How ScienceActually Works

Think about how you would hire someone for a job that actually mattered. You would look at what they have done before, whether it worked, and whether anyone independent can vouch for it. You would not hire them because their name kept appearing in your feed.

Most supplement formulation works the other way around. An ingredient trends, demand appears, and products get built to meet the demand. The question "what does the evidence actually support for this outcome" never really gets asked, because it is not the question driving the sale.

Selection done properly starts at the other end. You define what you are trying to support, you go to the human research, and the ingredient is the answer to a question rather than the reason for one. Sometimes that lands on something unglamorous that has been studied for forty years instead of the botanical everyone is posting about this quarter.

The tell is simple: ask a brand why this compound and not another. A research-led answer names outcomes and trials. A marketing-led answer names benefits and adjectives.

Wine makes this obvious. The same grape variety, planted on two hillsides a few miles apart, produces two genuinely different wines. Nobody finds that strange. Soil, sun, altitude and the day you pick all end up in the glass.

Plants used for supplements behave exactly the same way, but the label rarely admits it. A plant is a straw standing in dirt. Whatever the soil and water contain — minerals, or the things you do not want — gets pulled up and concentrated in the tissue. Two batches of the same species, one grown in mineral-rich volcanic soil and one in depleted farmland downstream of an old orchard, are not the same raw material in any meaningful sense.

Harvest timing compounds it. Many active compounds peak at a particular stage of the plant's life and fall away after. Harvest a fortnight late for logistics and you have a weaker input before processing has even started.

This is why "contains ashwagandha" tells you close to nothing. Which ashwagandha, grown where, picked when, and what did it test at on arrival.

A key that is the right metal and roughly the right size still will not open your front door. The shape has to match the lock. Nutrients work the same way — your gut has specific transporters, and a mineral has to arrive in a form those transporters recognise.

Magnesium is the everyday example. Magnesium oxide is cheap and it lets a label print a big milligram number, because oxide is mostly magnesium by weight. But it is poorly absorbed, so a large share travels straight through you, which is also why it has a reputation as a laxative. A better-absorbed form may show a smaller number on the label while delivering more magnesium to your cells. The bigger number is the worse product.

You cannot see any of this on the front of a bottle. It is on the back, in the fine print, in the words after the nutrient name — oxide, citrate, glycinate, bisglycinate. Those words are the whole story and they are printed in the smallest type.

A word of caution in the other direction: "enhanced delivery" and "liposomal" are marketing terms as often as they are technology. The right form is the one the body handles well, not the one with the most impressive name.

A scoop of coffee beans is not a dose of caffeine. Depending on the bean, the roast and the grind, two identical scoops can differ enormously in what they actually deliver. If you needed a precise amount of caffeine every morning, you would stop measuring scoops and start measuring caffeine.

That is the entire idea. Standardization means the manufacturer tests each batch and adjusts it so the active compound lands at a known concentration every time — not the weight of the herb, the amount of the thing that does the work.

Without it, potency rides on the harvest. A strong batch and a weak batch both go in the same bottle with the same label, and you have no way to tell which one you bought. It also makes the research meaningless: if a trial used an extract standardised to a particular concentration and your capsule is not standardised at all, you cannot assume the trial applies to what is in your hand.

This is the step most often skipped, because it costs money on every single batch forever, and no customer can see it missing.

Natural and clean are not the same word. A plant is an extremely efficient sponge — it drinks whatever is dissolved around its roots and stores it. Grow it in soil carrying lead or cadmium and the plant will faithfully concentrate them for you. Nothing has gone wrong; that is what plants do.

So the risk is highest in exactly the products people assume are safest. Root powders, greens, sea vegetables and clays are all things that spent their life accumulating from their surroundings.

Then there is what the factory adds. Extracts are pulled out of plant material using solvents, and traces can remain if the finishing is rushed. Botanicals are agricultural products that were recently outdoors, so microbial testing matters. And adulteration is real and documented — a cheaper species substituted for an expensive one, or an extract quietly bulked out with filler.

None of this is visible, smellable or tastable. It is only findable on instruments, which is why the machines further down this page matter, and why "third-party tested" with no detail behind it is a phrase rather than a fact.

There is a difference between an ingredient's résumé and this product's performance review. Most brands hand you the résumé.

Here is how the sleight of hand works. A study shows an ingredient does something useful — at a particular dose, in a particular form, in a particular group of people. A product includes that ingredient, often at a fraction of the studied dose, sometimes in a different chemical form. The marketing then borrows the study's conclusion wholesale. Every individual sentence is technically true and the overall impression is false.

Real substantiation means the trial was run on the finished product, at the dose in the bottle, and published where anyone can read it. That is rare and expensive, which is exactly why it is worth asking about.

Where it does not exist, the honest move is to say so and name what you do have — mechanism, traditional use, evidence on the ingredient rather than the product. That is why this site labels every claim by tier. A weaker tier stated plainly is worth more than a strong claim you cannot check.