The most common way a dose log dies is not neglect, it is ambition. Someone designs a fourteen column tracker, fills it perfectly for eleven days, and then stops because each entry has become a small chore.
The counter argument is that a log with five columns kept for a year beats a log with fourteen columns kept for a fortnight, by a very large margin. Almost every question you will actually ask of the record can be answered from date, dose in mg, units, vial, and one note field.
So the design question is not what would be nice to have. It is what you would still fill in on a bad week. Anything that fails that test should be optional, meaning it can be left blank without the row feeling incomplete.
Optional fields are fine as long as they are genuinely optional. What kills a log is a required field you cannot always supply, because then the honest response is to skip the whole row, and one skipped row makes the next one easier to skip.
Two additions that earn their place for most people: a one to five number for how the week felt, and weight if you track it anyway. Both are quick, both are comparable over time, and neither requires you to compose a sentence.
Everything else, symptoms in detail, meals, sleep, activity, is a research project rather than a dose log. Worth doing for a defined stretch when you are trying to answer a specific question, not worth carrying permanently.
I keep the permanent record small in the Peptide Tracker & Calculator and add detail only when there is a question to answer. The log survives that way, and a surviving log is the only kind that is any use.