SEO technique Checklist

TL;DR — A SEO technique checklist is the liste of technical boxes votre site has to tick so moteur de recherches peut trouver, lire, and montrer votre pages. Google’s short version: une page has to be crawlable, indexable, understandable, and renderable. Fonctionner the liste in priority order, and skip the sections que don’t appliquer to votre size of site — a 20-page brochure site ne fait pas besoin to worry à propos de the même choses a 500 000-product store fait.

Evidence for this claim Google's minimum technical requirements include accessible Googlebot crawling, a successful HTTP response, and indexable content. Scope: Eligibility prerequisites, not a guarantee of indexing or ranking. Confidence: high · Verified: 2026-07-14 Google Search Essentials: Technical requirements Evidence for this claim Meeting technical requirements does not guarantee that Google will crawl, index, or serve a page. Scope: Google Search eligibility and selection behavior. Confidence: high · Verified: 2026-07-14 Google Search Essentials: Technical requirements

Ce que ce checklist is (and isn’t)

SEO technique is the plumbing. Avant votre content peut rank, moteur de recherches have to be able to reach votre pages, download les, comprendre les, and fichier les away. A SEO technique checklist is simplement the ordered liste of choses que faire que possible.

It is pas the whole of SEO. It doesn’t cover writing bon content, targeting the correct keywords, or earning liens. Ceux matter enormously — they’re simplement a différent job. Si vous vouloir the everything-at-once version, that’s a complet SEO audit checklist, and it’s a separate, broader chose. Ce page stays technical.

The priority order

Do ces roughly top to bottom. The plus élevé up the liste, the plus it peut quietly break everything ci-dessous it.

1. Peut moteur de recherches explorer votre site? Vérifier votre robots.txt isn’t blocking pages vous vouloir trouvé. Fix broken pages (404s) and server errors (5xx).
2. Peut votre pages be indexé? Assurez-vous you’re pas accidentally telling Google to noindex pages importantes, and que duplicate versions of une page point to un preferred version.
3. Is votre site on HTTPS? The little padlock. It’s a baseline expectation now.
4. Fait it fonctionner bien on phones? Google indexes the mobile version of votre site, so votre phone version has to have the même content as votre desktop version.
5. Do vous have an XML sitemap? A simple liste of votre URLs vous hand to Google and Bing so ils don’t have to trouver everything by suivant liens.
6. Is votre données structurées valid? Optional, but it peut earn vous richer-looking résultats de recherche. (It won’t boost votre rankings — plus on que myth ci-dessous.)
7. How’s votre vitesse de page? Google’s Core Web Vitals mesurer chargement, réponse, and visual stability.

The two myths to know

• Blocking une page in robots.txt fait pas supprimer it from Google. It simplement arrête Google reading it. To en réalité supprimer une page, let Google explorer it and ajouter a noindex tag. (Blocking it in robots.txt and ajout noindex is a classic trap — the bot can’t explorer lune page, so it jamais sees the noindex.)
• Données structurées n’est pas a ranking boost. It peut faire votre result regarder richer, qui peut aider clicks — but ajout schema doesn’t déplacer vous up le résultats.

Pas everything s’applique to vous

The unique la plupart utile chose à propos de ce checklist is knowing ce que to skip. Si vous run a petit site, vous pouvez ignore crawl-budget management entirely — Google itself dit la plupart sites don’t besoin to think à propos de it. Vouloir the complet picture, notamment who fait besoin ceux avancé sections? Switch to the Avancé tab.

TL;DR — SEO technique is the floor, pas the ceiling: pages doit be crawlable → indexable → understandable → renderable avant content fonctionner peut pay off. Run the checklist in priority order and gate sections by site type — la plupart petit sites peut skip budget d’exploration, faceted-nav contrôler, log analysis, and JS-rendering engineering, qui Google’s propre crawl-budget doc confirms. Kill two myths on sight: robots.txt doesn’t deindex, and données structurées isn’t a ranking factor. Treat Core Web Vitals as targets, pas réussir/échouer gates — that’s Google’s propre “strive to” language. And meeting every box encore doesn’t guarantee indexation.

Evidence for this claim Google's minimum technical requirements include accessible Googlebot crawling, a successful HTTP response, and indexable content. Scope: Eligibility prerequisites, not a guarantee of indexing or ranking. Confidence: high · Verified: 2026-07-14 Google Search Essentials: Technical requirements Evidence for this claim Meeting technical requirements does not guarantee that Google will crawl, index, or serve a page. Scope: Google Search eligibility and selection behavior. Confidence: high · Verified: 2026-07-14 Google Search Essentials: Technical requirements

The mental model: four conditions

Google frames the whole of SEO technique autour four load-bearing conditions — une page has to be crawlable, indexable, understandable, and renderable. Its Search Essentials technical requirements boil que bas to three minimums: “Googlebot isn’t blocked,” “The page works” (served with an HTTP 200 status), and “The page has indexable content.” Everything on ce checklist is really in service of ceux.

