Langues

Guide : The Meta Charset Tag

Ce que <meta charset="utf-8"> fait, pourquoi the HTML spec veut it in the premier 1024 bytes, how a incorrect encoding causes mojibake, and pourquoi it's a rendering-correctness problème plutôt que a ranking factor.

Première publication : 2 juil. 2026 · Dernière mise à jour : 3 août 2026 · Advanced
Langues
AIRédigé parAI v2
1 indice probant sur cette page

The meta charset tag — <meta charset="utf-8"> — declares votre page's character encoding so navigateurs and robots d’exploration turn raw bytes into the correct characters. The HTML spec exige it dans the premier 1024 bytes of the document, and meilleur pratique is the literal premier child of <head>. Obtenir it incorrect (manquant, late, or a mismatched encoding) and vous obtenir mojibake: accented letters, curly quotes, em dashes, non-Latin scripts, and emoji render as garbage — qui peut corrupt what's affiché and indexé. It n’est pas a direct ranking factor; Google's seulement guidance is to 'utiliser Unicode/UTF-8 où possible.' UTF-8 is the near-universal, spec-required encoding pour HTML5 today. A server-sent Content-Type header charset overrides the in-page tag, qui is a courant source of migration bugs. Ce is un of le navigateur-facing tags in the meta-tags cluster.

TL;DR — <meta charset="utf-8"> declares the document’s character encoding. The WHATWG HTML spec exige the declaration to be serialized complètement dans the premier 1024 bytes of the document, and pour HTML5 the valeur doit match utf-8; meilleur pratique is placing it as the literal premier child of <head>. A manquant, late, or mismatched encoding produces mojibake — corrupted accented characters, curly quotes, non-Latin scripts, and emoji — qui is a rendering and indexing-correctness problem, pas a ranking signal. Google’s seulement public line is “use Unicode/UTF-8 where possible.” A server-sent Content-Type charset header overrides the in-document tag, qui is a classic post-migration mojibake bug. Seulement un charset meta element is allowed per document, and it has aucun effect in XML. A UTF-8 byte-order mark (BOM), si présent, wins over everything sinon; sinon the HTTP header wins over the in-page tag — complet precedence order ci-dessous.

Evidence for this claim For HTML documents, the charset declaration must identify UTF-8. Scope: Modern HTML conformance requirements. Confidence: high · Verified: WHATWG HTML: Character encoding declaration Evidence for this claim The complete character-encoding declaration must occur within the first 1024 bytes of the document. Scope: HTML serialization requirement intended to make encoding available early to parsers. Confidence: high · Verified: WHATWG HTML: Specifying the document's character encoding

Ce que the tag is

The charset declaration indique a parser qui character encoding to utiliser quand it turns the document’s bytes into text. The WHATWG HTML Living Standard puts it plainly: “The charset attribute specifies the character encoding utilisé by the document. Ce is a character encoding declaration.” MDN’s framing is the practical version: “Ce attribute declares the document’s character encoding.”

The modern syntax is the short formulaire:

<meta charset="utf-8">

There’s aussi a legacy pre-HTML5 formulaire you’ll encore voir in older templates:

<meta http-equiv="Content-Type" content="text/html; charset=utf-8">

Les deux declare the même chose. On a modern HTML5 document vous seulement besoin the short <meta charset="utf-8"> — en utilisant les deux is redundant, pas harmful, and the spec permet seulement un charset-declaring meta element per document anyway. (The http-equiv formulaire is meilleur thought of as legacy plutôt que something to ajouter fresh; si you’re auditing une page que has it, it isn’t broken, it’s simplement old.)

The spec requirements vous en réalité besoin to know

UTF-8 is effectively mandatory pour HTML5. MDN states it directement: the attribute’s “valeur doit be an ASCII case-insensitive match pour the string utf-8, parce que UTF-8 is the seulement valid encoding pour HTML5 documents.” The WHATWG spec goes plus loin and exige the document’s réel encoding to be UTF-8 regardless of what’s declared. UTF-8 covers essentially every script plus emoji, qui is pourquoi the ISO-8859-1 / Windows-1252 / Shift-JIS era of per-region encodings is over pour nouveau fonctionner — ceux survive seulement as legacy compatibility cas.

It doit land in the premier 1024 bytes of the document. Ce is a hard spec requirement, pas a soft suggestion. MDN: “<meta> elements qui declare a character encoding doit be located entirely dans the premier 1024 bytes of the document.” The raison is mechanical — the parser sniffs the byte stream pour an encoding avant it peut safely interpret the rest. Si votre declaration montre up aussi late, the parser may déjà have committed to a guessed encoding (or have to restart, qui costs performances). Remarque the framing carefully: it’s the premier 1024 bytes of the entier document, pas simplement of <head>.

Evidence for this claim The complete character-encoding declaration must occur within the first 1024 bytes of the document. Scope: HTML serialization requirement intended to make encoding available early to parsers. Confidence: high · Verified: WHATWG HTML: Specifying the document's character encoding

Meilleur pratique beats the spec minimum: faire it the premier child of <head>. Don’t settle pour “somewhere in the first 1024 bytes” — put <meta charset="utf-8"> avant votre <title>, <link>, <script>, <style>, and every autre tag. Ce is the placement modern tooling checks. There’s an ouvrir Lighthouse problème (#10023) proposing an audit que specifically checks si <meta charset> equals document.head.firstElementChild — i.e. flagging the tag quand it isn’t the literal premier element in head, pas merely quand it’s manquant. The direction of tooling is toward checking placement, pas simplement presence.

