You search for the best magnesium supplement and land on a page that looks like journalism. There’s a ranked list. There are star ratings. There’s a line about how many hours the team spent researching. Every single product name is a link.
Those links are the business model. On most “best of” pages in the health category, the publisher earns a percentage of whatever you buy after you click.
That arrangement has a name, affiliate marketing, and it is not illegal, not hidden from regulators, and not automatically dishonest. Some of the most careful consumer journalism in the country is funded this way. But it creates a particular shape of pressure, and almost nobody explains that shape to the reader.
The audience is enormous. A National Center for Health Statistics data brief, published in February 2021 and built on 2017 to 2018 national survey data, put supplement use among American adults 20 and older at 57.6 percent within the prior 30 days. The share taking four or more products reached 24.9 percent in the 60-and-over group. That’s the market a commission page is selling into.
How the money moves
The mechanics are simple, which is part of why they’re invisible.
A publisher joins a merchant’s affiliate program, either directly or through a network that brokers thousands of programs at once. The publisher gets a unique tracked link for each product. When you click it, your browser picks up an identifier. If you buy within the merchant’s attribution window, the sale is credited back to that publisher, and the publisher is paid a share of it.
You pay the same price either way. The commission comes out of the merchant’s margin, not out of your checkout total. That’s exactly why the arrangement feels like nothing is happening.
Rates are not uniform. They’re set by the merchant and they vary by product, by category and sometimes by whether you’re a new customer. Some programs credit the publisher for everything in your cart, not just the item that was linked. None of this is secret. It’s just not on the page you’re reading.
What the structure rewards, stated plainly
Here is the part that deserves to be said without euphemism.
Affiliate revenue pays for recommending. It does not pay for recommending well.
A page earns nothing from “you probably don’t need this.” Nothing from “the evidence here is thin and mostly short-term.” Nothing from “fix your sleep first and see if the symptom goes away.” Nothing from “ask your doctor whether your blood work shows a gap.” Every honest answer that ends with you not buying something is, to the page, a loss.
Three more consequences follow from the percentage:
Expensive beats cheap. A commission on a $60 bottle pays several times what the same click pays on a $12 bottle. Nobody has to decide to favor costly products. The arithmetic does it.
New beats proven. Newer brands often launch with higher rates to buy attention. A cheap, well-studied, widely available generic may have no affiliate program at all, which means a commercial page has no financial reason to mention it and no mechanism to link it.
Listed beats unlisted. If a product isn’t in any affiliate network, it can’t be monetized, so it tends not to appear. The best-evidenced option in a category can be absent from a “best of” page purely because nobody pays to list it.
None of that requires anyone to lie. That’s the point. An incentive doesn’t need a conspiracy to bend output, and treating affiliate funding as inherently corrupt gets the problem wrong. Ad-funded and subscription-funded publishing have their own distortions. The question is which direction the money pushes.
What the FTC requires of anyone using affiliate links
An affiliate relationship is a material connection. Under 16 CFR 255.5, a connection between an endorser and a seller that might materially affect the weight you give an endorsement, and that you wouldn’t reasonably expect, has to be disclosed clearly and conspicuously.
The FTC’s staff publication “The FTC’s Endorsement Guides: What People Are Asking” gets specific about wording. It suggests language like “I get commissions for purchases made through links in this post,” and says the closer the disclosure sits to the recommendation, the better.
What doesn’t clear the bar is more interesting. The FTC says the phrase “affiliate link” is not an adequate disclosure, because “Consumers might not understand that ‘affiliate link’ means that the person placing the link is getting paid.” The same goes for “commissionable link.” A “buy now” button is not a disclosure either. By contrast, the FTC says “paid link” placed right next to an affiliate link should be adequate.
Two more points from the same guidance. The rules travel: they apply to affiliate links in reviews posted on someone else’s site and in short social posts, not just on your own page. And there’s a genuine carve-out. A manufacturer that links to all of its authorized sellers, without favoring the ones paying referral fees, doesn’t need to disclose anything, because there’s nothing being hidden about why a seller appears.
All of this sits inside the broader disclosure rules for endorsements, which this library covers in its own article on what an influencer is required to tell you. The short version is that a commission is one kind of material connection among many, and the standard for revealing it is the same standard that applies to cash, free product or a family tie.
When the page itself is the problem
Disclosure handles the link. It does not handle a page built to look like something it isn’t.
