# Editorial standards, independence and corrections > Who writes here, who reviews it, how independence is kept and how mistakes are corrected. - Language: English - Page: https://smilehandbook.com/en/editorial-standards/ - Français: https://smilehandbook.com/fr/normes-editoriales/ - العربية: https://smilehandbook.com/ar/editorial-standards/ - All published pages: https://smilehandbook.com/llms-full.txt How Smile Handbook is written, checked, funded and corrected is what this page sets out. Its purpose is to make our claim of independence testable instead of something you have to take on trust. Where a safeguard is already in place, we say so. Where one is planned but not yet built, we say that too, in plain words, since no reader can judge our reliability from a description of the site we mean to become. What follows describes the site as it stands today. ## Who writes these pages Two named people stand behind the content pages. Dt. Esma Özkaya, a dentist, writes them; Muhammet Sadıç is the medical editor who checks them before publication. Both names sit at the foot of every page they are responsible for. Pages that describe the site itself rather than a treatment subject — this one among them — are the founder's own work and carry no such line, because attributing an account of our own method to a dentist would be false. No newsroom stands behind this site, and claiming one would be the first thing you could catch us on. We are not patient recruiters. The byline now has a written record behind it: the sign-off is dated 6 September 2026. That approval was given in conversation on 1 September 2026 rather than on paper. What that record was not is a page-by-page review; the line that works page by page came later: its first signature records were written on 3 September 2026, and our approval-line page sets out what it does. Every page you can open here that carries a clinical judgement has now been through it and holds a dated signature record, but the date sits in that record rather than on the page, so a reader still sees who checked a page and not when. The checks you hold over us are that record, the corrections log and the evidence audit. We do not treat patients, we do not arrange appointments, and we do not have a commercial relationship with any dental clinic in Turkey or anywhere else. By "appointments" we mean appointments for clinic treatment; we do not arrange travel bookings either. A separate free general-information conversation with a volunteer dentist may be offered when its contact channel is open. Our method is documentary and quantitative, not clinical. Take a representative example: across six dental clinics in Istanbul we gathered 2,115 publicly visible reviews ourselves and counted the star ratings by hand. Of those reviews, 93.85 percent carried five stars. Saturation that heavy traps the whole-corpus average mathematically between roughly 4.75 and 4.94, which leaves the score itself carrying almost no information. The bound holds for the corpus, not for each clinic inside it: three of the six clinic averages sit outside it, and the margins are far from equal — one far below at 4.20, and two above it by much smaller margins, 4.97 a little further out and 4.95 only barely, all counted review by review and rounded only for printing. We then recomputed each clinic's average from our own counts and set it against the figure the platform publishes. The divergence never exceeds 0.052; on the platform's one-decimal display five of the six land on the same figure, and on the sixth ours drops a notch — 4.9 against a displayed 5. We do not know when the platform computed the figure it shows. We also read 50 one-star reviews line by line — every one-star review in the corpus with enough text to describe what happened: at least seven describe a failure that appeared only after a delay, and at least seven describe losing contact with the clinic after returning home. That is work we can stand behind, and work of that kind is what we publish. ## What we looked for and did not find We searched that same body of reviews for the signatures of manufactured feedback. We did not find them. The median five-star review ran 187 characters, template-like phrasing appeared in 6.3 percent of texts, and reviewer-farm patterns — a single account reviewing more than one of the six clinics — surfaced in 14 accounts out of 2,085. Because we found no such evidence, we do not write that clinics commission fake reviews. An accusation we cannot support is not published here, however well it would perform. Review bursting, the clustering of reviews in time, is one thing we tried to measure and could not. Our data collection tool capped out at roughly 490 reviews per clinic, and 43.9 percent of the dates we did obtain were reported only as a year. So we do not say bursting is absent. We say our design could not measure it. ## What is not yet in place The medical editor position is filled. Muhammet Sadıç is the named medical editor, and Dt. Esma Özkaya, a dentist, writes the content pages he checks; both names sit at the foot of each page they are responsible for. The approval line behind those names now runs, and two limits stay behind it. It runs like this: a page carrying clinical judgement is read by the medical editor, and a dated signature record is then either written against that page or withheld; the page describing our approval line sets out what such a record holds. Its first signature records were written on 3 September 2026; the pages those first records stood against were still drafts. All 41 pages written here that carry clinical judgement have since been signed; twenty of those signatures were written on 6 September 2026, and the last sixteen signatures were written on 10 September 2026. The approval we held before the line was given in conversation on 1 September 2026, and put in writing only on 6 September 2026; it was given for the site as a whole and was not the output of that line, and the records written since are made one page at a time. The two limits are these. We publish no credential class for either name, because we hold no document to support one and would rather print nothing than print a qualification we cannot evidence. And no page published here prints a date of review, so the byline says a page was checked but not when: the date sits in the signature record and not on the page. The rule we apply still caps what we publish rather than asserting who checked it. Smile Handbook issues no clinical judgments. We do not tell readers what treatment they need, whether a proposed plan is appropriate, or what a particular symptom means. We report what we counted, describe how procedures and aftercare are generally organised, and send clinical questions to a treating clinician. A page written here that carries clinical judgment appears only once a signature record has been written against it, and if its clinical passages are rewritten afterwards the signature falls and the page comes down until it is signed again. ## Where the money comes from, and where it does not Smile Handbook earns no revenue at present. Advertising is the plan, later on. Once advertising starts, it will be labelled as advertising and kept apart from editorial content. The following will not be for sale at any price: inclusion on this site, position in any list, the wording of any page, the removal of an unflattering finding, or advance sight of anything we are preparing. No clinic can pay to appear here. We do not take commission, referral fees or booking revenue, and we do not operate a booking service for clinic treatment or travel. We publish no clinic rankings, no "best" lists, no price promises and no star badges. ## Declaring conflicts of interest Anyone who writes or reviews for this site declares any payment, gift, free or discounted treatment, travel, equity or family relationship tied to a clinic, laboratory, agency or supplier in this field. Once declared, an interest disqualifies that person from any work touching the party concerned. Where an interest exists and the work goes ahead anyway under supervision, the disclosure appears on the page itself, not only here. ## When you tell us we are wrong Write to us, set out what is wrong and, where you can, what the correct position is. Before we reply, we check it against our own records. If we were wrong, we fix the page and add a corrections log entry recording what the page said before, what it says now, and the date. No correction is made silently. If we were right, we say so and lay out the evidence. And where the matter resolves neither way, we record the point as disputed and describe both positions. Our corrections log is published, every entry in it is dated, and it only grows: an entry is added, never rewritten and never removed. The reason it is that short is that the log has only just been opened, not that nothing else on this site will ever need correcting. A log this young is a starting condition, and showing it beats not showing it at all. ## Sources used on this page - _calisma/bir-yildiz.json - _calisma/bir-yildiz.json · plan/16-YORUM-METNI-OLCUMU.md:355,360 - _calisma/plan-olcum/puan-band-olcum.json - _calisma/yorumlar/*.json - _calisma/plan-olcum/sablon-izi-olcum.json