The critical caveat, straight from the même doc: “Simplement parce que une page meets ces requirements doesn’t mean que une page va be indexé; indexation isn’t guaranteed.” SEO technique is a gate vous have to réussir via, pas a lever que guarantees results. Ce is pourquoi I’ve toujours argued the technical checklist is the floor — vous clair it so content and liens peut do leur job, pas au lieu de les.

Avant vous commencer: qui sections même appliquer to vous?

Every competing checklist organizes by topic and runs the whole chose on every site. That’s the incorrect par défaut. The meilleur organizing axis is site type, parce que Google itself gates its la plupart avancé guidance by site size. From the large-site crawl-budget guide: “Si votre site doesn’t have a grand number of pages que modifier rapidly, or si votre pages sembler to be crawled the même day que ils are publié, vous don’t besoin to lire ce guide.”

Google’s propre (deliberately rough) thresholds pour quand budget d’exploration starts to matter:

• Grand sites — 1 million+ unique pages with content modification à propos de weekly.
• Medium-or-larger sites — 10 000+ unique pages with very rapidly modification (daily) content.
• Sites with a big share of URLs stuck in Search Console’s “Découvert - currently non indexée” state.

So here’s the split I’d en réalité run:

Starter track — petit / brochure / local-business sites (< ~10K URLs): crawlability sanity vérifier, indexation/canonical sanity vérifier, HTTPS, mobile parity, un XML sitemap, un round of Core Web Vitals, valid données structurées où it earns a rich result. Skip budget d’exploration, faceted-nav contrôler, log-file analysis, and JS-rendering engineering entirely.

Avancé track — grand / ecommerce / JS-heavy / enterprise sites: everything in the Starter track plus crawl-budget management, faceted-navigation contrôler, JavaScript rendering audits, server-log analysis, and (si international) hreflang.

1. Crawlability

• robots.txt correctness. Confirmer vous aren’t disallowing anything vous vouloir indexé. Google: “A robots.txt fichier indique moteur de recherche robots d’exploration qui URLs the robot d’exploration peut accès on votre site. Ce is utilisé mainly to éviter overloading votre site with requêtes; it n’est pas a mechanism pour keeping a web page out of Google.” Utiliser it to garder bots out of low-value spaces (internal search, infinite parameter combinations), pas as a deindexing outil. Ce is the même territory the robots.txt deep dive covers in complet.
• The robots.txt + noindex contradiction trap. Do pas block une URL in robots.txt and rely on a noindex on it. Google: “Pendant que Google won’t explorer or index le contenu blocked by a robots.txt fichier, we pourrait encore trouver and index a disallowed URL si it is lié from autre places on the web.” The bot can’t explorer lune page, so it jamais sees the noindex, and l’URL peut encore surface bare in results. To supprimer une page: autoriser exploration + noindex, or password-protect it.
• Explorer errors. Fix unexpected 4xx and 5xx. Google seulement indexes pages served with a 200, and “Client and server error pages aren’t indexed.”
• Chaîne de redirectionss and loops. Collapse A→B→C→D bas to A→D. Chains waste explorer and leak a little on every hop. The exploration and redirections material goes deeper ici.

2. Indexability

• Index coverage. In Search Console’s Page Indexation report, reconcile ce que vous vouloir indexé contre ce que en réalité is. Investigate grand “Découvert/Crawled - currently non indexée” buckets.
• Canonicalization. Point duplicate and near-duplicate URLs at un preferred version. Google calls rel="canonical" “a strong signal que the specified URL devrait become canonical” — a signal, pas a directive it doit obey. And crucially: “Don’t use the robots.txt file for canonicalization purposes.” Vérifier vous aren’t sending conflicting canonical signals à travers HTML tag, HTTP header, and sitemap — the canonicalization deep dive walks via consolidating les.
• Contenu dupliqué. Parameters, print versions, staging leaks, http/https and www/non-www splits tout créer duplicates. Pick un, canonicalize or redirection the rest.

3. HTTPS

Baseline, pas optional. Serve the whole site over HTTPS, redirection http to https, and hunt bas mixed content (a secure page chargement an insecure image, script, or stylesheet). Chris Green’s SEO in 2026 reality-check pegs HTTPS adoption at “91%+” — vous don’t vouloir to be in the trailing 9%.

4. Mobile SEO

Google uses indexation mobile-first: “Google uses the mobile version of a site’s content, crawled with the smartphone agent, pour indexation and ranking.” The parity checklist, straight from Google’s mobile-first doc:

• “Make sure that your mobile site contains the same content as your desktop site.”
• “Assurez-vous que the title element and the meta description are equivalent à travers les deux versions of votre site.”
• “Make sure that your mobile and desktop sites have the same structured data.”
• “Use the same robots meta tags on the mobile and desktop site.”
• “Don’t lazy-load primary content upon user interaction.”
• “Assurez-vous que the mobile site has the même texte alternatif pour images as the desktop site.”

