Fluorescence microscopy is a powerful imaging tool that can be employed to track, localize and monitor fluorescently labelled biomolecules within cellular systems. However, using this technique only fluorescently labelled biomolecules can be studied within a given sample, while pertinent information on their immediate cellular environment remains ‘in the dark’. Electron microscopy enables in depth observations at a nanometre‐scale resolution of precisely such environmental details, including the cytoskeleton, membranes and even macromolecular complexes. Therefore, electron microscopy is often employed after fluorescence microscopy to ultrastructurally characterize the cellular environment of fluorescently labelled biomolecules of interest. In order to translate the information derived from fluorescence‐based studies for electron microscopy analysis, fluorescently labelled biomolecules must be identified in a manner compatible with the latter. Traditionally, this is achieved using electron dense gold markers equipped with antibodies recognizing fluorescently labelled biomolecules. Alternatively, the area of interest is imaged with both the fluorescence and electron microscope, followed by superimposition of the resulting data sets. In order to compare and contrast the outcomes of these two strategies, we compared the subcellular localization of fluorescent labels with that of gold labels. Surprisingly, we found that the distribution of the gold label does not always correspond with the distribution of the fluorescent label. The present study experimentally interrogates and characterizes this phenomenon with broad implications for combining fluorescence and electron microscopy techniques.