained on millions of images, including price histories. In over half of cases, the model predicted auction prices with surprising accuracy based solely on visual input. But accuracy collapsed without context. When stripped of metadata - artist name, provenance, gallery representation - the system faltered. It could analyse colour balance, composition, and texture, but it could not grasp reputation.
This finding is counterintuitive and should be reassuring in equal measure. The AI confirmed what dealers have always known: at the top of the market, more than 50% of contemporary auction value concentrates around a small circle of already-validated artists. Visibility begets visibility. But the same research points toward where the real opportunity lies.
Multi-modal deep learning delivers significant value precisely when valuation is hardest - namely, first-time sales - where historical anchors are absent. Visual embeddings provide a distinct and economically meaningful contribution for fresh-to-market works. In other words: AI is most valuable exactly where emerging artists need it most - when there is no sales history to fall back on, and the painting itself must carry the argument.
This is what platforms like Priceyourpainting are built to do. Not to replicate the auction house model - which was never designed for emerging artists - but to bring data-backed, market-informed valuation to the part of the market that has been navigating blind for six hundred years.
Image: Digital visualisation of AI analysis mapping compositional and stylistic cues in a painting. Illustration for reference.
Where valuation is going next
The convergence of visual AI, auction data, and artist career analytics is producing valuation tools that would have been unimaginable a decade ago. Platforms like Wondeur AI now track over 800,000 artists and predicted with 82% accuracy which ones would feature in major exhibitions. Behavioural analytics measure collector intent. Market indices track price movements by medium, style, and region in near real time.
For the emerging artist, the practical implication is this: the information gap that has historically allowed galleries and dealers to capture the majority of value in any transaction is closing. An artist who understands what their work is worth - backed by data rather than guesswork - negotiates from a fundamentally different position.
The Mona Lisa’s painter received a flat fee and no share of the painting’s future value. Five centuries later, the system that made that possible is finally changing.
We built Priceyourpainting to give every artist access to the data that was previously reserved for gallerists, dealers, and auction specialists. Your painting has a market value. You deserve to know what it is.
References
Patrons & Artists in Renaissance Italy - World History Encyclopedia, Mark Cartwright, September 2020. worldhistory.org
The Renaissance art market - Encyclopaedia Britannica Money. britannica.com
A Short History of Art Auctions - Artland Magazine, July 2022. magazine.artland.com
A History of Sotheby’s Auction House - MyArtBroker, January 2026. myartbroker.com
When AI Prices Art: Why Picasso Loses to the Unknown - Artmag, May 2026. artmag.org
Predictive Analytics in the Art Market - Dipayan Ghosh, May 2025. dipayang.com
Deep Learning for Art Market Valuation - Yang et al., arXiv, December 2025. arxiv.org
Made with love by artists, for artists 🫶🏼
Stay connected with us on Instagram for more artist stories and updates!
Comment on Substack
