<p>Recent years have seen a marked rise in fingerprint studies on ceramic artefacts, offering novel insights into the social organisation of production and the division of labour at a growing number of sites worldwide. However, methods adapted from forensic biometry and dermatoglyphics must be correctly calibrated for archaeological datasets, especially given the distorting effects of clay shrinkage on ridge breadth measurements. While clay shrinkage has long been recognised as a confounding variable, targeted experimental studies are increasingly needed to anchor biometric classifications of ‘plastic print’ impressions. This study presents the first shrinkage-controlled dataset of its kind, based on 420 experimental briquettes made from regionally relevant clays and temper types fired at successively higher temperatures. Shrinkage corrections are applied using a simple but novel ‘proportional-change formula’. The results of this study are used to calibrate a set of fingerprint impressions preserved on South Levantine Metallic Ware vessels recovered from an Early Bronze Age II complex at the Qedesh megasite in the Galilean highlands. The necessity of shrinkage-controlled reference groups for accurate biometric classification is demonstrated. The results underline the need for calibrations anchored in the physical properties of local raw materials, rather than generic predetermined rates, or statistical adjustments divorced from measured clay behaviour. By modelling the shrinkage limits of different clay bodies, this study also reconsiders fingerprint evidence from other major Early Bronze Age sites in the Levant.</p>