What is wrong with percentages

Apart from the remarks already done in the mass-volume percentage and volume-volume percentage sections there are at least two more reasons to use percent concentrations with caution.

Percent concentrations are regularly abused in several ways. Solution labelled 5% could be prepared by mixing 5 g of substance with 95 g of solvent - preferred, but almost never used method, by taking 5 g of substance and filling it up to 100 mL (in which case it is % w/v) or by taking 5g of substance and adding 100 mL of solvent. To make things worse preparation of % w/v solutions can be done either by taking 5 g of substance and filling it up to 100 mL, or by taking 5 g of substance, dissolving it in part of solvent and filling it up to 100 mL then. 50% v/v solution is usually made by mixing equal volumes of both liquids. In every case final result is different, in every case concentration is - in fact - unknown.

Usually that's not an issue as the difference between real and assumed concentration is below the level of accuracy of solution preparation, but using such solutions you should be aware of the fact that their composition may be far from what you are expecting.

Concentrations expressed as mass-volume and volume-volume percentages are also temperature dependent, which lowers their already doubtful precision.