Every change we make to the magnesium ranking, dated and explained — and the four things that cause one.

A ranking that never changes is a ranking nobody is maintaining. This page is the record of every change we make to the magnesium ranking — what moved, when, and why — so you can check our work instead of taking our word for it.
Four things, and only four. They are the four inputs of the score, applied identically to every product in every category. The full method is here.
Brands reformulate more often than anyone expects, and they rarely announce it. A change to the magnesium form, the elemental dose per serving, or the serving size itself changes the biggest input in the score — ingredients and dosing, at 35%. When we can verify the new label, we re-score against it. When we cannot, the product is marked unverified rather than quietly re-rated on a guess.
Price per serving is 20% of the score, and it is the input that moves most often. We re-check prices monthly. A permanent price change re-scores the product; a two-week promotion does not, because a ranking that chases discounts is a ranking that tells you nothing next month.
Verified user reviews are 35%. Volume and recency both matter, so a product with three hundred recent reviews carries more weight than one with thirty from four years ago. This input moves slowly and rarely flips a position on its own.
Gaining or losing a third-party certification changes what we can say about a product. It does not change the dose — a certified product can still be underdosed — but it changes our confidence that the label is true. What the seals mean →
Money. A brand can buy a labelled slot on this site; it cannot buy a position, a score, an award, or the deletion of a flaw we found. If that ever changed, this page would be the first place you would see it. Advertising disclosure →
Nothing before that date, because there was nothing before that date. We would rather have a short honest log than a long invented one.
Price per serving is 20% of the score, so it is worth saying exactly what that means. We take the price at a major US retailer, divide by the number of servings the label declares — not capsules, not grams — and re-check monthly.
Serving size is the part that catches people out. A product at "two capsules per serving" in a bottle of 60 is a 30-day supply, and comparing its shelf price against a one-a-day competitor's compares two different things. Our figures convert to the same unit first.
Brands reformulate quietly and often. When we spot a change, one of three things happens:
Label figures come from the product's filing in the NIH Dietary Supplement Label Database wherever it has one, in preference to marketing copy. The database records what was submitted to the regulator; a product page records what the marketing team wanted to say.
Six products, in a category with hundreds. That is deliberate. A product joins when we can rate it end to end — verified label, checked price, real review history — and not before. A list of forty products we half-checked would look more authoritative and be worth considerably less.
It also means the ranking will grow, and this page is where those additions get recorded.
The dose bar under each product is the whole argument in one line: the filled portion is the elemental magnesium on the label, and the vertical mark is where the studied range begins. But the bar cannot show absorption, and that is why the cheapest product here finishes last. Nature Made's oxide declares a real 250 mg of elemental magnesium — a longer bar than the glycinate above it — and still ranks bottom, because oxide is poorly absorbed. Read the bar and the form together. Why the form eats the dose →
Maintained by the NFS editorial team · Last updated 06 Aug 2026. Spot something wrong? Tell us and we will fix it with a dated note on this page.