What the tool does — and what it does not
The score rates the product as a product: whether the chemical form is the one that was studied, whether a daily serving reaches the dose that was tested, what can be verified about lab testing, whether the absorption claim is supported, how clean the formula is, and what an effective day of use costs.
The score does not say the supplement will do anything for you. It is not a medical claim, not a promise of an outcome, and not advice. "Should I take X?" is a question for a doctor, a pharmacist or a dietitian — not for this tool.
This tool sells no supplements, takes no affiliate commission, and takes no payment from brands. No product's score is connected to any revenue.
Six pillars, 100 points
| Pillar | Pts | What it measures |
|---|---|---|
| Form & absorption | 25 | Whether the chemical form on the label is the one the clinical research used. Ranked per nutrient against published literature. |
| Dose | 20 | The elemental or active amount in a realistic daily serving, against the studied dose range. Compound weight is always converted to active amount before scoring. |
| Testing & transparency | 20 | What can be independently verified, and at what resolution. A per-lot certificate of analysis from an ISO 17025-accredited lab is the ceiling; "third-party tested" with no detail is close to zero. |
| Delivery technology | 10 | Whether the absorption claim has human PK data for this nutrient, against a named comparator — and whether the nutrient has an absorption problem at all. A plain, correct formulation scores well; the absence of a technology is not a defect. |
| Formulation integrity | 10 | Proprietary blends, minerals competing for absorption in one capsule, ingredients at trace amounts, formats with known dose-stability problems. |
| Value & access | 15 | Shekels per day at the dose that actually works — not per pack — and how easy the product is to buy in Israel. |
Caps
A cap overrides the sum. A product can be excellent on five pillars and still be blocked, because one failure is enough to stop it working:
| Cap | When it applies |
|---|---|
| 45 | A realistic daily serving delivers under half the minimum effective dose |
| 50 | Omega-3 with no EPA+DHA figure per serving |
| 50 | A probiotic labelled at genus and species only, with no strain designation |
| 55 | A proprietary blend that hides the amount of the main active |
| 75 | An absorption claim with no named comparator, at a price premium of 2x or more |
| 40 | Exceeds an accepted safety upper limit with no indication — accompanied by a safety alert |
Grades: 85–100 Excellent · 70–84 Good · 55–69 Acceptable · 40–54 Weak · below 40 Avoid.
The evidence ceiling — when the ingredient is the limit
The six pillars rate the product: whether the form is the one that was studied, whether the serving reaches the dose that was tested, what can be verified, what it costs. They do not ask whether the ingredient does anything.
In most categories that does not matter — vitamin D, magnesium and omega-3 have an uncontested evidence base, and the only question is which product delivers them better. In some categories it matters a great deal: it is perfectly possible to make a clean, accurate, fully documented product out of an ingredient whose human research is weak or negative. Good formulation does not manufacture evidence.
So a category with a weak evidence base carries an evidence ceiling: the best product in the category cannot reach a grade the evidence does not support. The ceiling is not a truncation that flattens the category — above it the score is compressed into a narrow tail of at most four points, so you can still tell which product is better inside the category without the grade claiming more than the research shows. The ceiling appears on every product card with its reasoning and with links to the research itself.
| Ingredient | Ceiling | Evidence base |
|---|---|---|
| Nattokinase | 65 | Weak evidence |
| Serrapeptase | 46 | Almost no evidence |
| Lumbrokinase | 54 | Weak evidence |
| Bromelain | 62 | Weak evidence |
| Aged garlic (AGE) | 70 | Moderate evidence |
| Pine bark | 59 | Weak evidence |
| Grape seed | 59 | Weak evidence |
| Horse chestnut | 73 | Moderate evidence |
| Diosmin / hesperidin | 68 | Moderate evidence |
A ceiling is an assessment of the state of the research on the date at the top of this page, not a safety finding and not a claim about any company. If research has been published that changes the picture, that is exactly what the corrections route is for, and the ceiling will be updated with the change documented.
Two kinds of record — and the difference matters
verified — checked by hand
The label panel and the price were read by hand from the retailer's product page on the date shown in the card. Scored on all six pillars. Every source is linked from the card.
automatic — imported from a label database
The record was imported automatically from the NIH Dietary Supplement Label Database (DSLD), a public database of over 214,000 registered labels. In practice this means:
- Scored on five pillars only. The database carries no price, so the value pillar is not scored and the total is normalised over the remaining pillars.
- The testing pillar is set at brand level — a known company testing programme — not at the lot level of this specific product. It therefore scores below what a per-lot certificate would earn.
- The registered label has not been checked against the actual package, and may be older than or different from the product sold today.
So: before relying on an automatic record, open the source link in the card and verify against the product page. The tool has a filter for verified records only.
Where the data comes from
- Label panels — manufacturer and retailer product pages, and the NIH Dietary Supplement Label Database.
- Form rankings and dose ranges — published literature: controlled trials, meta-analyses, and the professional fact sheets of NIH ODS. Each nutrient's ranking rests on cited sources.
- Testing standards — the certifying bodies' own public registries: USP, NSF, Informed Sport, IFOS.
- Prices — retailer product pages, on the date recorded in the card. Israeli supplement prices move often and are promotion-driven; treat them as a band.
Fact, opinion, and our responsibility
The facts in every card — the chemical form, the amount per serving, what appears on the label, the price and its date, which certificates exist — are taken from public statements by the manufacturer, the retailer or the certifying body, and are linked to their source.
The score itself is an opinion — a professional assessment derived from those disclosed facts according to the rubric on this page. It is not a factual assertion about the company, its intentions, or the safety of its products.
We do not characterise a product as dangerous, harmful or counterfeit in our own voice. Where a documented risk exists, we cite the source that established it.
Corrections — for brands, importers and distributors
If a fact about your product is wrong — form, dose, price, a certificate that exists and was not counted, or a product no longer on the market — we are glad to correct it, and no process is required.
Contact: [email protected]
How it works:
- Send the product name, the incorrect figure, and the correct figure with supporting evidence (a label photograph, a certificate of analysis, a product page).
- We acknowledge receipt within 3 business days.
- A factual error is corrected within 14 days of receiving the evidence, and the score is recalculated automatically.
- If we disagree with the correction, we explain in writing what we disagree with, and we will publish your position alongside the card on request.
Every correction is recorded in that product's change log. The rubric itself is versioned: if a score falls because of a methodology change, the change is documented here and is never made against one product or brand in isolation.
Limitations we are aware of
- Israeli coverage is thin. Most records come from a US database and reflect products bought by personal import, not the local pharmacy shelf.
- The testing pillar measures transparency, not quality. A brand that publishes no certificates is not necessarily a bad brand — it is a brand that cannot be verified. In Israel almost no brand publishes a per-lot certificate, and that is a statement about the market rather than about any one company.
- Prices go stale. The check date appears on every card for exactly that reason.
- The rubric weighs what can be measured from the label. It cannot see what happens in the factory.