Research
What a score out of ten is actually counting
A number beside a seller looks like a measurement of the seller. On this site it is four yes-or-no facts divided by four, and knowing that changes what the number can be used for.
The arithmetic, in one line
The provider score on this site is the share of four declared facts that are true, printed on a ten-point display.
Four true facts make ten. Three make 7.5. Two make five. One makes 2.5, and none makes zero. That is the whole calculation.
Each fact is worth an equal quarter. Nothing is weighted, which means there are no weights to argue about and none to disclose.
That is worth saying plainly, because a weighted score is the more common shape. In a weighted score, the weights are the real editorial decision, and a reader who cannot see them cannot check anything at all.
A one-decimal number with five settings
The score prints with a decimal place, which suggests a fine gradation. Four yes-or-no inputs cannot produce one.
Only five values exist on this scale, and the gap between any two neighbors is one published fact. There is no such thing here as a seller half a point ahead of another.
So two sellers showing the same number are not close. They cleared exactly the same count of the same four questions.
The meter under the figure is drawn as ten separate segments rather than as a dial for this reason. A smooth arc would draw a continuum the rubric does not produce.
One of the four inputs separates nobody
Publishing a price somewhere on a seller's own site is true of the forty-five sellers here.
An input that every candidate clears adds the same quarter to every score and orders no one. It raises the whole field by one step and tells a reader nothing about which seller to pick.
While that holds, the lowest score this rubric can hand out in practice is 2.5 rather than zero. A floor is not a compliment.
The opposite case is also live. Including lab work inside the published price is rare enough on this roster that the flag behaves more like a bonus than a test. A seller clearing it moves further up than the fact by itself deserves.
A rubric input earns its place by splitting the field. Two of these four do most of that work, and a reader who knows which two can read the table faster than the number allows.
An absence and a denial cost exactly the same
A false flag records that a seller does not publish a fact. It is not a finding that the fact is untrue.
The score cannot hold that distinction. A seller that names no pharmacy because it uses several, and a seller that names none because it would rather not say, lose the identical quarter.
That is a real limit of any disclosure rubric, and it is the reason the four flags are printed beside the number rather than folded into it.
It also means the score is not evidence about a company. It is evidence about a company's website, and those are different objects.
The number belongs to the seller, never to the compound
A seller carries one score, and it is the same number on every board that seller appears on.
Nothing in it is about the compound in the row. A figure sitting beside a tesamorelin listing was earned partly by facts that have nothing to do with tesamorelin.
The published human evidence behind a compound is graded separately, by a different instrument, and that grade never moves with the seller. The two are meant to be read side by side and never merged.
What a rubric like this cannot reach at all
The four questions were chosen because a public page can answer them. That is a smaller set than the set of questions a buyer has.
Nothing in the number describes what happens after the card is charged. Not the cancellation, not a prescriber declining, not what arrives in the box.
Nothing in it describes what a seller says on the phone, which is often more than what it publishes.
And nothing in it is a clinical judgment. No score here says a company is safe, and none is positioned to.
What the federal rules say about a number like this, and what they do not
There is a federal trade regulation rule about reviews and ratings, made under the Commission's authority to define specific unfair or deceptive practices.
Read end to end, its operative sections prohibit six things. Fake or false reviews and testimonials. Buying reviews that express a chosen sentiment. Undisclosed insider reviews. Misrepresenting a review site the business controls as independent. Suppressing unfavorable reviews, and trading in fake indicators of social media influence.
One of its numbered sections is reserved and carries no text at all, which is worth knowing before anyone cites it.
Its definition of a consumer review is a consumer's evaluation, submitted by the consumer, and it says in terms that consumer reviews include consumer ratings whether or not they carry any words. A rating a publisher computes about a seller is not a consumer's evaluation, so it sits outside that definition.
Searched in one pass, the rule's text contains no use of rank, score, weight, formula, algorithm, methodology, ordering or comparison. Its only three uses of the word rating sit inside the definition of a consumer review and inside the suppression provision.
The plain reading is that nothing in it tells a publisher how to build a rating. That is the argument for reading a formula rather than trusting a number, on this site as on any other.
How to actually use it
Treat the number as an index into four facts, and then go read the four facts.
They sit in the same table. Which of them a seller cleared matters far more than how many, because they answer different questions and you probably care about one of them more than the rest.
If naming the pharmacy is the thing you want, a 5.0 that includes it beats a 7.5 that does not. The score cannot express that and the table can.
Key takeaways
- The score is the share of four declared facts that are true, each worth an equal quarter, with no weighting anywhere in it.
- Four yes-or-no inputs produce exactly five possible scores, so the smallest real difference between two sellers is one published fact.
- Publishing a price is true of the forty-five sellers here, so that input lifts everyone equally and separates nobody.
- A false flag means a fact is unpublished, and the arithmetic cannot tell that apart from a fact being untrue.
- The score describes a seller and never the compound in the row, which carries its own separate evidence grade.
- No federal rule prescribes how a publisher builds a rating, which is the reason to read the formula rather than trust the figure.
Frequently asked questions
Does a higher score mean a safer seller?
No. The score counts what a company publishes about itself, and publishing is not safety. A company can post every fact the rubric asks for and still be the wrong choice, and a company that posts almost nothing has not been shown to do anything wrong. Safety questions belong with a prescriber and with the state board that licenses the pharmacy, not with a comparison table.
Can a company pay to raise its score?
No, and the reason is structural rather than a promise. The function that produces the score reads four true-or-false fields and nothing else. Whether a company has a commercial relationship with this site is not one of them, so there is no input for money to move.
Why are there only five possible scores?
Because four yes-or-no inputs, each worth an equal quarter, can only produce five outcomes. None true, one true, two, three, or four. The decimal place in the display is a formatting choice, not a sign of finer measurement, and the ten-segment meter beside the figure is drawn to show the steps rather than hide them.
Does a low score mean a seller is doing something wrong?
No. Every one of the four inputs is a publication fact, so a low number means a seller publishes little. It may name its pharmacy on a call, ship to every state, and include bloodwork without saying so anywhere a buyer can read before paying. What the rubric measures is whether you can check before you pay.
Is this score a consumer rating in the legal sense?
No. The federal rule on reviews and testimonials defines a consumer review as a consumer's evaluation submitted by that consumer, and says consumer ratings count as consumer reviews for its purposes. A figure computed by a publisher from declared facts is not a consumer's evaluation and does not fall inside that definition. Nothing in that rule prescribes how such a figure should be built.
Sources
Each document below is named as it names itself, with the date printed on that document rather than the day it was read.
- Title 16 Code of Federal Regulations Part 465, Rule on the Use of Consumer Reviews and Testimonials, read in full — the definition of a consumer review and of consumer ratings at 465.1(d), the six operative prohibitions at 465.2, 465.4, 465.5, 465.6, 465.7 and 465.8, and the reserved section at 465.3 — Office of the Federal Register, Electronic Code of Federal Regulations, August 2024
- United States Code Title 15, Section 57a, Unfair or deceptive acts or practices rulemaking proceedings — subsection (a)(1)(B), the authority to prescribe rules which define with specificity acts or practices that are unfair or deceptive, and which is the authority cited for Part 465 — Office of the Law Revision Counsel, U.S. House of Representatives, January 2025