Most people who buy a supplement have never had the relevant blood test, and most people who order a large private test panel did not need one. The honest answer to "what should I test first" is short: ferritin, read together with a marker of inflammation, is the one blood test that reliably answers a supplement question. Vitamin D as 25-hydroxyvitamin D is the correct test when there is a real reason to look, though the 2024 Endocrine Society guideline suggests against testing it routinely in healthy people. Vitamin B12 needs a second-line marker before a borderline result means anything. And three popular tests, serum magnesium, serum zinc and hair mineral analysis, cannot tell you what most sellers imply they can.
This article explains what each marker measures, where it fails, and when the useful answer is a conversation with your doctor rather than another purchase. Nothing here is a diagnosis, and none of it replaces medical advice.
What a blood test can and cannot see
A blood test measures a concentration in blood at one moment. That is not the same as the amount stored in your body, and for several nutrients the two are only loosely related.
The body defends the blood concentration of some minerals tightly, moving them in and out of bone and cells so that the circulating number stays stable while stores change quietly in the background. Magnesium is the clearest case: only about 1 percent of total body magnesium is present in blood, so the serum value is a small window on a large room [8]. Zinc behaves in a related way. Cellular, tissue and whole-body zinc are tightly controlled across a wide range of intakes, which the BOND expert panel describes as the core reason zinc status is hard to assess in an individual at all [10].
Other markers behave much better, because they genuinely mirror a store. Ferritin tracks iron stores. 25-hydroxyvitamin D is the circulating reservoir form of vitamin D and has a half-life of roughly 15 days, long enough to represent recent weeks rather than this morning [5].
So the first question about any test is not "is it available" but "does this number move when my stores move".
Why "within the normal range" means less than you think
A reference interval is usually built from the spread of values measured in a reference population, not from clinical outcomes. That distinction sounds academic until you see how it plays out. The serum magnesium reference interval of 0.75 to 0.95 mmol/L still in general use was derived from the distribution of values in a survey population from 1974, rather than from what concentration is associated with better health, which is exactly the argument for revising it [9].
There is a second, quieter problem. A Turkish biological variation study measured the same 22 healthy volunteers weekly for five weeks and calculated how much each marker varies within one person compared with how much it varies between people. The individuality index came out at 0.04 for 25-hydroxyvitamin D3 and 0.13 for ferritin [12]. A low index means each person occupies their own narrow band inside a very wide population range. Practically: a result sitting comfortably inside the population interval can still be unusual for you, and the more informative comparison is your own previous result rather than the printed range.
The same study found within-person variation of 16.9 percent for ferritin and 8.6 percent for vitamin B12 between weekly samples [12]. A ferritin that moves from 45 to 52 has not necessarily moved at all.
Ferritin: the marker that earns its place
Ferritin is the exception that makes the rest of this article worth reading. A systematic overview of 55 studies that used bone marrow examination as the reference standard found serum ferritin to be by far the most powerful laboratory test for iron deficiency anaemia, with an area under the receiver operating characteristic curve of 0.95 [1]. Very few nutrition markers come close to that.
The WHO guideline sets the practical thresholds: in apparently healthy adults a ferritin concentration below 15 micrograms per litre indicates depleted iron stores, and in adult non-healthy populations a concentration above 500 micrograms per litre may indicate a risk of iron overload or other disease [2].
The catch: ferritin rises with inflammation
Ferritin is also an acute phase protein, so infection, inflammation and some chronic conditions push it up independently of iron. The BRINDA project quantified how badly this distorts the picture across 27,865 preschool children and 24,844 women of reproductive age. Estimated prevalence of depleted iron stores moved from 4 percent in the group with the highest C-reactive protein values to 30 percent in the group with the lowest, with a similar pattern for the second inflammation marker [3]. Same populations, same ferritin assay, wildly different conclusions depending on inflammation.
WHO handles this by shifting the threshold: in individuals with infection or inflammation, a ferritin below 70 micrograms per litre in adults and below 30 in children may be used to indicate iron deficiency [2]. In plain terms, ask for C-reactive protein alongside ferritin, and treat a "normal" ferritin measured during an infection with suspicion.
One more honest note, because it matters more than any of our products: we do not sell iron. Iron is the nutrient where guessing carries real risk in both directions, and it is the clearest case for testing first and letting a doctor interpret the result.
Vitamin D: the right test, and the case against ordering it
If vitamin D is measured, the correct analyte is 25-hydroxyvitamin D. The active hormone, 1,25-dihydroxyvitamin D, has a half-life of about 15 hours and is not a status marker [5]. Ordering the wrong one produces a number that looks scientific and means nothing about your reserves.
