AdSense approval checklist

AdSense will not approve the site while this is true. Fix it first.

• Bot check on the homepageAccess & availabilityPages like “Checking your browser…” stop Google's crawlers as well as people. The AdSense crawler cannot solve them, so the review sees no content.
• Google crawlers can open the siteAccess & availabilityAdSense needs its crawler (Mediapartners-Google) to read your pages. If a firewall or security plugin blocks it, the review fails as “Site down or unavailable”. We can only copy the crawler's name, not Google's IP addresses, so check your firewall logs to be sure.
• Content behind a loginAccess & availabilityGoogle says AdSense must be able to reach your site without a password. If content is behind a login, reviewers and crawlers see nothing to approve.
• Placeholder homepageAccess & availabilityGoogle does not approve sites that are under construction or have no publisher content. A parked, default or suspended page tells the reviewer there is no real site yet.
• AdSense crawler allowed in robots.txtAccess & availabilityGoogle says plainly that if robots.txt blocks the AdSense crawler, it cannot serve ads on the site. This is one of the most common causes of “Site down or unavailable”.
• robots.txt is reachableAccess & availabilityWhen robots.txt fails with a server error, Google's crawlers stop crawling the site until it works again. To AdSense, the site then looks blocked.
• Site is reachableAccess & availabilityIf Google cannot open your site, it rejects the application as “Site down or unavailable” before anyone looks at your content. This is an official requirement: the site must be live and published.
• Enough content to reviewContentGoogle's eligibility rules say a site must have enough content for its specialists to evaluate. A homepage, a few short pages or an empty template give the reviewer nothing to judge, and the application is turned down before anything else is looked at.
• Supported languageContentGoogle does not allow ads on content that is not mainly in a supported language. This is an official rule, so a site in an unsupported language cannot be approved.
• Asking visitors to click adsPolicyAdSense forbids encouraging clicks with phrases like “click the ads” or “support us”. It is one of the top reasons for account closure. We match phrases, so check the examples.
• Signs of a hacked sitePolicyGoogle's spam policies treat hacked content (injected pages, links, code and redirects) as a violation, and AdSense does not serve ads on malware or unwanted software. A cryptominer uses the visitor's computer without asking. Injected spam pages are often invisible from the homepage, so owners miss them.
• Hacking tools and fake generatorsPolicyGoogle does not allow ads next to hacking content, malware or fake generators that trick users. If you did not write these pages, your site may have been hacked.
• Prohibited topicsPolicySome topics, such as adult content, piracy or hacking, cannot carry Google ads at all. A site that covers them is not approved. Our topic match uses words and URLs, so check the examples.
• Google Safe Browsing statusPolicyGoogle does not allow ads on sites that spread malware, phishing or unwanted software. While Safe Browsing flags your site, browsers show visitors a red warning, and AdSense will not approve the site or show ads on it. If you did not set the site up to cause harm, a flag usually means it was hacked.

A documented rejection reason or a strong rejection signal. Fix before you apply.

• Same content for Google and visitorsAccess & availabilityIf Google's crawler sees an empty or different page, it cannot judge your content. Showing different content to Google on purpose is called cloaking, and it breaks Google's spam policies.
• HTTPS certificateAccess & availabilityGoogle's checklist for unreachable sites asks for a valid SSL certificate from a recognised authority, not a self-signed one. Browsers also show a security warning to visitors on sites without one.
• Homepage is indexableAccess & availabilityA noindex homepage often means the whole site was hidden from search by mistake, for example during development. Your articles then can't bring search visitors, and a site that has hidden itself from Google is a poor start for an AdSense review.
• Pages load without errorsAccess & availabilityReviewers click through your site. Many error pages make it look unfinished or hard to navigate, which Google lists as a reason it cannot set up an account.
• Googlebot allowed in robots.txtAccess & availabilityGoogle may turn off ads on content it cannot evaluate, and that includes pages blocked in robots.txt. Blocked pages also cannot appear in Google Search, so they get no search traffic.
• Server response timeAccess & availabilityA slow server can time out for Google's crawler and leads to “site doesn't load” problems. Visitors also leave slow sites. Google's web.dev guidance calls a time to first byte of 0.8 seconds or less good.