The FTC’s guide “Native Advertising: A Guide for Businesses” takes this on directly. A paid article that recommends or mostly shows a sponsor’s products is an advertisement, and the agency’s reasoning is that readers would otherwise assume it reflects independent views. The FTC judges an ad by its net impression, meaning the words, images, format and setting taken together, not element by element.
Its placement advice is unusually concrete. Put the disclosure immediately in front of or above the headline. Disclosures to the right of, below, or far from the headline are less likely to be noticed. Vague labels like “Promoted” are discouraged, and a logo or brand name alone is not enough.
In its guide for marketers on soliciting and paying for reviews, the FTC goes further and says pay-to-play comparison sites that sell better ratings or placement are deceptive, and marketers should not take part in them.
Then there’s the newest instrument. The Rule on the Use of Consumer Reviews and Testimonials, 16 CFR Part 465, took effect October 21, 2024. Section 465.6 makes it an unfair or deceptive practice for a business to misrepresent that a website, organization or entity it controls provides independent reviews or opinions about a category of businesses, products or services, including its own. In its questions-and-answers page, the FTC notes the provision reaches beyond review sites and can cover something like a certification program that hands seals to the company’s own products. Courts can impose civil penalties for knowing violations. There is no private right of action, so this is enforcement by the government, not by you.
The case that shows the mechanic
In 2020 the FTC settled with the operator of a loan comparison website. The agency alleged the site described its information as objective, accurate and unbiased while giving higher rankings and ratings to companies that paid for placement. The FTC announced the proposed settlement on February 3, 2020, and finalized it on May 26, 2020.
That was consumer finance, not supplements. The mechanic is identical. A site that sorts by who pays, while telling readers it sorts by merit, has a problem that no disclosure at the bottom of the page fixes.
What honest affiliate publishing looks like
Being fair here matters, because the reflex to distrust everything with a commission attached leaves you with no sources at all.
Three things separate a commercial page you can use from one you can’t.
Disclosure you find before the list, not after it. Placement is the tell. A site that means it puts the disclosure where you’ll read it first. A site hiding it in a footer or a legal page has made a decision about what it wants you to notice.
A methodology you can check. What products were considered? On what criteria were they ranked? What got excluded, and why? A page that describes its process in a way you could argue with is doing something different from a page that just asserts “we tested these.”
Evidence that the site says no. This is the strongest single signal and the rarest. Look for a product rated poorly, a category where the recommendation is to skip it, or a named pick the page tells you not to buy. A publication that has never once concluded “don’t” is not a publication that evaluates.
It’s also worth being honest about the alternative. A publication with no commercial relationship to anything it covers is rare, expensive to run, and usually smaller and slower than the pages that come up first in a search. Affiliate revenue is what makes detailed product coverage economically possible at all. The trade is real on both sides: you get coverage that would not otherwise exist, and the coverage has a thumb on the scale.
One warning against a comfortable assumption. Disclosure is necessary but it isn’t sufficient. A site that discloses its commissions and still ranks by commission is doing more damage than a site that ranks honestly and forgets to disclose. The disclosure tells you money is present. It says nothing about which way it pushed.
What researchers find when they grade these pages
The picture from the published literature is not flattering, though it’s also not the whole story.
A 2023 study in Integrative Medicine Research assessed 68 websites offering consumer information about dietary and herbal supplements for pain, drawn from top search results across four English-speaking countries. Reviewers scored them with DISCERN, a standard instrument for judging consumer health information. The mean overall quality rating came to 3.3 on a 5-point scale. Sites were weakest on describing uncertainty, on explaining what happens if you do nothing, and on effects on quality of life. The authors singled out commercial websites specifically, finding they frequently showed bias, failed to report the risks of these products, and offered little support for making a decision with a clinician.
Those three weak spots are not random. They’re precisely the places where honest content costs a commercial page money. Uncertainty makes a purchase less appealing. “What happens if you do nothing” is the free alternative. Risks are a reason to stop.
Readers are not naive about this. A 2026 discrete choice experiment published in Nutrients surveyed 732 adults in Thailand about supplements for cardiovascular prevention. Clarity of evidence was the most influential factor in their choices at 38.3 percent relative importance, ahead of who recommended the product at 24.3 percent, and far ahead of cost at 7.9 percent. Recommendations from social media figures were viewed unfavorably. That’s a non-U.S. sample and shouldn’t be stretched, but it suggests people are looking for evidence more than for a verdict.
There’s a complication worth sitting with. A 2026 experiment in Digital Health put this to the test with 289 participants and a set of long-form food videos, half of them sponsored. The sponsored versions made creators look more trustworthy and less authentic at the same time. Telling people about money does not automatically make them discount it.
