Time domain reflectometry (TDR) is a simple and popular method for measuring soil water content. The TDR measures a weighted average of the dielectric permittivity of the soil along and around a probe. Under some experimental or technical restrictions, the penetration of the probe into the studied domain is undesirable or impossible. Here, we experimentally investigate the feasibility of measuring a material’s water content by only touching it. Such a configuration, where the probe is not fully inserted into the measured domain, leads to a far-from-uniform distribution of the dielectric permittivity around the probe. A laboratory experiment was conducted, comparing wall-embedded TDR (wall probe [WP]) measurements with those of a regular TDR probe (standard probe [SP]). Five different soils were examined (sand, “Hamra” [red loamy sand], loess, pure quartz, and clayey soil). For all soils, the WP measurements apparently underestimated water content. This is because it averages soil permittivity of the domain of interest (soil) with that of the probe host material, typically of lower dielectric permittivity. A linear relation between the water content as measured by the WP and the SP was found for all the soils. For most soils and for the non-marginal water contents, the true soil permittivity (and hence the water content) was roughly twice that measured by the WP. However, the measurement was found to be highly sensitive to the exact location of the probe at the material interface. We conclude that the wall-assimilated probe can be used with minimal calibration to measure water content without penetration into the measured domain, but high caution should be used with respect to the exact location of the probe at the wall–soil interface.