About Know Your Mouth

What this site is

Know Your Mouth publishes plain-English explanations of ordinary dental questions — why breath goes bad, what bleeding gums mean, whether the whitening trick you saw online does anything. The aim is to answer the question properly and then stop, including when the honest answer is “the evidence here is thinner than people claim”.

What we are not

This is the important part, so it goes near the top rather than buried in a footer.

• We are not dentists. Nobody writing here holds a dental or medical qualification. We are not able to diagnose anything, and we do not try to.
• This is not medical advice, and it is not a second opinion. It is general information, written carefully, by people who read the research rather than people who treat patients.
• We have not tested any product in a laboratory. When we review something, we are analysing publicly available information about it — not measuring it.

If you want to know what is happening in your own mouth, you need someone who can look at it. Nothing on this site is a reason to delay that.

Who writes it

One person: Jose Pollman. No editorial board, no rotating freelancers, no “medically reviewed by” badge bought from an agency. If a page here is wrong, one person got it wrong, and the same person fixes it.

The relevant background is not clinical, and pretending otherwise would be the first thing to distrust. It is reading — specifically, reading claims against the documents they are supposedly based on and reporting where the two part company. The same work happens on the other sites I build: notspoofed.com checks whether a domain's published email policy does what its owner thinks it does, vetthisvendor.com checks whether a supplier's claims survive their own paperwork, and whattofixfirst.com measures web performance rather than repeating what a score said. Dental content is the same discipline pointed at a different literature.

Longer write-ups of that method, including the one that produced the flossing correction below, are published on Medium.

Why you should read us anyway

Because a lot of what is published in this niche is worse. Search almost any dental question and you will find pages assembled to rank, stuffed with claims nobody checked, pointing at a supplement. The bar we are trying to clear is simply: say what is actually established, say what is not, and be explicit about which is which.

Where we make a claim that matters, we say where it comes from. Where the research is weak, we say it is weak — even when a stronger claim would sell better. Our review method sets out how that works in practice.

How the site makes money

Affiliate commission, from exactly one product. Our ProDentim review contains an affiliate link, and if you buy through it we earn a commission from the seller at no extra cost to you. That is the only affiliate link on the site — no article contains one.

We would rather tell you that plainly than have you work it out. The full terms are in our affiliate and health disclosure.

Where the facts come from

The rule is that a claim on this site has to be traceable to something we actually opened. In practice that means the paper, the systematic review, or the regulator's own document — not a summary of it, not another dental site, and not a search-result snippet. Where the full text was behind a paywall or a bot check, we say which record we read instead, and we report the finding qualitatively rather than printing a number we could not see.

• Cochrane reviews first, where one exists on the question. They are explicit about certainty, and their “low-certainty” wording is usually the most honest sentence available on a subject.
• Then the abstract of record — PubMed, Europe PMC or PMC — for individual trials, so the volume, year and participant numbers come from the indexed record rather than from someone's blog post about it.
• Then national guidance, such as England's Delivering better oral health toolkit, when the question is what a health service actually advises.
• Never a claim we could only find on a page selling something.

Who paid for the research gets stated in the same sentence as the finding. Not in a footnote, not at the bottom — in the sentence. A great deal of the evidence in consumer dentistry is funded by the companies whose products are being tested, and a reader who learns that three paragraphs later has already formed a view. That rule is why the hydroxyapatite page names the manufacturer that funded the trials it relies on, and it applies to research we like as much as to research we do not.

The full method, including what we deliberately do not do, is on how we review products.

Corrections

The policy is short. If a page is wrong, we change it, we say what changed, and we leave the admission on the page — we do not quietly edit and re-date. Errors of fact get fixed as soon as we know about them. Where a claim turns out to be weaker than we made it sound, that is also an error and gets the same treatment.

The corrections we have published so far, and the reason this section exists:

• Water flossers, September 2026. Our page described the trials that share one co-author as “paid for by the manufacturer” and “sponsor-funded”. Only one of them says so: the 2025 paper states that its first author was a Water Pik, Inc. employee and that manufacturer money reached the research unit, and the 2018 paper lists a Water Pik affiliation. For the 2005, 2008, 2011 and 2013 papers the funder is not stated. Our funding ledger recorded that correctly and the article overstated it, so the article was brought into line with the ledger, not the other way round.
• Flossing, September 2026. Our page said the well-known review of interdental cleaning found that the studies were poor quality, rather than that flossing does not work. That is true, and it was incomplete. Reading the review's results properly turned up a stronger and more specific fact: across 35 trials and 3,929 adults, no trial measured decay between the teeth at all. We corrected how often should you floss, and wrote up what the review did and did not measure in what the flossing evidence actually measured. The longer account of finding it is on Medium.
• Brushing after acid, August 2026. Three pages carried the familiar “wait 30 minutes before brushing” advice as settled. It is not: a 2020 meta-analysis found no benefit to delaying. All three were corrected in one pass, and the erosion page now explains where the rule came from and what happened when it was tested. The one exception that survived is that you should not brush straight after vomiting.
• Tongue scraping, September 2026. Our page said scraping helps breath “more than almost anything else you could add”. That was too strong. The Cochrane review it rested on contained two trials in forty people, reported no data for its own primary outcomes, and was withdrawn in 2016; the effect it did measure could not be detected half an hour later. The page now carries that correction and the evidence behind it. The practical recommendation did not change — the confidence did.
• How many species live in your mouth, September 2026. We said “more than 700”. It is a figure repeated everywhere, and we could not find a first-party source for it. The Human Oral Microbiome Database itself lists 836 taxa across the mouth and aerodigestive tract, 525 of them primarily oral, and the paper describing that database puts it at “over 600 prevalent taxa at the species level”. We now say hundreds, and cite both. The point the number was making — that the mouth is a large community and most of it is not hostile — is unchanged. A number you cannot source is a number you should not print, including when it flatters your own argument.

The diagrams are free to reuse

Every illustration on this site is drawn here, as inline SVG, because the alternative in this niche is stock photography that looks like supplement advertising. They are released under CC BY 4.0, which means a hygienist can put one in a patient handout and a teacher can put one on a slide without asking us. The collection, with the attribution line to copy, is on the diagrams page.

Three small games

There are three games here, and each is about one idea. Marketing or evidence? deals ten advert claims a day and asks which are supported; every verdict is lifted from an article on this site and links to it, because the point is the reading, not the score. 48 Hours is an arcade game about the two-day window between plaque you can brush off and tartar you cannot. Mouth garden is a slow ecosystem game about the one idea the whole site rests on: a healthy mouth is a balanced community, not a sterile one, and the product that kills everything is the one that loses. None of them sends anything anywhere, none has a grade beyond the count, and all three say on the page what they get wrong on purpose.