Seven ways to read a “best of” page
Run these in about a minute, in this order.
- Count the no’s. Scroll the whole page and find one sentence telling you not to buy something. If there isn’t one, the page carries no downside risk and has no reason to warn you.
- Look for the losers. A real comparison names what it rejected and says why. A list with only winners is a catalog.
- Test whether the stated criterion survives. If the “best budget pick” costs more than three products that didn’t make the list, price was not the criterion.
- Check what doesn’t link. If every recommendation links out and the only unlinked mention is a competitor, you’ve learned the sort order.
- Chase the badge to the registry. If a product is described as verified or certified, look it up in the certifying body’s own public directory rather than trusting the logo on the page. We have a separate piece on how to tell a real certification scheme from a decorative one.
- Search the page for a disease word. If a page claims a product treats, prevents or cures anything, the commission is the smaller problem. Neither a seller nor anyone it pays can legally make that claim.
- Ask what the page would look like if the answer were “eat more fiber.” If that answer physically cannot fit on the page, the page cannot give it to you.
When a comparison table is the wrong tool entirely
There’s a moment where the whole exercise goes wrong, and it’s worth naming. It’s when you’re on a best-of page to deal with a symptom rather than to fill a gap you already know about.
Some things need evaluation, not a purchase. See a clinician instead of shopping if you have chest pressure or shortness of breath, unexplained weight loss, a new lump, blood in your stool or urine, a sudden severe headache, or symptoms that haven’t improved after a few weeks. A ranked list cannot examine you, and the delay while you try three products in sequence is the real cost.
Before adding anything, tell a clinician or pharmacist what you take. Three warnings from the NIH Office of Dietary Supplements explain why. Vitamin K undercuts warfarin. Add St. John’s wort and many medicines break down faster than they should. And some products thin the blood or leave anesthesia working differently on you before surgery. A comparison page has no idea what’s in your medicine cabinet.
The question to ask instead
Stop asking whether a site is paid. Most sites you’ll find in a product search are paid, in one form or another, and a page that earns nothing may simply be worse at the job.
Ask this instead: what does this site earn when the right answer is no?
If the answer is nothing, you now know which direction to correct for. Read the page for what it does well, which is usually narrowing a huge category down to a handful of real products, and then go get the part it can’t sell you. That part is in the clinical evidence, in the certifier’s registry, and in the five minutes with a clinician or pharmacist that no commission has ever paid for.
Keep exploring: sponsored research.
Sources
- Federal Trade Commission. “The FTC's Endorsement Guides: What People Are Asking.” ftc.gov
- Federal Trade Commission. “Native Advertising: A Guide for Businesses.” ftc.gov
- Federal Trade Commission. “Soliciting and Paying for Online Reviews: A Guide for Marketers.” ftc.gov
- Federal Trade Commission. “The Consumer Reviews and Testimonials Rule: Questions and Answers.” ftc.gov
- Electronic Code of Federal Regulations. “16 CFR Part 465, Rule on the Use of Consumer Reviews and Testimonials.” ecfr.gov
- Electronic Code of Federal Regulations. “16 CFR Part 255, Guides Concerning the Use of Endorsements and Testimonials in Advertising.” ecfr.gov
- Federal Trade Commission. “LendEDU, et al., In the Matter of (Matter No. 182-3180).” ftc.gov
- CDC National Center for Health Statistics. “Dietary Supplement Use Among Adults: United States, 2017-2018 (NCHS Data Brief No. 399).” cdc.gov
- NIH Office of Dietary Supplements. “Dietary Supplements: What You Need to Know.” ods.od.nih.gov
- PubMed. “Ng et al. "Dietary and herbal supplement consumer health information for pain: A cross-sectional survey and quality assessment of online content." Integrative Medicine Research, 2023. DOI: 10.1016/j.imr.2023.100996.” doi.org
- PubMed. “Rayanakorn et al. "Consumer Preference and Willingness to Pay for Nutraceuticals to Prevent Cardiovascular Diseases in Thailand: A Discrete Choice Experiment." Nutrients, 2026. DOI: 10.3390/nu18183083.” doi.org
- PubMed. “Pfender et al. "Your media diet is impacting your actual diet: The effects of influencer 'What I Eat in a Day' YouTube videos on influencer perceptions and nutrition behaviors." Digital Health, 2026. DOI: 10.1177/20552076261416379.” doi.org