The la plupart courant échec is a stripped-down mobile template que quietly drops content, liens, or données structurées présent on desktop — Google indexes the thinner version.

5. Sitemaps and discovery

• XML sitemap hygiene. Utiliser absolute, URL canoniques; stay sous the 50 MB / 50 000-URL per-file limite; liste seulement indexable, URL canoniques; garder lastmod accurate. Référence it in robots.txt (Sitemap: https://example.com/sitemap.xml) so engines découvrir it automatically. Complet treatment in the XML sitemaps material.
• Bing encore cares. From Bing’s July 2025 guidance: “Sitemaps remain a foundational signal pour ensuring comprehensive URL coverage à travers votre site,” “XML remains the preferred format pour sitemaps,” and “The lastmod field in votre sitemap remains a clé signal, helping Bing prioritize URLs pour recrawling.”
• IndexNow. La plupart Google-centric checklists omit it, but it’s a live, free, one-line win: Bing’s advice is to “Utiliser IndexNow pour real-time URL submission, instantly notifying Bing and participating moteur de recherches” quand content changements. It complements sitemaps plutôt que replacing les. (Remarque: Google ne fait pas utiliser IndexNow pour general pages.)

6. Données structurées

• Ce que it fait: earns rich-result eligibility and helps machines (and LLMs) comprendre votre page. “Ajout données structurées peut enable résultats de recherche que are plus engaging to utilisateurs… qui are appelé résultats enrichis.”
• Ce que it fait pas do: boost rankings. Google’s docs frame schema strictly as rich-result eligibility and machine understanding — pas a ranking signal. Ne faites pas sell it, or budget pour it, as a ranking play.
• Format: “In general, Google recommends en utilisant JSON-LD pour données structurées si votre site’s setup permet it, as it’s the easiest solution pour website owners to implement and maintain at scale.” Validate with the Résultats enrichis Tester. The données structurées material covers the spécifique types worth implementing.

7. Core Web Vitals and page experience

Targets, pas gates — Google’s réel wording is “strive to,” qui la plupart checklists overstate as hard réussir/échouer cutoffs:

• LCP — “strive to have LCP occur dans the premier 2,5 seconds of lune page starting to charger.”
• INP — “strive to have an INP of less than 200 milliseconds.”
• CLS — “strive to have a CLS score of less than 0.1.”

And the relationship to ranking, qui personnes badly over-weight: “Recherche Google toujours seeks to montrer the la plupart relevant content, même si lune page experience is sub-par.” Bon Core Web Vitals are a tiebreaker among relevant results, pas an override of relevance. Mesurer with real-user (CrUX/field) données, pas simplement lab scores.

8. Avancé additions (grand / ecommerce / JS-heavy seulement)

• Budget d’exploration. Seulement si vous cleared Google’s thresholds ci-dessus. Capacity + demand; vous gain budget by removing waste (parameter explosions, faceted-nav combinations, spider traps, duplicate URLs) far plus que by trying to faire Google explorer “more.” Voir the budget d’exploration deep dive.
• JavaScript rendering. Confirmer critical content and liens exist in the rendered HTML and are reachable via réel <a href> liens, pas click-only navigation. Ce is JavaScript SEO territory.
• Faceted navigation contrôler. Decide qui filter/sort combinations are crawlable/indexable and contrôler the rest.
• Log-file analysis. The ground truth pour ce que bots en réalité récupérer, how souvent, and ce que code d’états ils hit.
• hreflang — seulement si you’re genuinely multi-regional/multilingual. That’s a big suffisant topic to live in its propre international-SEO material; don’t bolt it on half-done.

9. AI / LLM robot d’exploration accès (short, scoped)

Two table-stakes items in 2026, and aucun plus — the deep GEO/AEO fonctionner lives in the AI-search material, pas ici:

• Decide AI-crawler accès in robots.txt. Explicitly autoriser or disallow the AI user-agents vous care à propos de (training vs. AI-search vs. user-triggered fetchers are différent bots). Chris Green’s framing: “Robots.txt is ne … plus simplement explorer housekeeping. It’s becoming a policy surface.”
• Données structurées doubles as machine context pour LLMs — a bonus raison to obtenir votre schema valid, pas a nouveau workstream.

10. How to prioritize ce que vous trouver

Ce is où la plupart checklists échouer: ils hand vous 90 items with aucun weighting. Don’t fix everything — fix ce que moves the needle. My buddy Patrick’s advice on client audits, qui I garder coming back to: “Si clients are coming to vous asking pour an audit, ils déjà have a pain point. Talk to les. Solve que un chose and they’ll be happy with the audit.” The même SEO Audit Template frames it as “sweating the small stuff rarely does much for your rankings” — meilleur to spend “80% of your time fixing the 20% of things that matter.”

