The problem, briefly
When we started collecting external opinion on the four brands we cover, DoNotAge, IM8, NOVOS, and Thorne, we found that a large majority of the top-ranking YouTube reviews and blog write-ups had a discount code, a coupon, or a purchase link embedded in the same post. That's not automatically dishonest, but it's structurally biased: it is very hard to earn commission from a product while giving it a 4/10.
How we handled it
- We built a three-tier source hierarchy, A, B, C, plus an openly listed Rejected tier. Any source with an affiliate link is marked accordingly on its card.
- Affiliate-linked sources can still contribute facts (ingredient names, doses printed on labels, formulation histories) if the facts are independently checkable. They cannot contribute a rating.
- Ratings and star scores from affiliate-linked sources, for example, biohacker-style blueprint sites that give a 9.4/10 next to a buy button, are excluded from the ranking. We list them under Rejected so you can see what we excluded and why.
A specific case: Forbes
Forbes has written favourably about both IM8 and NOVOS, and both brands quote Forbes in their marketing. Forbes is a real publication, so it isn't obviously in the same category as a coupon blog. But: the Forbes coverage in question is business journalism about founders and companies, not a product test. It doesn't measure a dose or open a capsule. We put it in the Rejected tier and explain why, rather than pretending it doesn't exist.
Business journalism about a founder isn't a product test. We list it, then we set it aside.
What we accepted
In Tier A we accepted: Innerbody Research (long-form independent reviews with methodology disclosed), El País and Daily Mail (independent press citing named experts), Healthline for Thorne specifically, and, the strongest signal in the whole set, the independent German lab that assayed DoNotAge's NMN and confirmed roughly 99% purity.
In Tier B we accepted registered dietitians and physicians reviewing on YouTube without affiliate codes, and one fitness-focused amino-blend teardown that produced the citable 1.58 g figure on IM8's amino component.
Tier C, Trustpilot, Reddit sentiment, is used as background only. It cannot move a product up the leaderboard. It exists on the page so you can see the operational picture.
What this changes for you as a reader
If you look at a product on our site and its ranking looks lower than what you see elsewhere, the most common reason is that we did not count the affiliate-linked reviews other sites did. If it looks higher, it usually means a Tier A source (Innerbody, an independent lab report, an unaffiliated dietitian) had a positive read that other sites missed.