Most of these products fail their own label test.

On 26 February 2025, ChromaDex published Quantitative Analysis of Nicotinamide Riboside Consumer Products: 39 US NR products, bought off the shelf, run by validated HPLC-UV assay.

Two caveats. First: ChromaDex sells NR — they are the company behind Tru Niagen — so this is a test commissioned by an interested party. Hold it exactly as you should be holding this article. Second, and more important: the reputable brands passed. Tru Niagen, both Thorne products and Elysium all measured over claim. This is not a story about everybody lying. It's a story about a long tail — the $25 bottle of "NAD+" that is mostly filler — sitting on the same results page as products that are exactly what they say, indistinguishable from the outside.

Which leaves one practical rule: the ingredient deck is not the product. The product is what's inside the capsule, and there's one way to know the difference — a third-party assay tied to the lot number on the bottle in your hand. Not a generic PDF. A batch-matched certificate from a named lab.

That standard includes us. Our current lot is out for independent assay and we'll publish the number when it lands. Until then, treat this as a standard we're asking to be held to, not one we claim to have cleared.

A bigger number on the front is a warning sign, not a feature.

In that same analysis, every single 1,500 mg product tested — all four of them — had no detectable NR. Not "some." None.

No conspiracy is required to explain this. Nicotinamide riboside is one of the more expensive raw materials on a supplement shelf; the number printed on the front of the bottle costs nothing. When a 1,500 mg claim arrives attached to a $25 price, the arithmetic has already told you something the label hasn't — which is why the experienced end of this market now reads mg-inflation as a scam signal rather than as value.

Related tell: products promising to deliver NAD+ itself rather than a precursor are making an absorption claim the evidence doesn't obviously support. The serious formulas sell a precursor — NR or NMN — and say so on the label.

The number on the front of the bottle is the cheapest thing in the box.

"Proprietary blend" means you are not permitted to check the dose.

A worked example, provable from the label alone. Life Extension's NAD+ Cell Regenerator with Resveratrol Elite lists "Resveratrol Elite™ Proprietary Blend — 55 mg." Fifty-five milligrams sounds like a resveratrol dose. Inside that blend, the actual trans-resveratrol is 10 mg. Same structure on the same panel: "Bio-Quercetin® — 14 mg" contains 5 mg of quercetin.

None of this is unlawful, and Life Extension publishes the real figures if you go looking. But the blend name is doing work the dose can't, and a buyer who reads "55 mg" and moves on has been told something untrue by a label that never said anything false.

Two labels can print different numbers and mean nearly the same thing. Read the form, not just the figure.

Nicotinamide riboside is not sold as free NR. It is sold as a salt, and the milligram figure on the panel includes the salt. Thorne's ResveraCel prints 415 mg — nicotinamide riboside hydrogen malate. Ours prints 500 mg — nicotinamide riboside chloride. The fraction that is actual nicotinamide riboside differs by form, and the two forms are not the same.

We could tell you what that works out to. We're not going to, because we haven't verified it against a physical Thorne panel in our own hands, and unverified arithmetic presented as fact is much of how this category lost its audience. Instead, both labels, side by side.

Table 1 — swipe to compare →

Transcribed from each brand's published label and current shelf pricing, July 2026. Thorne's range reflects its own label, which specifies two to four capsules daily. Note what the table makes visible: the two products disclosing every dose can be compared with each other. The one with blends can only be compared with itself.

Nobody has shown these products raise NAD+ inside the tissue everyone is talking about.

This is the sentence the whole category is careful not to say, so we'll say it. Elhassan and colleagues, 2019: in older men taking NR, whole-blood NAD+ more than doubled. In the same men, in the same trial, NAD+ measured in muscle tissue was statistically unchanged — p = 0.22.

So "replenishes NAD+ in your cells" is a claim nobody in this aisle has earned, and it is on a great many of these pages. "Raises NAD+ in the blood" has been earned repeatedly, in humans, in controlled trials. Two different sentences, and the difference is not pedantry — it is the open question of the category.

And the trials that came back empty. Dollerup 2018 — 12 weeks at 2,000 mg a day, euglycaemic clamp — null across every metabolic endpoint measured. Three separate randomised cognition trials — null on their primary outcomes. Martens 2018's blood-pressure signal, one of the most-quoted results in NR marketing, was exploratory and not statistically significant; the same lab did not replicate it in 2026.

The direction where the human evidence is least thin is inflammatory markers — Elhassan reported reductions in IL-6 and TNF-α, the one area even the sceptics' review of the NR literature endorses. If a sales page claims substantially more than that, the page is running ahead of the data.

The dose-response is real, and it plateaus — so the problem is at the bottom of the range, not the top.

Conze, Crespo-Barreda & Kruger, Scientific Reports 9:9772 (2019). 140 adults, randomised, double-blind, eight weeks, three doses of NR. Whole-blood NAD+:

And the detail that matters more than any of those three numbers: the 100 mg arm decayed. By week eight it had drifted back to roughly +10%. The 300 mg and 1,000 mg arms held.

So the lesson isn't "more is better" — the curve flattens and there's no reward for chasing the top of it. The lesson is that there's a floor, it sits somewhere around 300 mg, and a great many products on the shelf are dosed below it. Much of the "I took it and felt nothing" in this category may simply be people who bought 100 mg.

