In a charge-transfer (CT) transition, electron density moves from one end of the molecule (donor) to the other end (acceptor). This type of transition is of paramount importance in nature, for example, in photosynthesis, and it governs the excitation of several protein biochromophores and luminophores such as the oxyluciferin anion that accounts for light emission from fireflies. Both transition energy and oscillator strength are linked to the coupling between the donor and acceptor groups: The weaker the coupling, the smaller the excitation energy. But a weak coupling necessarily also causes a low oscillator strength possibly preventing direct excitation (basically zero probability in the noncoupling case). The coupling is determined by the actual spacer between the two groups, and whether the spacer acts as an insulator or a conductor. However, it can be difficult or even impossible to distinguish the effect of the spacer from that of local solvent molecules that often cause large solvent shifts due to different ground-state and excited-state stabilization. This calls for gas-phase spectroscopy experiments where absorption by the isolated molecule is identified to unequivocally establish the intrinsic molecular properties with no perturbations from a microenvironment. From such insight, the effect of a protein microenvironment on the CT excited state can be deduced. In this Account, we review our results over the last 5 years from mass spectroscopy experiments using specially designed apparatus on several charged donor-acceptor ions that are based on the nitrophenolate moiety and π-extended derivatives, which are textbook examples of donor-acceptor chromophores. The phenolate oxygen is the donor, and the nitro group is the acceptor. The choice of this system is also based on the fact that phenolate is a common structural motif of biochromophores and luminophores, for example, it is a constituent of the oxyluciferin anion. A presentation of the setups used for gas-phase ion spectroscopy in Aarhus is given, and we address issues of whether double bonds or triple bonds best convey electronic coupling between the phenolate oxygen and the nitro group, the significance of separating the donor and acceptor spatially, the influence of cross-conjugation versus linear conjugation, and along this line ortho versus meta versus para configuration, and not least the effect of a single solvent molecule (water, methanol, or acetonitrile). From systematic studies, a clear picture has emerged that has been supported by high-level calculations of electronically excited states. Our work shows that CC2 coupled-cluster calculations of vertical excitation energies are within 0.2 eV of experimental band maxima, and importantly, that the theoretical method is excellent in predicting the relative order of excitation energies of a series of nitrophenolates. Finally, we discuss future challenges such as following the change in absorption as a function of the number of solvent molecules and when gradually approaching the bulk limit.