And zoom out on the whole exercise: a checklist obtient vous to okay. Google’s John Mueller has repeatedly made the point que fundamentals alone obtenir vous fine-but-not-great results — réel dominance comes from topical depth and authority, pas from ticking every technical box. The checklist clears the floor; content and liens construire the house.

Vouloir the full-site version?

Ce page is technical-only by design. Si vous vouloir the broader audit — technical plus on-page, content, and off-page — that’s the SEO Audit Checklist, a separate, wider chose. Don’t essayer to faire ce un page do les deux jobs.

AI summary

A condensed prendre on the Avancé version:

• SEO technique = the floor. Google’s four conditions: crawlable, indexable, understandable, renderable. Meeting les is necessary but pas sufficient — “indexation isn’t guaranteed.”
• Gate by site type, pas simplement topic. La plupart petit sites peut skip budget d’exploration, faceted-nav contrôler, log analysis, and JS-rendering engineering. Google’s propre thresholds: 1M+ pages modification weekly, or 10K+ modification daily, or lots of “Discovered - currently not indexed.”
• Priority order: crawlability (robots.txt, 4xx/5xx, chaîne de redirectionss) → indexation (coverage, canonicalization, duplicates) → HTTPS + mixed content → mobile-first parity → XML sitemap + IndexNow → données structurées → Core Web Vitals.
• Two myths to kill: robots.txt fait pas deindex (blocked-but-linked URLs peut encore apparaître bare); données structurées is pas a ranking factor (rich-result eligibility seulement).
• CWV are targets, pas gates — Google’s “strive to” language: LCP < 2,5s, INP < 200ms, CLS < 0,1. And “Recherche Google toujours seeks to montrer the la plupart relevant content, même si lune page experience is sub-par.”
• Mobile-first parity: même content, titles/meta, données structurées, robots meta tags, and texte alternatif à travers mobile and desktop; don’t lazy-load principal content on interaction.
• Prioritize by impact (solve the client’s réel pain point; 80/20). A checklist obtient vous “okay”; topical depth wins.
• Full-site version (technical + content + liens) = the separate SEO Audit Checklist.

Documentation officielle

Primary-source documentation from the moteur de recherches.

Google

• Recherche Google technical requirements — the crawlable/fonctionne/indexable minimums, and the “indexing isn’t guaranteed” caveat.
• Introduction to robots.txt — ce que robots.txt fait and doesn’t do.
• Construire and submit a sitemap — formats, the 50MB/50 000-URL limite, and referencing it in robots.txt.
• Consolidate duplicate URLs — canonicalization signals and “don’t use robots.txt for canonicalization.”
• Intro to données structurées markup — résultats enrichis, and the JSON-LD recommendation.
• Understanding Core Web Vitals — the “strive to” LCP/INP/CLS targets.
• Page experience dans la recherche Google results — relevance vs. page experience.
• Indexation mobile-first meilleur practices — the mobile/desktop parity requirements.
• Optimize votre budget d’exploration — who en réalité nécessite crawl-budget management, with thresholds (formerly titled “Large site owner’s guide to managing crawl budget”; the doc déplacé sous Google’s Exploration Infrastructure docs).

Bing / Microsoft

• Keeping Content Discoverable with Sitemaps in AI-Powered Search — Bing’s July 2025 sitemap + IndexNow guidance.
• Bing Webmaster Guidelines — exploration, indexation, ranking, and quality (cite by référence; lune page is JS-rendered).

Quotes from the source

On-the-record statements from Google and Bing. Où a lien is a deep lien, it jumps to the quoted passage on the source page.

Google — the minimum requirements

• “Client and server error pages aren’t indexed.” — Recherche Google Central, Recherche Google technical requirements. Jump to quote
• “Just because a page meets these requirements doesn’t mean that a page will be indexed; indexing isn’t guaranteed.” Jump to quote

Google — robots.txt and canonicalization

• “A robots.txt file tells search engine crawlers which URLs the crawler can access on your site. This is used mainly to avoid overloading your site with requests; it is not a mechanism for keeping a web page out of Google.” Jump to quote
• “rel=“canonical” link annotations are a strong signal that the specified URL should become canonical… Don’t use the robots.txt file for canonicalization purposes.” — Google, Consolidate duplicate URLs. Lire the doc

Google — données structurées and Core Web Vitals

• “Adding structured data can enable search results that are more engaging to users… which are called rich results.” — Google, Intro to données structurées markup. Lire the doc
• “In general, Google recommends using JSON-LD for structured data if your site’s setup allows it, as it’s the easiest solution for website owners to implement and maintain at scale.” Lire the doc
• “strive to have LCP occur within the first 2.5 seconds”; “strive to have an INP of less than 200 milliseconds”; “strive to have a CLS score of less than 0.1.” — Google, Understanding Core Web Vitals. Lire the doc
• “Google Search always seeks to show the most relevant content, even if the page experience is sub-par.” — Google, Page experience. Lire the doc

