Silicon is being increasingly studied as the next-generation anode material for Li-ion batteries because of its ten times higher gravimetric capacity compared with the widely-used graphite. While nanoparticles and other nanostructured silicon materials often exhibit good cyclability, their volumetric capacity tends to be worse or similar than that of graphite. Furthermore, these materials are commonly complicated and expensive to produce. An effortless way to produce nanostructured silicon is electrochemical anodization. However, there is no systematic study how various material properties affect its performance in LIBs. In the present study, the effects of particle size, surface passivation and boron doping degree were evaluated for the mesoporous silicon with relatively low porosity of 50%. This porosity value was estimated to be the lowest value for the silicon material that still can accommodate the substantial volume change during the charge/discharge cycling. The optimal particle size was between 10–20 µm, the carbide layer enhanced the rate capability by improving the lithiation kinetics, and higher levels of boron doping were beneficial for obtaining higher specific capacity at lower rates. Comparison of pristine and cycled electrodes revealed the loss of electrical contact and electrolyte decay to be the major contributors to the capacity decay.