How We Grade Longevity Providers: Our Methodology

Longevity medicine is a field selling care ahead of its own evidence. No intervention marketed by a longevity clinic — not rapamycin, not NAD+, not peptides, not plasma exchange — has been shown in a completed randomized trial to extend human lifespan, in part because the methodology to run true human healthspan trials is only now maturing and validated aging endpoints barely exist yet¹. That gap is why a ranking site in this niche has to grade differently than one ranking, say, mattresses. Nobody can honestly grade these providers on whether their therapies work, because nobody knows. What can be graded — precisely, repeatably, from the provider's own published pages — is whether the business is straight with you about what it sells and what it costs.

This page is the rubric, and it describes the calculation that actually runs. That distinction matters, because until July 2026 this page described a different one: five weighted axes — evidence honesty 30%, clinical oversight 25%, transparency 20%, price-to-value 15%, conflicts of interest 10% — scored A through F. Three of those axes had no stored value for any provider anywhere in our data. The letter on our provider rankings was never computed from them and could not have been, and the only way to make the page true would have been to invent twenty evidence-honesty scores to justify grades that were produced some other way. We rewrote the page instead. What follows is a smaller claim than the old one, and it has the advantage of being checkable: every input below is a field you can verify on the provider's own site in about two minutes.

The five factors

| Factor | What it measures | Points | |---|---|---| | Price transparency | One all-in figure, visible before you commit, that a month-to-month buyer actually pays — halved where the provider's own terms disclaim it | 4 | | Longevity line | Whether the provider actually sells NAD+, peptides/sermorelin, hormone optimization or longevity diagnostics | 4 | | Clinical oversight | A licensed clinician's review included in the published price, not billed on top or left unstated | 2 | | Pharmacy quality | What the provider publishes about its dispensing pharmacy | 2 | | Human support | Nurses, dietitians, coaching or unlimited clinical messaging inside the price | 2 |

Fourteen points in total. Nothing else scores — not the size of the menu, not the brand, not whether we are paid by the provider, and not how much we like the copy.

Why price transparency is worth four

Price is the failure this category commits most often, and it is the one a reader cannot check without handing over a card number. Across every provider we grade, the published headline is frequently not a price at all: it is a twelve-month prepay rate presented as a monthly figure, an introductory first month that steps up at the first refill, an intro-code discount, or a number the site itself supersedes with "final price after clinical review". Several providers in the current ranking advertise a monthly figure that no month-to-month buyer is ever charged. Two make the point without any interpretation from us: a "no hidden fees" headline that the site's own footnote reveals is a yearly-subscription rate, and a storefront that shows a struck-through list price beside a code-gated one and then says the real amount is set after clinical review.

So the four points go to one thing: a single all-in figure, visible before signup, that a buyer with no commitment actually pays. A number reachable only on a six- or twelve-month prepay scores zero. A number quoted only inside the portal scores zero.

There is one middle rung, added on 8 August 2026, and it changed several grades here. Some providers publish exactly the figure this factor asks for — ungated, month-to-month, all-in — and then, in their own terms or their own fine print, tell you it may not be what you are charged. "The final charge to your credit card may fluctuate contingent upon the prescribed medication and the chosen pharmacy." "All prescription medications require a valid and complete online consultation prior to approval and final pricing is determined." "Price may vary depending on your particular biochemistry and provider recommendation." Those are three real sentences from three rows here, each sitting underneath a real published price.

A figure the seller does not stand behind is materially less transparent than one it honors — and until this rule existed the rubric could not tell the two apart, which meant it actively rewarded publishing a price and then disclaiming it. So the factor now has three rungs:

• •Four points — an all-in figure, published before you commit, that the provider stands behind.
• •Two points — the same figure, published and then disclaimed. Better than publishing nothing, because you have a number to start from and to hold them to. Worse than a price that binds.
• •Zero — no such figure exists before commitment at all.

Two guardrails on it, both deliberate. It is not triggered by "prices are subject to change" or "we may modify our fees": a forward-looking right to reprice is universal boilerplate, it appears on most rows here, and docking for it would hit the whole ranking and measure nothing. And it is not triggered by a caveat about a different product from the one we grade — a marketplace whose terms say the cost of a prescription depends on what is prescribed is still publishing a firm price for the appointment we rank it on. The rule was applied by sweeping every row here that publishes a price — twenty of them, page by page, on the day it was written — rather than to the row that prompted it. Three came back positive, and the dock is taken inside this factor and floored at zero, so a provider that publishes no price cannot be charged twice for the same failure.

Why the longevity line is worth four

This is the factor added in July 2026, and it exists because the rubric could not previously tell a longevity provider from a weight-loss shop with tidy billing. A GLP-1 operator with one clean monthly price, an included consult and a support line could score full marks on every factor and land at the top of a page a reader opened to find NAD+ and peptides. The letter was measuring billing hygiene and calling it longevity.