Google — indexation mobile-first

• “Google uses the mobile version of a site’s content, crawled with the smartphone agent, for indexing and ranking.” — Google, Indexation mobile-first meilleur practices. Lire the doc

Google — budget d’exploration (who nécessite it)

• “If your site doesn’t have a large number of pages that change rapidly, or if your pages seem to be crawled the same day that they are published, you don’t need to read this guide.” — Google, Optimize votre budget d’exploration. Jump to quote

Bing — sitemaps and IndexNow (July 2025)

• “XML remains the preferred format for sitemaps.” / “The lastmod field in your sitemap remains a key signal, helping Bing prioritize URLs for recrawling.” — Fabrice Canel & Krishna Madhavan, Bing Webmaster Blog. Lire the post

Patrick Stox — how to prioritize an audit

• “If clients are coming to you asking for an audit, they already have a pain point. Talk to them. Solve that one thing and they’ll be happy with the audit.” — Patrick Stox, quoted in Ahrefs’ Free SEO Audit Template. Jump to quote

Qui checklist devrait I en réalité run?

Don’t run the même 90-item liste on every site. Commencer ici.

Q1. How nombreux URLs fait votre site have, and how fast fait content modifier?

• Sous ~10 000 URLs, modification occasionally → run the Starter track and arrêter: crawlability sanity vérifier, indexation/canonical sanity vérifier, HTTPS, mobile parity, un XML sitemap, un round of Core Web Vitals, valid données structurées où it earns a rich result. Skip budget d’exploration, faceted-nav contrôler, log analysis, and JS-rendering engineering. Go to Q3.
• 10 000+ URLs modification daily, or 1M+ modification weekly, or lots of “Découvert - currently non indexée” → run the Avancé track (Starter + budget d’exploration + faceted nav + JS rendering + logs). Go to Q2.

Q2. Is le site JavaScript-heavy or international?

• JS-heavy (SPA, client-rendered content/liens) → ajouter a rendering audit: confirmer critical content and <a href> liens exist in the rendered HTML. Ce is the item la plupart probable to be silently costing vous indexation.
• Genuinely multi-region / multi-language → ajouter hreflang, fait correctement, in its propre workstream. Si you’re pas truly international, skip it.
• Neither → proceed to Q3.

Q3. Ce que did the audit surface — and ce que devrait vous fix premier?

• A explorer/index blocker (robots.txt disallow on pages importantes, mass noindex, incorrect canonical, site-wide 5xx) → fix premier, toujours. Ces gate everything ci-dessous.
• Lots of medium problèmes, limited temps → appliquer the 80/20: fix the ~20% que moves the needle, and si it’s a client, fix the spécifique pain point ils came to vous with premier. Don’t hand over 90 undifferentiated line items.
• Seulement cosmetic/edge problèmes left → you’ve cleared the floor. Arrêter optimizing plumbing and go do content and liens.

Q4. Do vous en réalité vouloir the full-site version?

• Yes — technical + content + on-page + liens → ce technical checklist isn’t it. Run the broader SEO Audit Checklist à la place; ce page is technical-only by design.
• Aucun — simplement the technical floor → you’re on the correct page.

One-line version: petit + stable → Starter track seulement; big/JS/international → ajouter the avancé sections; alors fix explorer/index blockers premier and prioritize the rest by impact.

The SEO technique checklist

Two tracks. Run the Starter track on quelconque site; ajouter the Avancé items seulement si vous cleared le site-type gate (10K+ URLs modification daily, 1M+ weekly, or JS-heavy/international).

Starter track — every site

Crawlability

• robots.txt doesn’t disallow anything vous vouloir indexé (and fait block low-value spaces comme internal search).
• Aucun important URL is blocked in robots.txt and relying on a noindex (the trap).
• Unexpected 4xx/5xx fixed; pages importantes retourner 200.
• Chaîne de redirectionss/loops collapsed to a unique hop.

Indexability

• Page Indexation report reconciled — ce que vous vouloir indexé en réalité is.
• Canonicals point to un preferred version; aucun conflicting signals à travers HTML tag, HTTP header, and sitemap.
• http→https and www/non-www consolidated to un version.

HTTPS

• Whole site on HTTPS; http redirections to https.
• Aucun mixed content (secure page chargement insecure assets).

Mobile

• Mobile version has the même content, titles/meta, données structurées, robots meta tags, and texte alternatif as desktop.
• Principal content isn’t lazy-loaded on utilisateur interaction.

Sitemaps

• XML sitemap listes seulement canonical, indexable URLs; accurate lastmod; sous 50MB/50 000 URLs per fichier.
• Sitemap submitted in Search Console and Bing Webmaster Outils, and referenced in robots.txt.

Données structurées

• Schema (JSON-LD) valid in the Résultats enrichis Tester, and seulement utilisé où it earns a rich result.

Core Web Vitals

• LCP, INP, CLS reviewed on field/CrUX données contre Google’s targets (2,5s / 200ms / 0,1) — treated as targets, pas réussir/échouer gates.