What we won't do with that chart. No trial has ever measured chronic blood NAD+ at exactly 500 mg, so any brand quoting a percentage for a 500 mg dose — ours included — is interpolating between two data points and calling it data. Mark down anyone who does. What is on the record at 500 mg is safety, not effect size, and we'll label it as such: EFSA concluded in 2021 that 500 mg/day of NR chloride is safe for adults; Martens 2018 and the 400-patient NOPARK phase III trial both used 500 mg as their per-dose unit; NR-SAFE found no moderate or severe adverse events at 3,000 mg/day.

Cost per day is the only honest comparison in this category, and almost nobody publishes one.

Prices here are printed per bottle, and bottles hold different numbers of servings at different label doses. A $40 bottle and a $95 bottle can cost the same to take, and frequently do. The only figure that compares anything is price divided by days, at each label's own stated daily dose. Nothing below is a brand's claim; each is arithmetic on a live price.

Table 2 — swipe for full range →

Computed as price ÷ days at each label's stated daily dose, live pricing July 2026. Where a brand runs a discounted first order, we used the ongoing price, not the loss-leader.

Two honest notes on our own table. The floor is real: bulk sublingual NMN powder genuinely costs about a quarter a day, and if cost is your only variable, that's the answer — it's just a different product form, weighed out on a scale, and most people don't stay with it. The ceiling is real too: one IV session costs more than a year of most products above it. Between those poles sits the actual market, in a narrow band — about $1.20 to $2.50 a day, for a promise that is nominally identical across all of them.

Celluvate, held to the checklist above.

We built the formula before we wrote the checklist, so parts of this are luck rather than virtue. Here is where it lands.

Dose disclosure: every active named with its milligram figure and botanical source. No blends. Salt form: printed on the panel — chloride, the form EFSA assessed. The floor: 500 mg sits above the 300 mg level that held for eight weeks in Conze, and well below the 3,000 mg at which NR-SAFE found no moderate or severe adverse events. Cost per day: $0.88 to $1.33 — the whole reason we were willing to print Table 2 at all. Where it doesn't clear: the batch assay isn't published yet. Until it is, point one is unanswered for us, and we won't pretend otherwise.

Three things we're not going to tell you

• That the ingredients work synergistically. The NICE trial (McDermott et al., Nature Communications, 2024) tested that exact combination. NR alone: +17.6 m of walking distance. NR with resveratrol: +3.65 m, non-significant, with more nausea and diarrhoea. The authors' words — resveratrol did not add benefit to NR alone. Here they are companions to the NR, and that's all we claim.
• That resveratrol extends anything. The Interventions Testing Program found no lifespan extension, and the original sirtuin finding is now attributed to an artifact of the dye assay used to measure it. Anyone still selling resveratrol on that story is selling a withdrawn result.
• That the galactomannans multiply absorption. There is a real human study behind that claim, run at roughly 2:1 carrier-to-active. Ours runs about 1:20. Importing that data would be the most misleading thing available to us. So: a carrier, nothing more.

Nobody will tell you when you'll feel something. So here it is, hedged honestly.

Across the fifteen largest brands in this category, not one publishes a straight answer to the most obvious question a buyer has. Elysium tells you that you won't feel anything. Most go quiet. One offers a week-by-week timeline built on an uncontrolled customer survey written up in the grammar of a clinical trial.

We have no trial data on how this formula feels, because no such trial exists — not for ours, not for anyone's. So what follows is not a promise. It's what the published trials and our own small number of reviews suggest, carrying the uncertainty they actually have.

• Weeks 1–2 — most likely nothing. One trial (Dolopikou, 2020) found a same-day functional effect from a single 500 mg dose, but that's one measurement in a laboratory, not a description of an ordinary Tuesday.
• Weeks 3–6 — if anything shows up, usually here. The vocabulary people reach for is stubbornly unglamorous: steadier mornings, an afternoon that doesn't collapse into the flat stretch where sentences get shorter and the second screen is too much information. Nobody describes it as a stimulant.
• Months 2–3 — the reports become about absence rather than presence. Something that used to happen at a particular hour has stopped happening, and it is hard to notice a thing that isn't there.
• The most reliable pattern in the consumer literature: people don't become convinced when they start. They become convinced when they run out.

You probably won't notice it starting. You may notice it stopping.

Which is why our guarantee runs 60 days rather than 14 — and why the most useful instruction we can give is this: take it consistently for six weeks, then stop for two, and pay attention to the second week — a schedule that finishes inside the 60-day guarantee rather than a fortnight past it.

Some people will do exactly that and notice nothing. Here is one of them, copied from our own review page, where she still is:

We could have removed that. We have twelve reviews in total; taking one down would move the average visibly. It stays because a review page with no three-star reviews isn't a review page, it's an advertisement — and because Adelaide's experience is a genuine outcome this product has for some people, and you ought to be able to price it in before you spend $39.99 rather than after.

Four more, chosen for the same reason:

4.7 out of 5, across 12 reviews. Twelve. We'd rather show you the number than a badge implying a bigger one.

One price, printed per day, so you can put it back into Table 2.

Every bottle is 60 capsules — 30 days at two capsules a day.

One-time purchases. No subscription, no auto-refill, nothing to cancel later. Free US shipping.

See Celluvate →