The bigger surprise is who should be tested at all. The 2024 Endocrine Society clinical practice guideline found no clinical trial evidence supporting routine 25-hydroxyvitamin D screening in the general population, including in people with obesity or darker skin, and no clear evidence defining an optimal target concentration for disease prevention. On that basis the panel suggests against routine 25-hydroxyvitamin D testing in all the populations it considered [4]. The same guideline still suggests empiric supplementation for specific groups, including children and adolescents aged 1 to 18, adults aged 75 and older, people who are pregnant, and people with high-risk prediabetes, and it suggests against empiric supplementation above the dietary reference intakes to lower disease risk in healthy adults under 75 [4].
Read that carefully, because it inverts the usual sales logic: for many people the guideline supports a modest intake without a test, not a test that then justifies a large dose. If you want the seasonal picture, we covered it separately in how much vitamin D you need in summer versus winter. Our own vitamin D3 and K2 drops are dosed as 0.25 ml every three days rather than daily, and vitamin D contributes to the normal function of the immune system and to the maintenance of normal bones.
Vitamin B12: one number is not enough
Total serum B12 remains the most commonly used and most widely available test, but the choice often reflects availability rather than diagnostic performance, and its sensitivity and specificity are compromised by factors unrelated to B12 stores [6]. That is why holotranscobalamin, serum methylmalonic acid and plasma homocysteine have moved into routine practice as second-line markers [6].
The practical rule: a clearly low or clearly high total B12 is informative, a borderline value is not, and a borderline value is exactly where a second marker changes the answer. We looked at the age side of this in which nutrients absorb worse as you get older, where the failure is release of food-bound B12 rather than absorption of the crystalline form used in supplements.
Homocysteine deserves a warning of its own. It is a functional marker, but it is not a clean readout of B vitamin status. In 16,176 adults in the Hordaland Homocysteine Study, plasma total homocysteine was higher in men than women, rose with age, increased markedly with the number of cigarettes smoked, and was related to cholesterol, blood pressure, heart rate and physical activity [7]. A raised homocysteine is a reason to ask questions, not an answer by itself.
If you are simply looking for an everyday intake of the eight B vitamins in their bioactive forms, that is a dietary decision rather than a test-driven one: our bioactive vitamin B-complex contributes to normal energy-yielding metabolism and to the reduction of tiredness and fatigue.
Magnesium: the test most likely to mislead you
Serum magnesium is the test doctors order because it is the test that exists, not because it answers the question you are asking. With roughly 1 percent of body magnesium in blood, and bone releasing magnesium to defend the circulating concentration, a person can sit inside the reference interval while total body magnesium is drifting down [8]. The argument for a revised, outcome-based reference interval rests on exactly this: subclinical deficiency can exist despite a status that the current interval calls normal [9].
Two consequences follow. A normal serum magnesium does not prove your intake is adequate, and a low one is a genuine finding that belongs with your doctor. For most people the useful lever is not a test but the plate and, if intake is short, a sensible supplement. Our magnesium 7 in 1 delivers 251 mg of elemental magnesium in one daily serving, and magnesium contributes to normal muscle and nervous-system function and to the reduction of tiredness and fatigue.
Zinc: a population tool, not a personal one
Plasma zinc concentration does respond to severe dietary restriction and to supplementation, and cutoffs exist to identify populations at risk [10]. That is the level at which it works. For one individual on one morning, the BOND panel lists the interference plainly: plasma zinc responds less to zinc in food than to a supplement taken between meals, varies considerably between individuals for the same intake, and is influenced by recent meals, time of day, inflammation, and certain drugs and hormones [10]. Population reviews reach the same conclusion, recommending the percentage of low serum zinc values as a population indicator alongside intake data and child growth, rather than as an individual diagnosis [11].
So a single zinc result taken after lunch, during a cold, tells you close to nothing. Zinc contributes to the normal function of the immune system, and our zinc picolinate capsules supply 30 mg per capsule, but that is an intake decision rather than something a routine test can steer.
The test to skip entirely: hair mineral analysis
Hair mineral analysis is still sold as a way to map your mineral status and toxic exposures. A JAMA study split a single hair sample from one healthy volunteer and sent the pieces to six commercial laboratories that between them analysed about 90 percent of United States samples. Differences between the highest and lowest reported concentrations exceeded tenfold for 12 minerals, statistically significant extreme values appeared for 14 of the 31 minerals analysed by three or more laboratories, and the laboratories' own reference ranges varied greatly [13].
One sample, one head of hair, six answers. Whatever that measures, it is not your mineral status.
How to act on a result without fooling yourself
Three practical rules follow from the evidence above.
Retest on the marker's own timescale. 25-hydroxyvitamin D has a half-life around 15 days [5], so a change in intake needs roughly two to three months before a repeat test reflects it. Retesting after two weeks measures noise.
Compare against yourself, not the printed range. With individuality indices of 0.04 for 25-hydroxyvitamin D3 and 0.13 for ferritin, your own previous result is the more meaningful comparison, and week-to-week movement of 17 percent in ferritin can be ordinary biological variation [12].