Avancé track — grand / ecommerce / JS-heavy / international seulement

• Budget d’exploration reviewed (seulement si past Google’s thresholds); waste supprimé (parameters, facets, traps, duplicates).
• JS rendering audited — critical content and <a href> liens présent in rendered HTML.
• Faceted navigation: decided qui combinations are crawlable/indexable.
• Server logs analyzed pour explorer waste and uncrawled important URLs.
• IndexNow wired up pour real-time modifier submission to Bing and participating engines.
• hreflang correct and reciprocal (seulement si genuinely multi-region/multilingual).
• AI-crawler accès explicitly decided in robots.txt.

The mental models

1. The four conditions — crawlable → indexable → understandable → renderable. Every technical item sert un of ces. Quand une page isn’t performing, trouver qui condition it’s failing avant vous touch anything: is it même crawlable? Indexable? Understood? Rendered?

2. Floor, pas ceiling. SEO technique clears the floor so content and liens peut do leur fonctionner. “Indexation isn’t guaranteed” even when you pass — so don’t treat a green technical audit as “SEO fait.”

3. Gate by site type. The unique la plupart utile déplacer: decide up front qui sections don’t appliquer. Petit/stable site → Starter track, skip the avancé plumbing. Grand/JS/international → ajouter it. Google gates its propre crawl-budget guidance ce façon; vous devez aussi.

4. Signals vs. directives. Know qui contrôle Google doit obey and qui are merely strong signals. noindex is a directive; rel=canonical is “a strong signal” Google peut override; CWV targets are “strive to” goals, pas gates. Mislabeling a signal as a guarantee is où la plupart bad advice comes from.

5. Impact over completeness (80/20). A 90-item checklist is a menu, pas a to-do liste. Fix the ~20% que moves the needle, solve the réel pain point premier, and arrêter optimizing plumbing une fois the floor is clair.

6. Ce is the technical slice. Explorer/index/render/serve mechanics live ici; content quality, keyword targeting, and liens live in the broader SEO Audit Checklist. Garder the boundary clean so neither job obtient half-done.

SEO technique — quick-reference

Ce que chaque contrôler fait

Core Web Vitals targets (Google’s “strive to” numbers)

Crawl-budget “do I care?” thresholds (Google’s propre, deliberately rough)

• ~1M+ pages modification à propos de weekly → yes.
• 10K+ pages modification daily → yes.
• Big “Discovered - currently not indexed” bucket → yes.
• Sinon → “you don’t need to read this guide.”

Fast facts

• Sitemap limite: 50 MB / 50 000 URLs per fichier; XML preferred; référence it in robots.txt.
• Données structurées: résultats enrichis, pas rankings. JSON-LD recommended.
• IndexNow: Bing/Yandex/others — pas Google. Google indexes the mobile version.
• Mobile parity: même content, titles/meta, données structurées, robots meta tags, texte alternatif.

Mythes et erreurs à éviter

The traps que come up la plupart — several are widely-repeated myths worth correcting:

• “Blocking a URL in robots.txt keeps it out of Google.” Aucun. Google: “it n’est pas a mechanism pour keeping a web page out of Google.” A disallowed URL peut encore be indexé (bare, aucun snippet) si it’s lié externally. Utiliser noindex or password protection.
• “Block it in robots.txt and noindex it for extra safety.” The unique la plupart courant self-inflicted wound. Si l’URL is blocked, Googlebot jamais crawls it, so it jamais sees the noindex — and l’URL peut encore surface bare. Pick un: autoriser explorer + noindex, or disallow (accepting it may encore apparaître).
• “Adding schema boosts rankings.” Aucun. Google frames données structurées purely as rich-result eligibility and machine understanding — pas a ranking signal. Sell it as richer results, pas plus élevé positions.
• “Passing Core Web Vitals beats a more relevant competitor.” Aucun. “Recherche Google toujours seeks to montrer the la plupart relevant content, même si lune page experience is sub-par.” CWV is a tiebreaker-ish signal, pas an override of relevance.
• “CWV thresholds are hard pass/fail gates.” Google’s propre word is “strive to” — they’re targets. Chasing a perfect lab score at the expense of everything sinon is misspent effort.
• “Crawl budget matters for every site.” Aucun. Google gates its propre guide to 1M+ pages (weekly changements) or 10K+ (daily). La plupart sites are told outright “vous don’t besoin to lire ce guide.” Don’t burn a small-site engagement on crawl-budget theater.
• “rel=canonical is a directive Google must obey.” Aucun — it’s “a strong signal.” Google peut and fait pick a différent canonical quand autre signals conflict. Reduce conflicting signals plutôt que assuming the tag wins.
• “A green technical audit means SEO is done.” Aucun. “Indexing isn’t guaranteed” même quand vous réussir. SEO technique is the floor; content and liens encore have to montrer up.
• “Run the same 90-item list on every site.” The differentiator is knowing ce que to skip. Gate sections by site type, and prioritize by impact, or you’ll drown clients in irrelevant line items.

