To rationally design new molecular materials with desirable linear optical properties, such as refractive index or birefringence, we investigated how atomic and functional-group polarizability tensors of prototypical molecules respond to crystal field effects. By building finite aggregates of urea, succinic acid, p-nitroaniline, 4-mercaptopyridine, or methylbenzoate, and by partitioning the cluster electronic density using quantum theory of atoms in molecules, we could extract atoms and functional groups from the aggregates and estimate their polarizability enhancements with respect to values calculated for molecules in isolation. The isotropic polarizability and its anisotropy for the molecular building blocks are used to understand the functional-group sources of optical properties in these model systems, which could help the synthetic chemist to fabricate efficient materials. This analysis is complemented by benchmarking density functionals for atomic distributed polarizabilities in gas phase, by comparing the results with refractive-index calculations under periodic boundary conditions, and by estimating functional-group optical properties from a classical electrostatic atom-dipole interaction model.