Do not order a marker you cannot act on. The value of ferritin is that a low result changes what happens next, under medical supervision. A serum magnesium in the middle of its range changes nothing, which is why it rarely deserves a place on a self-ordered panel.
Frequently Asked Questions
Do I need a blood test before taking a supplement?
Usually not. For nutrients where a modest intake is sensible and safe, such as vitamin D within the dietary reference intakes, guidelines increasingly favour an appropriate intake over routine testing, and the 2024 Endocrine Society guideline suggests against routine 25-hydroxyvitamin D testing in the populations it considered [4]. Iron is the clear exception: test first and let a doctor interpret the result.
Which single test gives the most useful information?
Ferritin, requested together with C-reactive protein. Ferritin had an area under the curve of 0.95 for iron deficiency anaemia against bone marrow examination [1], and the inflammation marker is what stops a falsely reassuring result, since ferritin rises with inflammation [2][3].
Why is serum magnesium considered a weak test?
Because only about 1 percent of body magnesium is in blood and the circulating concentration is defended by release from bone and cells, a person can be inside the current reference interval while total stores fall [8]. That interval was also derived from the distribution of values in a population survey rather than from health outcomes [9].
Is a serum zinc test worth ordering?
For an individual it is hard to interpret. Plasma zinc is influenced by recent meals, time of day, inflammation and certain drugs and hormones, and varies considerably between individuals at the same intake [10]. It is recommended as a population indicator rather than a personal diagnosis [11].
My vitamin B12 is borderline. What now?
A borderline total B12 is not conclusive. Holotranscobalamin, methylmalonic acid and homocysteine are used as second-line markers precisely for that situation [6], and homocysteine on its own is affected by age, sex, smoking and other factors [7]. This is a conversation for your doctor, not a reason to self-prescribe.
What about hair mineral analysis or large private panels?
Hair mineral analysis performed poorly when one split sample went to six commercial laboratories, with more than tenfold differences for 12 minerals [13]. For any large panel, remember that reference intervals describe a population and individual markers vary from week to week [12], so more numbers mostly means more things to misread.
The Bottom Line
Testing is not the same as knowing. Ferritin, interpreted alongside inflammation, is the marker that genuinely changes decisions. 25-hydroxyvitamin D is the correct vitamin D test when there is a reason to order it, and current guidance says most healthy people do not need it. B12 needs a second marker before a borderline value means anything. Serum magnesium and serum zinc are widely available and widely over-read, and hair mineral analysis is not a measurement in any useful sense.
The most honest use of this article is subtractive: it should stop you buying two or three tests, not encourage you to buy five. And where a test does show something, the next step is your doctor, not our shop.
Sources
- Guyatt GH et al. Laboratory diagnosis of iron-deficiency anemia: an overview. Journal of General Internal Medicine, 1992.
- WHO guideline on use of ferritin concentrations to assess iron status in individuals and populations. World Health Organization, 2020.
- Namaste SM et al. Adjusting ferritin concentrations for inflammation: Biomarkers Reflecting Inflammation and Nutritional Determinants of Anemia (BRINDA) project. The American Journal of Clinical Nutrition, 2017.
- Demay MB et al. Vitamin D for the Prevention of Disease: An Endocrine Society Clinical Practice Guideline. The Journal of Clinical Endocrinology and Metabolism, 2024.
- Jones G. Pharmacokinetics of vitamin D toxicity. The American Journal of Clinical Nutrition, 2008.
- Sobczynska-Malefora A et al. Vitamin B12 status in health and disease: a critical review. Diagnosis of deficiency and insufficiency, clinical and laboratory pitfalls. Critical Reviews in Clinical Laboratory Sciences, 2021.
- Nygard O et al. Total plasma homocysteine and cardiovascular risk profile. The Hordaland Homocysteine Study. JAMA, 1995.
- Elin RJ. Assessment of magnesium status for diagnosis and therapy. Magnesium Research, 2010.
- Costello RB et al. Perspective: The Case for an Evidence-Based Reference Interval for Serum Magnesium: The Time Has Come. Advances in Nutrition, 2016.
- King JC et al. Biomarkers of Nutrition for Development (BOND) Zinc Review. The Journal of Nutrition, 2015.
- Gibson RS et al. Indicators of zinc status at the population level: a review of the evidence. British Journal of Nutrition, 2008.
- Ozkanay H et al. Biological variation of plasma 25-hydroxyvitamin D3, serum vitamin B12, folate and ferritin in Turkish healthy subjects. Scandinavian Journal of Clinical and Laboratory Investigation, 2023.
- Seidel S et al. Assessment of commercial laboratories performing hair mineral analysis. JAMA, 2001.
This article is information, not medical advice. Blood tests and their interpretation belong with a qualified healthcare professional. Food supplements are not a substitute for a varied and balanced diet and a healthy lifestyle.