SOP: run a SEO technique checklist (recurring)

A repeatable réussir. Roughly 60–90 minutes pour a petit site; a day-plus pour grand/JS-heavy. Run it quarterly, or après quelconque big migration, redesign, or CMS modifier.

1. Confirmer le site type and pick a track. Count URLs (Search Console → Pages, or a explorer) and remarque how fast content changements. Sous ~10K and stable → Starter track seulement. Over the thresholds, or JS-heavy/international → ajouter the Avancé sections. Ce decides ce que vous skip.
2. Explorer le site with Screaming Frog, Ahrefs Site Audit, or similaire. Capture status codes, chaîne de redirectionss, blocked URLs, canonicals, indexability, and depth.
3. Vérifier robots.txt premier. Récupérer /robots.txt, confirmer nothing important is disallowed, and vérifier pour the disallow-plus-noindex trap. Confirmer the Sitemap: line is présent.
4. Reconcile indexation. In Search Console’s Page Indexation report, comparer what’s indexé to ce que vous vouloir indexé. Investigate grand “Discovered/Crawled - currently not indexed” and “Duplicate” buckets.
5. Vérifier HTTPS and mobile parity. Confirmer site-wide HTTPS with aucun mixed content. Alors comparer the mobile-rendered page to desktop pour content, titles/meta, données structurées, robots meta tags, and texte alternatif (Google indexes the mobile version).
6. Validate sitemaps and données structurées. Confirmer the XML sitemap listes seulement canonical, indexable URLs with accurate lastmod; run clé templates via the Résultats enrichis Tester.
7. Pull Core Web Vitals from field données. Utiliser the CrUX/field report in Search Console or PageSpeed Insights — pas simplement lab scores — and comparer contre the 2,5s / 200ms / 0,1 targets.
8. (Avancé track) Analyze logs and budget d’exploration. Lire server logs pour explorer waste and uncrawled important URLs; audit JS rendering and faceted-nav combinations.
9. Prioritize the findings by impact, pas count. Put explorer/index blockers at the top; appliquer 80/20 to the rest; si it’s a client, lead with leur réel pain point.
10. Log a baseline and re-check suivant cycle. Record the state so suivant quarter’s réussir measures progress, pas a fresh commencer.

Ready-to-use AI prompts

Copy-paste starting points pour triaging a SEO technique checklist with an LLM. Toujours vérifier LLM output contre principal docs and votre propre explorer données — treat ces as drafting aids, pas sources of truth.

Pick the correct track pour a site

I run a website with à propos de [N] URLs, and content changements roughly [how souvent]. It’s construit on [CMS/framework], is [single-language / multi-region], and is [static / JavaScript-rendered]. Fondé on Google’s propre site-size thresholds pour budget d’exploration, tell me qui sections of a SEO technique checklist genuinely appliquer to me and qui I peut safely skip. Be explicit à propos de ce que to skip and pourquoi.

Triage a explorer export by impact

Ici is a CSV export from a site explorer [paste columns: URL, code d’état, indexability, canonical, chaîne de redirections, robots.txt status]. Groupe the problèmes into (1) explorer/index blockers to fix premier, (2) medium-impact fixes, (3) cosmetic/low-impact. Pour chaque groupe, expliquer the SEO consequence in un sentence. Ne faites pas tell me to “fix everything” — rank by impact.

Expliquer a spécifique indexation status

A batch of my URLs montre “[exact Search Console status, e.g. Découvert - currently pas indexé]”. Expliquer the probable causes in priority order, how to diagnose chaque, and the concrete fix. Flag anything que is a symptom of budget d’exploration vs. a per-page problem.

Sanity-check pour the robots.txt + noindex trap

Ici is my robots.txt [paste] and a liste of URLs I’m trying to garder out of Google [paste]. Pour chaque URL, tell me si my current setup va en réalité garder it out of the index or fall into the “blocked in robots.txt but still indexable via links” trap, and give the correct fix (allow-crawl + noindex, or password protection).

Draft a prioritized remediation plan

Turn ces confirmed technical findings [paste] into a prioritized remediation plan pour a developer: blockers premier, alors impact-ranked, chaque with the spécifique modifier to faire and a one-line rationale. The client’s stated pain point is [X] — surface fixes connexe to que premier.

Outils pour working the SEO technique checklist

Commencer with focused tests pour the échec vous are checking. A unique all-in-one score souvent hides the difference entre explorer accès, réponse behavior, index signals, rendering, and field performances.

