From catalogs it seems that they are superior to C8 and C18 in the analysis of small molecules like nucleosides, narcotics (morphine, codeine, methadone) anesthetics (lidocaine, prilocaline, benzocaine, other drugs (oxaxepam, alprazolam, diazepam), fluoroquinolones, tetracyclines, benzodiazepines, steroids, etc. i.e. smal molecules wiht one or more rings.

I've had some success with "difficult" antibiotics (Massimo mentioned the tetracyclines) and vitamins on phenyl columns. Even if your compound isn't "phenolic" in the broader sense, it can help your separation by offering a different selectivity (i.e. shifting your peaks around...)

The major compound that can be separated using phenyls phase (Biphenyl) is aromatic compounds with electron withdrawing groups like benzene /nitrobenzene, phenyl phases like  O-terphenyl /n-butyl benzene, steroids hormones etc.