• Hidden links to other sitesContentLinks hidden with CSS (display:none, off-screen text, zero size) are link abuse under Google's spam policies. They are almost always injected by a hack or sold to another site, and both put an ad account at risk.
• Placeholder textContentPlaceholder text is a clear sign that the site is still under construction. Google does not approve sites that are under construction.
• Content imported from other sitesContentGoogle does not allow ads on replicated content: copied, scraped or republished from other sources without adding value. RSS auto-blogs are the most common case and a frequent reason for a “Low value content” rejection.
• Thin articlesContentGoogle does not allow ads on screens with low-value content. Google has no minimum word count, but very short articles rarely give a reader enough. “Low value content” is the most common AdSense rejection.
• Reused expired domainDomainGoogle's spam policies name expired domain abuse: buying an old domain and filling it with low-value content to profit from its history. If you bought the domain for its old reputation, reviewers may see it that way. This is our estimate from public archives.
• Free domain extensionDomainFree extensions such as .tk or .ml are widely used for spam. Sites on them are often rejected or hard to verify. This is a common pattern, not a written rule.
• Hosted platformDomainTo use AdSense you must be able to put code in the head of your pages. Many free hosted platforms do not allow this, and AdSense only accepts sites on domains or on approved platforms.
• No pop-up covering the page on phonesUser experienceGoogle says intrusive interstitials make content less accessible on mobile, especially pop-ups that cover the main content right after a visitor arrives. The Better Ads Standards count large pop-ups as a bad ad experience. Banners required by law, such as cookie consent and age checks, are exempt and not counted here. We report a pop-up only when it still covers the screen two seconds later, on both phone sizes and on a reload.
• Topic focusTopic & nicheSites that jump between unrelated topics are often rejected as low value, because they look like they were written to collect traffic. A clear focus helps readers and reviewers. This is our estimate.
• Different content for GooglePolicyShowing search engines different content from visitors is cloaking, which Google's spam policies forbid. Hacked sites often do this to show pharma, casino or Japanese spam only to Googlebot, so the owner never notices. Google Publisher Policies apply the spam policies to every page with Google ads.
• Copyright and piracyPolicyGoogle does not allow ads on content that infringes copyright, such as pirated downloads or streams. This is a common reason for rejection and closure. We match words, so a page about fighting piracy can also show up here.
• Hateful or derogatory contentPolicyGoogle does not allow ads on content that incites hatred, promotes discrimination or disparages people for who they are. We match words only, so a page that reports on or criticises hate can also show up here.
• Enabling dishonest behaviourPolicyGoogle does not allow ads on content that helps people deceive others, such as fake documents, essay mills, bought reviews or spy apps.
• Paid link shortenersPolicyPaid shorteners such as adf.ly, ouo.io or Linkvertise show ads and countdowns before the real link. AdSense forbids misleading navigation and pages that redirect visitors to unwanted sites, and these links are typical of download and piracy sites.
• Redirect to another sitePolicyGoogle's policies forbid sites that redirect users to unwanted sites. A redirect you did not add usually means a hacked plugin, theme or ad script.
• Restricted topicsPolicyRestricted content is allowed, but fewer advertisers can bid on it, so you earn less. When it is the main topic of the site, rejection is also more likely.
• Shocking or violent contentPolicyGruesome or graphic content gets restricted ads or none at all. We match words only, so news and education pages can show up here too.
• Redirects for some visitors onlyPolicyGoogle's spam policies call this a sneaky redirect: a normal page for some visitors and another site for others, often only for phone users or people who arrive from search. Google Publisher Policies apply the spam policies to every page with Google ads. A redirect you did not add almost always means the site was hacked.
• Links to gambling, pharma or adult sitesPolicySite-wide links to casinos, betting, pharmacies or adult sites on an unrelated site are typical of sold links or of a hack. Buying or selling links is link spam under Google's spam policies, and linking to these topics also ties your site to restricted or prohibited content.
• Paid-to-click and traffic exchangesPolicyAdSense forbids traffic from paid-to-click, autosurf and click-exchange programs. Promoting them on your site is a red flag for reviewers.
