Dendrometers have been widely studied for irrigation management in apple orchards, yet the high variability of dendrometer-derived parameters among trees remains a challenge for irrigation control. This study evaluated the variability of key dendrometer-derived parameters across eight trees and four stem positions per tree, with particular focus on their applicability for irrigation optimisation. Continuous measurement of trunk water potential (Ψtrunk) obtained with microtensiometers and conventional stem water potential measurements (Ψstem) obtained using a Scholander pressure chamber were compared with dendrometer-derived growth and tree water deficit (TWD) parameters. The number of sensors required to achieve a 0.15 MPa (Ψtrunk) standard error was then calculated. All measured variables showed higher variability than Ψstem. Dendrometer parameters showed greater variability within trees than between trees, indicating substantial heterogeneity across stem positions. When using TWDmax as Ѱtrunk predictor, nine sensors were required to achieve 0.15 MPa standard error (coefficient of variation = 0.123). Five dendrometers were needed to accommodate a larger variance of 0.2 MPa. Daily Ψtrunk minimum and TWDmax were well correlated (R2m and R²c = 0.73), although this correlation was weaker and tree dependent for Ψtrunk maximum and TWDmin (R2m = 0.06, R²c = 0.32). This reflects their shared measurement location conditions and supports the suitability of dendrometers for monitoring tree water status, especially in the middle of the day. Overall under the here presented experiment conditions, dendrometers represent a practical, economical, and reliable tool for operational water stress assessment, and options to reduce observed variability should be studied.