The Gathas are among the hardest texts in any ancient language, and any reader using a translation should know why.
The corpus is small, about six thousand words, which means there is very little context for establishing the meaning of rare terms. The transmission was oral for a very long period before it was written down, and the priests who preserved it were reciting a language they no longer spoke. The surviving manuscripts are late, the earliest of substance from the thirteenth century CE. The Middle Persian translation and commentary, the Zand, was made when the original was already imperfectly understood, and it frequently glosses rather than translates.
The consequence is that translations of the same stanza can differ so completely that a reader comparing two will not recognise them as the same text. This is not a matter of style. Two competing methods have been used: one working from the Zoroastrian exegetical tradition, associated with the Bombay school and with Kanga and others, and one working from Vedic comparison, associated with European philology and with figures from Bartholomae to Humbach and Insler. They produce different texts.
M. L. West's verse translation is accessible and is explicitly a poet's rendering with philological notes.
The practical advice for a reader is to use at least two translations, to prefer editions that mark uncertainty, and to be suspicious of any version that reads smoothly throughout. The Gathas do not read smoothly in Avestan.