Superresolution techniques in imaging are widely known for exploiting various special properties of matter or illumination source to uncover fascinating proprities of imaged objects, important across disciplines such as biology or material sciences. Recent developements in quantum information theory pointed out that even in a simple fluorescence imaging setup, or equivalently even in a astronomical telescope, resolution may be improved by designing an optimal measurements of incoming photons. These theoretical ideas led to practical questions of implementation, which we have been answering in our lab. In particuar, we explore the specific situations where the informational gain may be achieved. In a most recent expeirment, we translated the ideas from imaging to spectroscopy, and used our quantum memory setup to implement an optimized measurement. We have shown that the same spectral resolution (precision) may be obtained using our setup with 20 times less light than with a conventional device.