Sorting is not mainly about order. It is about putting similar things next to each other so the eye can catch what a filter would miss.
What surfaces when you sort
Near-duplicates that differ by a character. Inconsistent capitalisation. Trailing spaces. Two spellings of the same brand. None of these are findable by scanning an unsorted list, and all of them are obvious once neighbours are adjacent.
Sorting both directions shows different ends
Ascending puts one end of the range in view, descending the other. Anomalies frequently live at the extremes, and checking both takes seconds.
Do it before every deduplication
A quick sorted read tells you what the deduplication is about to miss — the variants that differ slightly and will therefore both survive. That is the moment to normalise, not after the count has already been reported.
Try it: Sort Z to A on SeoWolf's Notepad.