Patrick’s free outils

• Google Index Checker checks observable status, redirection, noindex, and canonical blockers, alors points vous to Inspection d’URL pour Google’s réel indexé state.
• robots.txt Tester tests URLs contre bot rules and montre the winning autoriser or disallow rule. Utiliser it avant modification explorer contrôle.
• XML Sitemap Validator checks sitemap syntax, URL inventory, and fichier problems avant vous submit the fichier.
• Canonicalization Checker compares HTML and HTTP canonical signals and tests l’URL cible pour conflicts.
• Bulk Code d’état HTTP Checker checks code d’états, destinations, chains, and loops à travers une URL définir.
• Balisage de données structurées Validator validates données structurées syntax and Google rich-result requirements avant release.
• Render Gap compares initial HTML with rendered output pour JavaScript-dependent content, liens, canonicals, and robots directives.
• Core Web Vitals Checker separates disponible field données from a current performances vérifier so vous ne faites pas mistake un lab run pour utilisateur données.
• Mobile-Friendly Tester checks viewport, responsive layout, tap targets, and connexe mobile implementation signals.
• Log Fichier Analyzer turns server logs into evidence of ce que search bots en réalité récupéré, qui is la plupart utile on grand sites.

Search-engine outils

• Recherche Google Console provides Page Indexation, Inspection d’URL, Statistiques d’exploration, Sitemaps, Core Web Vitals, rich-result reports, manual actions, and security problèmes.
• Bing Webmaster Outils adds Bing’s first-party index and performances views, IndexNow, Site Explorer, Site Scan, and Explorer Contrôler.

Robots d’exploration and navigateur outils

• Ahrefs Site Audit or Screaming Frog SEO Spider is the scalable couche pour exploration templates, code d’états, directives, lien internes, and données structurées.
• Chrome DevTools exposes the network réponse, rendered DOM, console errors, and performances trace pour individual pages.
• PageSpeed Insights and CrUX provide Google’s lab and field performances views; utiliser field données pour the standing user-experience baseline quand it is disponible.

Testez vos connaissances: SEO technique Checklist

Five rapide questions on ce que belongs on a SEO technique checklist and how to prioritize it. Pick an réponse pour chaque, alors vérifier.

Ressources utiles

My connexe writing

• The Beginner’s Guide to SEO technique — the bigger picture ce checklist sits à l’intérieur.
• SEO Audit Template — my 80/20, impact-first framing, and the “solve their actual pain point” advice; aussi the closest chose to the broader full-site audit.
• The Story of Blocking 2 High-Ranking Pages With Robots.txt — my first-party experiment on ce que robots.txt blocking en réalité fait.
• Indexé, though blocked by robots.txt — the mechanism behind the disallow-plus-noindex trap.
• Robots.txt and SEO: Everything Vous devez Know.

My speaking

• How Search Fonctionne (SlideShare) — my walkthrough of exploration, rendering, indexation, and ranking, qui is the pipeline ce whole checklist is protecting.

My publié fonctionner worth citing

• Web Almanac 2021 — SEO chapter — I was lead author; the source pour cross-signal canonical-conflict rates à travers the réel web.
• Rankable Ep. 65 — Ranking SEO technique Priorities (iPullRank) — me on how to prioritize technical problèmes on a roadmap, qui is exactly ce checklist’s organizing principle.

From autour the industry

• Recherche Google technical requirements — the definitive statement of the minimums, and the “indexing isn’t guaranteed” caveat.
• Optimize votre budget d’exploration (Google) — the source pour le site-size gate (page renamed from “Large site owner’s guide to managing crawl budget” and déplacé sous Exploration Infrastructure).
• Keeping Content Discoverable with Sitemaps in AI-Powered Search (Bing) — le sitemap + IndexNow guidance la plupart Google-only checklists skip.
• SEO in 2026: Plus élevé standards, AI influence, and a web encore catching up (Chris Green, Moteur de recherche Land) — a reality-check on où sites en réalité stand (HTTPS 91%+, robots.txt as a “policy surface”).
• Web Almanac 2022 — SEO chapter (HTTP Archive) — canonicalization-method and schema-adoption données à travers the web.
• SEO technique Checklist: The Complet Guide (DebugBear) — a strong performance-leaning prendre on the même territory.
• SEO technique checklist (90+ points) (Kristina Azarenko) — a thorough, prioritization-minded alternative checklist.

Stats worth citing

• HTTPS adoption ~91%+. Où sites en réalité stand on the HTTPS baseline heading into 2026 (HTTP Archive données, via Chris Green). Source
• Canonical adoption rose from 65% (2024) to 67%+ (2025). Coverage of the canonical tag is climbing but far from universal. Source
• ~67% of images lack a chargement attribute; 91%+ of iframes lack un. Low-hanging performances wins la plupart sites encore leave on the table. Source
• Conflicting canonical signals appeared on ~0,3–0,4% of pages. Petit but réel — a raison plus grand sites devrait vérifier pour cross-signal canonical conflicts (Web Almanac 2021 SEO chapter, qui I led). Source
• Sitemap fichier limite: 50 MB / 50 000 URLs per fichier — the hard ceiling to split grand sitemaps contre (Google). Source

Journal des modifications