Every price on this site comes from sales we can name, run through one published model, and reviewed under a fixed standard before release. The full process is public, including where the data is still thin.
21,516
published price ranges, built from 63,835 recorded sales
Four steps, the same for every price on the site. Why you can lean on these numbers.
63,835
recorded sales, each with its source
Each figure traces to a recorded auction result or gallery price, with its citation attached.
One published model weighs
The model, its inputs, and every source are published in full, open to review by anyone.
Nothing publishes until sources, model inputs, and artist identity have been checked against Okasen's published review standard.
21,516
published price ranges
A verified price is a range with a confidence interval. A wide range signals thinner data, not a hidden risk.
See where an artist’s market sits before you bid or sell. A tight range and a wide one should change how much weight you put on the figure.
A reference you can put in front of a client, with citations they can follow. Useful for internal valuations as well.
A record of the African art market over time that you can cite, with the method attached to every figure.
3,543 artists
In: auction sales, read for
Out: a price
A range with a confidence interval, plus a liquidity score
7,542 artists
In: signs of a market, read for
Out: a score, not a price
A momentum score, because the data cannot carry a price
Low end$22
The amber bandWhere a sale would plausibly land
High end$1,596
Not a floor, and not a promise: a band, set on a log scale, so each step on the axis is ten times the last.
The same artist's work can sell for very different sums from one lot to the next. The range is the band we think a sale would plausibly land in. It is not a floor, and it is not a promise.
One number would look more precise than the data really is. So every estimate shows its range and its tier. There is no bare figure anywhere on this site, on purpose.
Plenty of recent, consistent sales make it narrow. Thin or scattered data makes it wide. A wide interval is us being honest, not us being sloppy.
Subscribe once to unlock the methodology whitepaper
Subscribe once. You get the methodology paper the day it lands, plus every other download on the site right now. One form, and we will not ask again.