Seulement un charset meta element per document, and the charset attribute has aucun effect in XML/XHTML documents (it’s permitted là seulement to ease migration to and from XML). Worth a caveat si you’re working with XHTML-served content or RSS/Atom-adjacent templates.

BOM, HTTP header, meta tag — the precedence order

Navigateurs don’t simplement lire the meta tag in isolation; the encoding-sniffing algorithm checks three sources in a fixed order, and the premier un que donne an réponse wins:

  1. A UTF-8 byte-order mark (BOM) — a few bytes at the very commencer of the fichier. Si le navigateur detects a BOM, que determines the encoding with certainty; nothing sinon is consulted.
  2. The HTTP Content-Type header’s charset, si le serveur sends un and there’s aucun BOM. Ce takes precedence over the in-document meta declaration.
  3. The in-document <meta charset> (or legacy http-equiv) declaration, vérifié seulement si neither of the ci-dessus supplied an encoding.
Evidence for this claim A UTF-8 BOM takes precedence over HTTP and in-document declarations; otherwise an HTTP charset has higher precedence than meta, so server and document declarations must agree. Scope: HTML documents, HTTP delivery and rendered metadata as applicable Confidence: high · Verified: Declaring character encodings in HTML

En pratique, BOMs are rare on hand-authored HTML (they’re plus courant as an artifact of certain text editors or file-export outils), so the header-vs-tag conflict is the un que bites la plupart souvent: une page que correctement declares <meta charset="utf-8"> peut encore render garbled si a CDN, reverse proxy, or misconfigured server sends a différent charset in the header. It’s a classic symptom correct après a server or CDN migration — the HTML didn’t modifier, but the header did, and now the header is fighting the tag. Quand you’re debugging mojibake, vérifier pour a BOM and la réponse header charset, pas simplement lune page source.

Is meta charset an SEO ranking factor?

Aucun — and it’s worth being blunt parce que fear-based audit-tool copy parfois implies sinon. Ce is a rendering- and indexation-correctness prerequisite, pas a ranking signal.

Google’s guidance ici is thin and indirect comparé to the tags it discusses constantly (title, meta description, robots, canonical). Là is aucun dedicated Google Search Central page à propos de character encoding — it’s un entry à l’intérieur the general meta tags Google supports référence, sous “Content-Type and charset.” Google’s seulement on-record line is a recommendation, pas a ranking claim: “We recommend en utilisant Unicode/UTF-8 où possible.” Aucun verbatim statement from Mueller, Illyes, Splitt, or Canel specifically naming “meta charset” or “mojibake” surfaces in the trade press or Search Off the Record archives — charset is treated as basic web-standards hygiene, table stakes comme valid markup, plutôt que a topic warranting SEO commentary.

Bing has aucun distinct public position on the on-page tag soit; its documentation références UTF-8 seulement pour its propre API/feed formats (IndexNow clé fichiers, Webmaster API requête headers), pas as guidance à propos de the HTML <meta charset> on votre pages. Since Bingbot is a standard HTML parser, the practical implication is the même: follow the HTML spec’s UTF-8 / 1024-byte rule.

So où peut it hurt vous? Indirectly, and seulement quand the encoding is genuinely broken: garbled text is a content-quality and UX problem, it peut corrupt ce que apparaît in snippets, and severely broken output peut regarder broken to Google’s indexation systems aussi. The industry consensus, as Ahrefs’ propre meta-tags guide (by Joshua Hardwick) puts it, is que “unless votre page is severely broken as a result of charset problèmes (qui is unlikely), the impact is going to be quite minimal.” Fix it parce que broken text is bad, pas parce que vous expect a ranking bump.

Comment vérifier and fix it

A rapide diagnostic chemin quand vous suspect an encoding problem:

  • View source / DevTools. Confirmer <meta charset="utf-8"> exists and is the premier child of <head>. In DevTools, vérifier the Content-Type réponse header pour a charset valeur — si it disagrees with the tag, the header wins and is votre probable culprit. Rule out a BOM aussi: it’s rarer, but si présent it beats les deux the header and the tag.
  • Validators and robots d’exploration flag it. The W3C validator and rule-based checkers (e.g. Rocket Validator’s “charset after the first 1024 bytes” rule) va appel out a late or manquant declaration. Site audits in Ahrefs Site Audit and Screaming Frog surface charset problèmes à travers a whole site.
  • Fix the correct couche. Si the placement is incorrect, déplacer the tag to the top of head. Si the encoding is incorrect (the bytes themselves aren’t UTF-8, or the header sends a conflicting charset), fixing the meta tag alone won’t aider — vous have to re-encode the fichier as UTF-8 and/or correct le serveur’s Content-Type header so the header and tag agree.

The fix is almost toujours trivial une fois you’ve identified qui couche is at fault. Ce is a stable, long-settled partie of the HTML spec — there’s aucun recent deprecation or platform-behavior modifier to track; the seulement evolving nuance is tooling increasingly checking placement, pas simplement presence.

Où ce sits

The charset tag is un of the browser-facing head elements — comme the viewport tag, it’s à propos de rendering, pas ranking, qui puts it in a différent bucket from the SEO-active tags in the meta-tags cluster (the title element, the meta description, and the robots family). It’s adjacent to the internationalization fonctionner I spend a lot of temps on: encoding is the couche underneath hreflang and multi-script content — hreflang indique Google qui language/region version to serve, but si the encoding is incorrect the text in que version is garbled regardless. Pour the complet map of head elements grouped by the job ils do, voir the meta tags hub.

Add an expert note

Pin an expert quote

New person? Create their unclaimed profile at /admin/experts/ → Pin a quote first.