Water is one of the most important erosive agents in roadside hillslopes. When these are built with ineffective drainage systems, erosion occurs, reducing road’s service life. However, these systems are not receiving the appropriate importance, given their strategic value. Therefore, a new drainage system called ‘branched’ is proposed in this study. Its technical and economic feasibility is compared with those of the traditional system, which consists of drainages with lines that follow maximum hillslope, to assess differences in relation to erosion, construction and maintenance costs, and service life. Different parameters were analysed, such as the average velocity of water (mm⁻¹) running through the channels, its average specific energy (kJ), and its drag force (N). A scale model was constructed and used to test these factors before implementing it in natural terrain for testing it under field conditions. According to the theoretical and measured results, these factors were lower in the branched drainage than in the traditional one (from 24% to 34% in speed, from 37% to 60% in energy, and from 51% to 73% in force). The service life of hillslopes with a branched system of up to 0.5 m high and 1:2 grade is significantly longer than in those with a traditional drainage. Although the initial economic expense for the construction of the branched system is higher (€3534/m³ as opposed to €2930/m³ for the traditional one), its maintenance cost will be lower than the traditional one (€1230/m³ per year for the branched one as opposed to €1332/m³ per year for the traditional one). Consequently, under our experimental conditions, the proposed drainage will be profitable from the eighth year of construction, saving on the road maintenance in the following 15 years of service life.