Inflation is presently considered to be the most compelling paradigm to describe the origin of perturbations in the early universe. The remarkable success of the inflationary paradigm can be attributed to its simplicity and efficiency. However, the efficacy of the inflationary scenario also seems to be responsible for a major drawback. Despite the strong constraints from Planck on primordial non-Gaussianities, a rather large set of inflationary models seems to remain consistent with the cosmological data, even as new models continue to be proposed. In such a situation, it seems imperative to explore viable alternatives to inflation. In this talk, after an introduction to inflation and the inflationary observables, I briefly discuss the performance of a handful of inflationary models against the cosmological data. Thereafter, I shall discuss as to how bouncing scenarios offer an alternative to inflation to overcome the horizon problem and also lead to scale invariant spectra. Specifically, I shall describe our recent work on a matter bounce scenario that leads to a viable tensor-to-scalar ratio and small tensor non-Gaussianities. I shall conclude by highlighting the typical theoretical challenges faced in constructing well motivated bouncing models that are also consistent with the observations.