• Spam in commentsPolicyYou are responsible for everything on a page that shows ads, including comments. Google's AdSense guidance asks you to moderate comments, and its spam policies list user-generated spam. Reviewers who see casino or pharma links in comments read the site as unmoderated.
• Template pagesProgrammatic SEO & duplicationMany pages that differ only in a city or keyword are doorway pages under Google's spam policies. They are a typical cause of “Low value content”. This is our estimate from text overlap.
• Overall page qualityPage qualityWe score each article with questions based on Google's quality guidance: effort, structure, readability, who wrote it and when. It is our estimate, not Google's score, but low scores across the site match the “Low value content” pattern.
• Script from a compromised sourceSecurityIn 2024 the polyfill.io domain and several sister domains were bought by a new owner and began to serve code that sent some visitors to scam and malware sites. Every page that still loads a script from them can do the same to your visitors, and Google may flag the site as harmful (no ads are served on sites with malware).
• Exposed private filesSecurityBots request addresses like /.env, /.git/HEAD and /backup.zip on every site, every day. An environment file usually holds the database password and API keys; a readable .git folder lets anyone download the full source code; a backup or SQL dump is the whole site and its users. We only call a file exposed when it answers normally (status 200, no redirect) and its content has the format of that file type (for example KEY=value lines for .env). We never store the contents: the evidence shows the address and what kind of file it is.
• Broken linksSite structureGoogle's policies name “linking to content that doesn't exist” as misleading navigation. Broken links also make a site look neglected.
• Navigation menuSite structureGoogle's program policies say sites with Google ads should be easy to navigate. “Difficult to navigate” is also listed as a reason Google cannot set up an account.
• Canonical points to another siteTechnical SEOA canonical tag to another site tells Google the original content is elsewhere. That looks like copied or syndicated content, and Google may treat your pages as duplicates.
• Outdated or vulnerable softwareTechnical SEOHacked sites are rejected: Google's spam policies cover hacked content, and AdSense does not serve ads on sites with malware. Most break-ins go through software with a published security hole, and attackers scan the web for exactly the version numbers a site shows. We only read versions your site announces itself (server headers, the generator tag, file addresses and what the browser reports) and compare them with the vendors' security advisories. A version we cannot see is never reported. Critical means the hole is actively exploited or lets attackers run code on your server. Web servers and PHP are often patched by the hosting company without changing the number, so ask your host before acting on those.
• Mobile viewportTechnical SEOWithout a viewport tag, phones show a tiny desktop page. Most visitors come from phones, and Google asks for a good experience on the site.
• Privacy policyTrust & required pagesA privacy policy is required for AdSense. It must explain that third parties, including Google, may use cookies to serve ads. This is an official rule.
• Thin affiliate contentTrust & required pagesGoogle's spam policies name thin affiliate pages: product text copied from the seller with nothing added. Google's AdSense guide also warns against affiliate pages with little original content.
• Publisher ID matches ads.txtVerification & ads.txtGoogle's policy says you must not place Google ads on a domain whose ads.txt does not list you as an authorised seller. A wrong ID in ads.txt means ads may not show even after approval.
• Pop-under or aggressive ad networksVerification & ads.txtAdSense does not allow Google ads on sites that contain or trigger pop-unders. Reviewers see these networks during review, and they also annoy visitors.

Weakens your application. Fix it before you apply if you can.

• www and non-www versionsAccess & availabilityMany people type your domain with or without “www”. If one version fails, Google can treat the URL you submitted as wrong or unreachable. If both load, you have two copies of the same site.
• Certificate expiryAccess & availabilityWhen a certificate expires, browsers block your site with a warning and Google may see it as unreachable. If this happens during review, it can cause a rejection.
• HTTP to HTTPS redirectAccess & availabilityGoogle's own checklist asks whether your site redirects HTTP to HTTPS. Without a permanent redirect, Google and visitors can reach two copies of every page.
• Redirect chainAccess & availabilityEach redirect adds delay and another place where things can break. Long chains can make crawlers give up, and it is harder to confirm the URL you submitted is the real site.
