The question is one of optimization. There are arguably optimal doses of particular nutrients (i.e. Vitamin D) that would be extremely hard to achieve through

The question is one of optimization. There are arguably optimal doses of particular nutrients (i.e. Vitamin D) that would be extremely hard to achieve through "a balanced diet". Also most people do not and many can not eat a "balanced diet", as that requires a ton of effort to maintain. To put it another way, if the benefits of exercise were available in pill form on grocery store shelves, everyone would be taking that pill. Saying "you could save some money by just exercising normally!" is making a lot of assumptions about what people are willing and able to do, and frankly you'd be a fool to waste time exercising if the benefits were achievable through a magic pill. Supplements aren't quite that magical, but they make optimal, hard-to-achieve doses achievable, even on a bad diet. Granted they can also be misused, or doses can be so low/ingredients not bio-available such that they have no impact. They're a tool like anything else, and can be used correctly or incorrectly. Part of what bugs me about the meta-analysis saying "supplements don't reduce all-cause mortality!" is that they lump wildly different supplements with different make-ups, taken in different doses on different schedules from different starting diets with different amounts of discipline/regularity. Of course at a population level the number of people using them wrong probably outnumber the people who do their research and maintain a disciplined approach. Your average person has issues getting their car's oil changed and understanding the difference between synthetic and conventional oil. A skilled individual could properly apply and see benefits from supplements, just as a skilled individual could successfully change their own oil.