The factor asks one question of the provider's own catalog: does it sell what this page ranks — NAD+, peptides or sermorelin, hormone optimization or TRT, or longevity diagnostics? A published line of the business earns all four points. Something adjacent but none of those — hormone therapy for symptom relief, or a microdosed GLP-1 the provider itself frames as a longevity protocol — earns two. A business selling weight loss and nothing else earns zero.

The weight is what makes it bite. A provider with no longevity line can reach at most 10 of 14 even with a flawless price, an included consult, a published 503A standard and included support — and the A band starts at 11. That is deliberate: on a longevity ranking, the top band is closed to a business that does not sell longevity medicine, regardless of how well run it is.

It is worth being clear about what this factor is not. It is not a claim that the products in a "longevity line" work. GLP-1 receptor agonists have the strongest human outcome data in the entire field — the SELECT trial cut major cardiovascular events by roughly 20% in overweight and obese adults without diabetes² — and NAD+ and sermorelin have nothing remotely comparable behind them. A provider selling only GLP-1s may well be selling the better-evidenced product. It is simply not selling what this site ranks, and we would rather say that plainly than quietly rank it first. We trace the evidence in GLP-1s for healthspan and longevity.

This factor is also researched rather than assumed. Values are set by opening the provider's own catalog and reading it. That check has already corrected us once: a provider we had written off as GLP-1-only turned out to run its own anti-aging category with injectable NAD+ at two strengths and injectable sermorelin in it, priced month-to-month. A brand's marketing angle is not its menu.

Clinical oversight — two points

Is a licensed clinician's review included in the price, or is it billed on top, gated behind a membership, or simply not mentioned? Two points for included; nothing otherwise. This is the axis that separates a program that prescribes from a lab that only measures, and it is scored on the narrow, checkable question rather than on the quality of the clinician, which no outsider can assess from a website.

Pharmacy quality — two points, with a middle rung

Nearly everything in this category is compounded, so who compounds it matters. The points ladder is about disclosure, and it turns on one question: what has the provider itself published, and where?

• •Two points — the provider publishes a 503A claim on a surface it controls: its own site, its help center, or its official press material. Naming the facility is not required, and prominence does not matter — a line in a footer counts exactly as much as a line on a product page.
• •One point — the provider identifies the dispensing pharmacy by name, address and phone, without classifying it.
• •Zero — "partner pharmacies", "a wide network across all 50 states", "compounded in the USA", "a licensed U.S. pharmacy". A category is not an identification.

Two things do not count as a 503A claim. A third-party review, directory entry or newswire write-up is not a surface the provider controls, however confident it sounds — several providers here are described as 503A operations by other websites and say nothing of the kind themselves. And a general explanation of what Sections 503A and 503B are, or a compliance page listing "503A requirements where applicable" among the regulations a company observes, is not a statement about who fills your prescription. We read the sentence around the keyword before scoring it.

The middle rung was added in July 2026 for a specific reason. One provider here printed both of its dispensing pharmacies with street addresses and phone numbers — at that point the most pharmacy-transparent row we graded — and scored zero, because neither was described as a 503A facility. Disclosure that specific has to be worth something. It is worth less than the standard being stated, so it sits in the middle rather than at the top. Several rows have since joined it there, including one that prints four pharmacies with phone numbers alongside two prescribing physicians and their NPI numbers and still never says under which regime any of them compounds.

We had been applying this inconsistently, and the drift favored us

In July 2026 we re-verified every row here against this one test and found we had not been applying it evenly. Our own #1 provider was scored full marks for a 503A line in its help center. Two other providers made the same claim — one in its site-wide footer, one on the product page you actually buy from — and were scored zero. A third provider's FAQ stated that every pharmacy it works with is 503A certified, and nobody had ever looked at it; it had been scored zero for four months and moved up a whole grade when we did.

Fourteen rows now qualify for the top rung and nine sit on the middle one, after a board-wide re-audit on 8 August 2026 that found several of our own "no pharmacy named" findings were wrong — read from one document when the answer was on another page of the same site. Applied evenly, the test does not flatter us: our own provider's 503A evidence is among the thinnest that qualifies — a help-center line, naming no facility — while several others publish the claim more plainly than we do. The consequence is that our #1 pick no longer holds the top score on its own; two providers now tie it. We are leaving that alone. A rule rewritten until the favorite is back on top by itself is not a rule, and the whole point of writing this test down is that it has to be able to produce an answer we did not want.

Where a provider states the standard but still will not say which pharmacy, its row says so. That distinction is real and it is worth knowing: "a 503A pharmacy, we won't say which" and "here is our pharmacy, at this address" are different kinds of disclosure, and only a handful of providers here do the second.

