What is your procedure for using comparison stars for photometry or spectroscopy?
How many do you use? Do you combine multiple stars? Once you have obtained your observations, do you find that you are able to use all of your chosen stars for analysis? What about testing for stellar variability?
In our case, we are performing timeseries spectroscopy and we notice that the lack of atmospheric dispersion correction causes the wavelength solution of all stars to vary over time. As a result, we found that using a single comparison star of similar spectral type to the target improved precision compared to using multiple stars of different spectral type (because spectral shifts cause similar flux variations in 2 stars with similar spectra, while the same shift could cause very different flux variations for stars with different spectra).