I had a professor who suggested measuring out mobile phase components by weight rather than volume for greater consistency between runs. However, at work I typically prepare 2-5 L or more, so it's not very practical. Using volumetric glassware is also usually not practical because you end up using more than one flask. For example, how would you measure out 1750 mL? You would have to use at least three volumetric flasks and besides being inconvenient I doubt the accuracy would be that much better than a graduated cylinder.