Human support — two points

Nurses, registered dietitians, coaching or unlimited clinical messaging, included in the price rather than sold as an upgrade. The distinction being drawn is between a program and a shipped vial. Plenty of providers ship a good product with nobody attached to it; that is a legitimate business, and it costs two points here.

What we deliberately do not score

Required lab work scores neither way. It is printed on every row because it changes the experience, but a lab-gated program is making a defensible clinical trade-off, not committing a failure. Rewarding or punishing it would smuggle a preference into a rubric that is supposed to measure disclosure.

Evidence honesty, price-to-value and conflicts of interest are not scored either — the three axes this page used to claim carried 60% of the grade between them. Not because they do not matter; they matter enormously. They are not scored because we have no honest way to reduce them to a per-provider number, and a rubric that assigns a provider a "27 out of 30 on evidence honesty" is publishing a precision it does not have.

They live in the written reviews instead, where a judgment can be argued rather than scored:

• •Growth-hormone marketing. A provider selling sermorelin or other GH-secretagogues as anti-aging is running ahead of a landmark systematic review that found small body-composition changes but no proven functional benefit and significantly more adverse events in healthy older adults³. That is written into the provider's review; it does not move the letter.
• •Biological-age claims. Epigenetic clocks like GrimAge genuinely predict lifespan and healthspan at the population level⁴, but individual test-retest reliability has been weak enough that a single reading can swing by years on re-measurement — a problem the field has had to engineer around⁵. A provider selling one clock reading as a precise "true age" is overselling the instrument. We unpack that in do epigenetic clocks actually work.
• •Over-screening dressed as thoroughness. The whole-body MRI anchoring many concierge pitches is a double-edged screen: a 2026 systematic review of more than 9,000 asymptomatic adults found a confirmed-cancer detection rate of just 1.57%, alongside frequent incidental findings, unstandardized protocols, and no long-term outcome or cost-effectiveness data, concluding the scans "may lead to unnecessary investigations"⁶.

If you want the full proven-versus-hyped map we hold every claim against, it is in longevity medicine: what's proven vs hyped.

From points to a letter

Scores convert to letters on two lines that were fixed before any provider was scored:

• •A — 11 to 14. One point above the 10 a provider with no longevity line can reach with everything else perfect.
• •B — 8 to 10. More than half the scale, short of the A floor.
• •C — 0 to 7. Half the scale or less.

Neither line is allowed to move to change who sits on which side of it. A rubric adjusted until the partners win is not a rubric. Both cut-points were fixed before any provider was scored and neither has been touched since — including through the July 2026 pharmacy re-verification described above, which moved two providers' letters. When that happened we moved the providers, not the lines.

That re-verification is the honest test of whether this holds, so here is what it did. Two rows changed band, and neither change helped us: an affiliate partner rose from B to A, and an unpaid provider we had scored C for four months rose to B once we actually read its FAQ. One partner gained points and stayed C regardless — the standard it publishes is real, but a hedged "we use both 503A and 503B pharmacies, we won't say which" does not fix a program that will not publish the price you actually pay.

The floor is real, and it lands on paid providers. Every C grade in the current ranking belongs to an affiliate partner: providers that publish no true month-to-month price, or do not state whether a consult is included. We are paid by all of them and they are graded C anyway, because the calculation is not shown who pays us.

Grading and ordering are two different decisions

This is the part most ranking sites blur, so we will be blunt about it.

The letter is computed from the five factors above and from nothing else. The function that computes it is not given the provider's rank, is not given whether the provider is a paid partner, and is not given whether it is our featured pick — those fields are not in its inputs, and an automated test flips every commercial flag on every row and fails the build if any letter moves.

The order is not computed the same way. Affiliate partners are listed first on the rankings page, ahead of the providers we hold no commercial relationship with. That is a placement, we are paid for it, and we disclose it — the same separation of commercial placement from editorial judgment that our editorial policy commits us to, and that who we are and how this site is funded sets out in full. Within each block — partners, then everyone else — rows are ordered by their rubric score, highest first.

So a paid row can and does sit above a better-graded unpaid one. The position tells you who pays us. The letter tells you what the provider publishes. If the two ever seem to agree suspiciously well, the letters are the thing to check, because they are the thing we cannot move.

We re-audit graded providers periodically, because prices, protocols and product lines all change — and when a headline price turns out to be a prepay rate, or an anti-aging category quietly appears on a weight-loss site, the score follows the field rather than the other way round.

For the rankings this methodology produces, see our graded longevity provider shortlist; for the full write-up behind every grade, browse our provider reviews; and for whether any of it is worth buying, are longevity clinics worth it.