Trust and evidence
Why This Site Publishes No Star Rating
This site publishes no star rating because the review data supplied with the product it sells was marked as layout copy rather than real customer feedback. A rating is the most persuasive element on a product page and the hardest thing for a reader to verify, which is exactly why publishing an unverified one would have been dishonest.
What we were given, and what we did
Product packs sent to retail sites usually include review material: a star score, a review count, an order count, a distribution across five stars, and a handful of named customers with ages, cities and day counts. Ours did too.
It also carried a line most packs do not: a note stating that the star scores, order counts and reviews were layout copy. In other words, placeholder text for designers, not customer feedback.
So we published none of it. Not the score, not the count, not the distribution, and not the eight named people. What is on our reviews page instead is an explanation, a description of what we were sent, and a guide to reading reviews elsewhere.
Why write a post about it
Because the decision is more useful to you than the outcome. Every reader of this post will meet a supplement rating this week, and most of them will have no idea what stood behind it.
Why ratings are so persuasive
A star rating does something no other element on a page can: it converts a stranger's judgement into a number, and numbers feel like facts. Three things make it unusually powerful in supplements specifically.
The product is unmeasurable
You cannot test a supplement at home. In the absence of a measurement, other people's impressions fill the gap.
The claims are vague by law
"Supports energy" is not falsifiable. A 4.6 out of 5 feels like it answers a question the label legally cannot.
Nobody checks
A rating with a four-figure count and no link is accepted by almost every reader, almost every time.
Put those together and the rating is often the single most load-bearing element on a supplement page — and the cheapest to produce.
How supplement reviews go wrong
| Problem | What it looks like | How to spot it |
|---|---|---|
| Placeholder copy published as real | A tidy score with round-number counts | Names with no orders behind them, identical phrasing across reviews |
| Only positives shown | A page of five-star quotes | No star distribution, or one with no one- and two-star rows |
| Incentivised reviews | "Leave a review for 10% off" | Disclosure buried, or absent |
| Reviews of a different product | A brand with several SKUs pooling them | Review text mentioning a format the page does not sell |
| Repeated stock phrasing | Several reviewers using the same sentence | Read three in a row and the rhythm gives it away |
The first row is the one worth dwelling on. It rarely begins as deception: a designer fills a template with sample data, and somewhere downstream the sample data is published. The result is identical to a fabrication, which is why the only safe rule is to publish nothing you cannot trace to an order.
What a trustworthy rating looks like
- Every review tied to a real order.
- The data held by a platform rather than by the seller.
- A published moderation policy: what gets removed and why.
- The full star distribution, with the one- and two-star text shown at the same length as the five-star text.
- A way for a reader to verify any of the above without taking the seller's word for it.
That is not a high bar. It is simply the bar a number has to clear before it means anything, and most supplement pages clear none of it.
How to read reviews anywhere
Read the one- and two-star reviews first
They are where the specific information lives. Five-star reviews tend to be short and generic.
Check the day count
A review written on day three is about packaging and taste.
Look for the complaint inside the praise
"Good, but nothing in week one" is worth ten enthusiastic sentences.
Distrust numbers in review text
Nobody measures their own hormones at home.
Ask who benefits
If the seller controls the data, the data is marketing until proven otherwise.
What to judge a product on instead
Ratings summarise other people's expectations. These three are facts about the product, and all three can be checked without trusting anyone.
The label
What is printed and what is not. Serving size, amounts, sugar, base.
The research
Every ingredient with a reference page at NIH, NCCIH or MSK, linked and quoted.
The refund terms
How long, from which date, and who answers the phone.
On the product this site sells, those three are on the label facts, ingredients and guarantee pages. Two of the three do not flatter it.
What we would publish
Verified reviews, the day they exist: tied to real orders, with the low scores shown at the same size as the high ones, and a note saying where the data is held.
Until then, a blank space with an explanation is more informative than a number with nothing behind it. If you disagree, that is a reasonable position — but it is worth asking what the numbers on the next supplement page you read are actually standing on.
What this cost, in plain terms
Worth being concrete rather than self-congratulatory. A star rating with a review count is the highest-converting element on a typical supplement page. Removing it is the most expensive editorial decision on this website by a wide margin.
There was also an easier route available, and it is worth naming because plenty of sites take it: publish the numbers with a small note saying "sample data". That keeps the persuasive effect of a 4.6 out of 5 while technically disclosing it, because almost nobody reads the note under the stars.
Which is why we did not do that either
A disclosure that relies on not being read is not a disclosure. If the data is not real, the honest presentation is no data.
The one number we do publish
There is a score on this site: our own editorial score, currently 2.9 out of 5. It is worth explaining how that is different, because a seller scoring its own product is not obviously more trustworthy than a seller quoting customers.
The difference is that ours is checkable. Each of the seven rows names what it is scoring and why, the low rows say exactly what is missing, and every factual claim behind them is linked to a source or to the brand's own artwork. You can disagree with a weighting; you cannot be misled about what it rests on.
A customer rating works the opposite way: a single number, no reasoning, and no route back to the evidence. That is fine when it is real and tied to orders. It is worthless when it is not, and indistinguishable from the outside.
The test to run on any product page
Four questions, in order of how quickly they settle the matter. Where is the rating held — on the seller's own site, or on a platform? Can you see the one- and two-star reviews, at the same length as the five-star ones? Do the reviews mention specific, awkward details, or are they short and interchangeable? And does the page ever reach a conclusion that costs the seller money?
A page that passes all four is rare and worth trusting. A page that fails the first is not necessarily dishonest, but it is asking for a kind of trust it has not given you any way to verify — and that is the position every supplement buyer is in far more often than they realise.
One final thought
The uncomfortable part of writing this post is that it only exists because a seller pack happened to label its sample data honestly. Had that line not been there, we might have published the numbers in good faith — and so, presumably, do plenty of other retailers every day.
That is worth sitting with the next time you read a rating anywhere. The question is rarely whether a seller is lying. It is whether anyone in the chain ever checked.
Judge it on the checkable parts
The label, the research and the refund terms are all on this site.
60-day money-back guarantee · Questions first? Call +1 (302) 200-3480