The High Energy Modular Ensemble of Satellites (HERMES) project intends to build an all-sky monitor operating from keV to MeV for the detection and localisation of transient events, like gamma ray bursts. HERMES is a modular observatory composed by detectors on-board of nanosatellites. HERMES aims to revolutionise the world of multi-messenger astrophysics thanks to the innovative concept of a modular instrument based on small satellites and characterised by reduced design and development times and low costs, in the face of high technological content and scientific profile of the mission.A first part of the project, HERMES Technological Pathfinder, composed of three CubeSats, aims to demonstrate the feasibility of detecting transient phenomena in high energy with small satellites. The next phase of the project, HERMES Scientific Pathfinder, will expand the constellation up to six satellites, allowing routinely accurate triangulation measurements.This paper presents an analysis that aims to be propaedeutic for the design of the final HERMES Full Constellation, which will be an all-sky monitor made up of tens/hundreds of nanosatellites in Low Earth Orbit with a total effective area of ∼m2. An adequate number of nanosatellites, simultaneously detecting a transient, provides a source localisation accuracy of the order of magnitude of few arcmin and large effective area. The objective of this work is to define possible strategies of injection in orbit of the complete constellation. In this first analysis we used only the number of triangulable Gamma Ray Bursts (simultaneously detected by at least three nanosatellites) as discriminating factor in order to maximise the constellation performance. By achieving the goal of at least a mean number of 20 Gamma Ray Bursts triangulated per nanosatellite over the two years life-time, optimal configurations for the complete constellation, consisting of seven to fourteen CubeSats, have been identified.