Due to late time non-linearities, the location of the acoustic peak in the two-point galaxy correlation function is a redshift-dependent quantity, thus it cannot be simply employed as a cosmological standard ruler. This has motivated the recent proposal of a novel ruler, also located in the Baryon Acoustic Oscillation range of scales of the correlation function, dubbed the "linear point". Unlike the peak, it is insensitive at the $0.5\%$ level to many of the non-linear effects that distort the clustering correlation function and shift the peak. However, this is not enough to make the linear point a useful standard ruler. In addition, we require a model-independent method to estimate its value from real data, avoiding the need to deploy a poorly known non-linear model of the correlation function. In this manuscript, we precisely validate a procedure for model-independent estimation of the linear point. We also identify the optimal set-up to estimate the linear point from the correlation function using galaxy catalogs. The methodology developed here is of general validity, and can be applied to any galaxy correlation-function data. As a working example, we apply this procedure to the LOWZ and CMASS galaxy samples of the Twelfth Data Release (DR12) of the Baryon Oscillation Spectroscopic Survey (BOSS), for which the estimates of cosmic distances using the linear point have been presented in Anselmi et al. (2017) [1].