In this thesis, we will investigate the collider phenomenology and cosmological consequences of extensions of the Standard Model (SM) with hidden sectors coupled to the SM via a Higgs portal coupling. We will explore how models with classical scale invariance, where all mass scales are dynamically generated, can address the shortcomings of the SM without destabilising the Higgs mass. The matter-antimatter asymmetry in the Universe and the tiny masses of active neutrinos are addressed in a U(1)$_{B-L}$ extension of the SM with GeV scale right-handed neutrinos. We then investigate a range of models with both Abelian and non-Abelian gauge groups in the hidden sector to show how we can stabilise the Higgs potential and at the same time provide phenomenologically viable dark matter candidates where all scales in the theory have a common origin. For non-Abelian gauge groups in the hidden sector, we also show that hidden magnetic monopoles can make up a significant fraction of dark matter. The dark matter in this model, which consists of both magnetic monopoles and gauge bosons, has long-range self-interactions which could explain the too-big-to-fail-problem at small scales in the standard cold dark matter scenario. We then study the collider phenomenology of hidden sector models with dark matter candidates through a simplified model framework both at the LHC and at a future 100 TeV collider. Hidden sector extensions of the SM with a Higgs portal coupling give a rich and predictive model building framework for BSM physics without introducing a large hierarchy of scales.