Reviews, and what they are worth
This site did not collect a single review. What follows are the seller's own figures, published as the seller's, plus an audit of the sample text that came with them. No review markup is emitted on any page of this site.
The seller publishes 4.3 / 5 from 803 written reviews against 6,386 orders. Those are the seller's figures, published here as the seller's.
Eight sample reviews were supplied. They contain two distinct bodies of text, four instances of each, word for word. This site emits no review or rating markup on any page.
What the seller published
- Score
- 4.3 / 5
- Written reviews
- 803
- Orders
- 6,386
- Would recommend
- 90%
- Most common wait before noticing anything
- 3 to 4 weeks
- Most popular pack
- 6 bottles
- Most mentioned positive
- the daily serving, and the 180-day window
- Most mentioned negative
- slow start, wanted more in week one
Read that as the seller's claim
Every figure above came from the seller. We cannot verify any of it, we have no access to the underlying reviews, and we are not presenting it as independent. It is published because withholding it would be worse: a buyer is entitled to see what the seller claims and to weigh it accordingly.
Why no stars appear in search results for this site
Structured data can attach a star rating to a page so that search engines show it. Doing that with figures we did not collect and cannot verify would be passing off the seller's claim as this site's finding.
- No
AggregateRatingmarkup is emitted on any page. - No
Reviewmarkup is emitted on any page. - The sample text supplied with the figures is not published as customer testimony anywhere on this site.
- The
Productmarkup on this site carries an offer block, because prices exist and were published. It carries no rating.
That decision costs this site the star rating in search results. It is the right trade and it is the same decision we would make again.
The audit
Eight sample reviews were supplied with the figures. We ran the simplest possible check on them: do any two say the same thing in the same words?
Eight entries. Two distinct bodies of text. Four instances of each, word for word, attached to eight different names, ages, cities, pack sizes and day counts.
That is not a finding about the product. It is a finding about the sample text, and the most likely explanation is template copy assembled for layout rather than eight real reviews. Which is precisely why this site does not publish them as customer reviews.
What the numbers would mean if they are real
Set the audit aside for a moment and read the figures as given.
- 803 written reviews against 6,386 orders is a 12.6% review rate. That is high for this category, where low single digits is typical.
- A 4.3 average with 66% five stars and 5% one or two stars is a plausible distribution rather than an implausibly perfect one.
- 90% would recommend against a 4.3 average is internally consistent.
- A most-common wait of 3 to 4 weeks is notably honest for a seller to publish. It undercuts the impulse purchase.
- Six bottles as the most popular pack is what you would expect when the six-pack is 38% cheaper per day and carries the same 180-day window.
The internal consistency is worth noting in both directions: it is what real data looks like, and it is also what carefully constructed data looks like. We cannot tell from here, so we are telling you what we can and cannot tell.
How to read supplement reviews generally
- Read the three-star reviews first. They are the least likely to be either incentivised or furious, and they usually contain the specifics.
- Look for a mechanism in the text. 'I sleep better' is a claim; 'I stopped waking at 3am after week three' is an observation.
- Check whether the reviewer says how long they took it. A review at day four of a twelve-week ingredient is not evidence of anything.
- Watch for repeated phrasing across reviewers, which is the check run above and takes thirty seconds.
- Discount your own reading of them. Reviews are the weakest evidence in this category and a label you can do arithmetic on is the strongest.
That last point is the reason this site leads with the label rather than the reviews. The panel is a document the manufacturer signed its name to. A review is a sentence.
See the Cognizil packs
Two, three or six bottles, priced plainly, with a 180-day window on the sales page kits. Read the best case first, then decide.
Shipping is calculated at checkout. This site publishes no figure it was not given.
The eight entries, as supplied
For completeness, here is what was supplied, with the duplication shown rather than described. These are not published as customer reviews anywhere on this site and they carry no schema.
| Name and age | Stars | Pack | Day | Body text |
|---|---|---|---|---|
| Luis P., 64, Scottsdale AZ | 5 | 3 bottles | 16 | Text A |
| Pete N., 56, Nashville TN | 3 | 6 bottles | 47 | Text B |
| Kevin C., 59, San Antonio TX | 4 | 6 bottles | 41 | Text B |
| Nick K., 41, Houston TX | 5 | 6 bottles | 18 | Text A |
| Bill H., 65, Denver CO | 3 | 2 bottles | 38 | Text B |
| Omar J., 60, Miami FL | 4 | 2 bottles | 21 | Text B |
| Miguel P., 57, Chicago IL | 5 | 6 bottles | 46 | Text A |
| Kevin E., 53, Detroit MI | 3 | 6 bottles | 52 | Text A |
Scroll this table sideways →
- Text A
- “Daily habit stuck. Reordered before the last bottle.” — four entries, word for word
- Text B
- “Fair at a month. Wanted more in week one.” — four entries, word for word
Note the star ratings
Text B appears on a three-star entry and on a four-star entry. Text A appears on three five-star entries and one three-star entry. The same sentence carries three different ratings across the sample, which is the clearest indication that the text and the metadata were generated separately.
What we would need to publish reviews properly
This site would publish customer reviews, with schema, if the following were true. None of it is, currently.
- The reviews are collected by a system that verifies a purchase, so that a review is tied to an order rather than to a form.
- The full set is published, not a selection. A curated subset is a marketing asset rather than a dataset.
- Negative reviews appear with the same prominence as positive ones.
- No review is edited for content, and any moderation policy is stated.
- The collection method is described, including whether anything was offered in exchange.
- The text is not templated. A duplicate-wording check of the kind run above passes.
Six conditions, none of them unusual, and several review platforms meet all six routinely. Until this product's reviews meet them, the figures on this page stay labelled as the seller's and no star rating from this site will appear in a search result.
What the review figures cannot tell you
Even a perfectly collected, perfectly honest review set has a limit that applies to this whole category, and it is worth stating because it is not about honesty at all.
- Nobody reviewing a supplement has a control group. A reviewer cannot know what would have happened had they taken nothing.
- Reviews are written by people still taking it. The ones who stopped at week two are systematically absent.
- The outcome being reported — memory, clarity, focus — is the single outcome most affected by expectation and by attention.
- Reviews cluster at the extremes. The middle, where most experience actually sits, is under-represented.
- A review at day 16 is a review of the first two weeks of a twelve-week ingredient.
That last row applies to three of the eight entries supplied here. It is not a criticism of the reviewers; it is a reason the label is a better document than the review section on any product in this category, including this one.
How this site would rather be judged
Reviews are the weakest evidence available in this category and the easiest to manufacture. A label is a document a manufacturer signed its name to, and arithmetic on a published total is something any reader can recompute.
So if you take one thing from this page, let it be that the review figures at the top are the least useful information on this site, and that the least useful information is the part every competing product page leads with.