The generation of renewable and sustainable electricity can be achieved through fuel cells by the reaction hydrogen (H2) with oxygen (O2). In order to meet the current energy demand with fuel cells, a large feedstock of H2 is necessary. Alkaline water electrolysis is the only mature energy conversion technology for the renewable production of H2 through the hydrogen evolution reaction (HER). The efficiency of water-splitting is limited by its other half reaction, the oxygen evolution reaction (OER), due to the multiple proton-coupled electron-transfer (PCET) steps involved in it. Currently, ruthenium- and iridium-based catalysts and elemental platinum still represent the benchmark catalysts for practical applications of OER and HER, respectively. Nonetheless, their high cost and scarce availability limit their use in large-scale applications. Transition metal pnictides have become an attractive choice as OER and HER electrocatalysts due to their low hydrogen adsorption energies, faster reactions kinetics, and electroconductivity. Common synthesis of transition metal pnictides are based in solid-state methods, which lead to materials with uncontrolled aggregation, heterogeneous morphology, and random composition. The design of nanostructured electrocatalysts of specific size, stoichiometry, and morphology can be achieved by the molecular decomposition of single-source precursors (SSP). The tuning of the experimental conditions (i.e., decomposition method, reaction time, temperature, solvent) leads to nanostructures of different characteristics. In this work, a molecular synthetic approach has been used to access transition metal phosphides (TM-Ps) and arsenides (TM-As). Deeper investigations have been carried out to understand the kinetics and electronic processes during catalysis and relate them to the chemical and electronic transformations of the materials. Amorphous materials are known for their large electrochemically active surface (ECSA), which usually allows higher electrocatalytic activities in comparison to their crystalline counterparts. Keeping this in mind, amorphous and crystalline cobalt phosphide (CoP) have been synthesized by the thermolysis (hot-injection and pyrolysis) of a SSP. Amorphous CoP notably displayed higher activity for all tested electrocatalytic reactions in comparison to the crystalline CoP. During OER, the electrochemical corrosion and oxidation of phosphorus led to the formation on both materials of a CoIII (oxy)hydroxide shell. The presence of CoIV OER active sites in the surface, in combination with CoIII led to distorted CoO6 octahedra (Jahn-Teller distortion) that facilitated the absorption of oxygen intermediates. During HER, a complete loss of surface phosphorus and oxidation of the surface led to the formation of a Co-enriched (oxy)hydroxide phase, followed by an in situ reduction to generate Co0 active sites for HER. The difference in activity was related to unique electronic properties and surface characteristics of the high ECSA of amorphous materials, that offers a larger number of active centers and larger flexibility upon catalysis. TM-As have received little attention in water-splitting applications due to the high toxicity of As. However, their high electrical conductivity render these materials active for OER electrocatalysis. Crystalline FeAs nanoparticles were accessed through the hot-injection of a SSP. The electrochemical exploration showed high OER catalytic activity and stability in comparison to the other Fe-based reference electrodes and Fe-based electrocatalysts reported in the literature. During OER, the material underwent corrosion, which caused the complete oxidation and dissolution of As into the electrolyte. Ex-situ characterizations, as well as quasi in situ XAS, revealed the total electroconversion of the FeAs to a highly porous nanocrystalline iron (oxy)hydroxide phase, the 2-line ferrihydrite. The semiconductor nature of this phase enabled faster electron transport. Moreover, the generated phase contained Jahn-Teller distorted FeIII atoms that behaved as active centers for OER. Importantly, the dissolved As was successfully recaptured at the cathode making the complete process sustainable and energy-efficient. The influence of the non-metal (pnictogen) in the activity was studied by preparing transition metal pnictides with equal metal content, morphology, and composition. A room temperature salt-metathesis was used to access amorphous NiAs and NiP. Electrochemical experiments revealed a higher OER activity of NiP in comparison to NiAs and other Ni-based materials reported in the literature. The post-operando analysis revealed oxidation and dissolution of the pnictogen during OER and generation of a γ-NiIIIOOH phase. This phase contained sheets of edge-sharing NiO6 octahedra separated by a large interlayer distance (7 Å), which provided the space for the intercalation of water and anions which were exposed to the NiIII active centers. The difference in activity was attributed to the higher loss of the pnictogen element in NiP (86 %) in comparison to the NiAs (45 %), which resulted in a larger structural transformation of the NiP to a γ-NiIIIOOH phase. This work illuminates that easy preparation routes towards nanoscaled TM-As and TM-Ps of different morphology and composition which could ensure the physical and electronic properties necessary for OER, HER and OWS. The studied materials showed electrocatalytic activities under alkaline conditions and comparable to the ones of precious-metal-based catalysts. Moreover, the developed methods are flexible and shows a wide range of attainable products depending on the preparation route. Therefore, it opens the route to the facile preparation of transition metal pnictides and their further application in other areas.