• Number of articlesContentGoogle asks for “enough content for our specialists to evaluate” but gives no number. Sites with fewer than about 20 useful articles are often rejected as “Low value content”. The count matters less than the quality of each article.
• Default CMS contentContentDefault posts and demo pages make a site look unfinished. Reviewers notice them at once.
• Duplicate titlesContentThe same title on several pages makes it hard for readers and Google to tell them apart. It often points to copied or templated posts.
• Embed-only articlesContentGoogle's replicated-content policy covers pages that embed others' content without adding value. Google asks you to add your own contribution.
• Generic, look-alike articlesContentGoogle asks whether content offers original information, research or experience, and its spam policies cover content produced at scale with little value. For each article we look for concrete signs of real work: numbers and measurements, specific names, your own photos, first-hand experience, an author and a date, and sources. We flag only articles that have almost none of them and also read like a formula (many stock phrases, or the same heading outline as many other posts). This is our estimate; it does not say who or what wrote the text.
• Hidden textContentText hidden from visitors but present for search engines breaks Google's spam policies. It can also be a sign that the site was hacked. Accordions and tabs are fine, so check the examples.
• Keyword stuffingContentRepeating a keyword again and again to rank is keyword stuffing, which breaks Google's spam policies. It also makes the text hard to read.
• Near-duplicate pagesContentGoogle's policy on replicated content covers copied or rewritten text that adds no value. Many near-identical pages look like mass-produced content to a reviewer.
• Repeated sentences across pagesContentThe same paragraph on many articles makes pages look generated from a template. Boilerplate in the main text also lowers the share of original content on each page.
• Tool pages with little explanationContentA tool page that is only a widget gives a reviewer nothing to read. Google does not serve ads on screens without publisher content, and a tool with no explanation can look like low-value inventory. We count the page's own words around the tool, without menus, footers, tables and option lists.
• Tool site without editorial contentContentAdSense has no blog requirement, but a site made almost only of tools is a common “low value content” or “no publisher content” rejection: the reviewer sees widgets, not content you wrote. A visible guides or blog section with a few genuinely useful, in-depth articles, plus real explanations under each tool, makes the site look substantial. This is a community-reported pattern, not a written rule.
• Domain in a restricted countryDomainGoogle publisher products are not available in Crimea, Cuba, Iran, North Korea and the occupied regions of Donetsk and Luhansk. A country domain does not prove where you live, but the account holder's address and payment details must be outside these places.
• Layout stays still while loading (our phone test)User experienceThis is our own lab measurement of Cumulative Layout Shift, without scrolling or tapping, on two phone sizes and a reload. It can differ from the real-user Core Web Vitals that Google shows elsewhere in this report and in Search Console. Good is 0.1 or less, poor is above 0.25. Jumping content makes readers lose their place and tap the wrong thing, often an ad.
• Scripts, styles, images and fonts loadUser experienceA missing script or stylesheet can break the layout and features; missing images and fonts leave gaps. Reviewers see the same broken page. We count only files on your own domain that returned an HTTP error on all three loads; rate limits and bot checks are ignored.
• Pages fit the phone screenUser experienceGoogle's mobile guidance asks for pages that fit the screen without sideways scrolling or zooming. A page that wobbles sideways looks broken, and most visitors use phones. We report it only when a real swipe moved the page sideways on both screen sizes and on a reload, and we name the element that sticks out.
• Mobile viewport on every pageUser experienceWithout a viewport tag set to width=device-width, phones lay the page out about 980px wide and shrink it, so text is tiny and links are hard to tap. Google's mobile guidance asks for it on every page, not only the homepage.
• Low-value nicheTopic & nicheSites like these are usually built from copied or easily copied content. They are rejected for “Low value content” very often. This is a common pattern, not a written ban.
• Off-topic articlesTopic & nicheA few off-topic posts are normal. Many of them make the site look like it chases traffic rather than serving one audience. This is our estimate from topic words.
• Many ad units, little contentPolicyGoogle does not allow more ads than publisher content on a screen, and the Better Ads Standards treat high ad density as a bad experience. Pages built around ad units rather than content are the “made for advertising” pattern reviewers reject. We count units in the page source, so ads added later by scripts are not included.
• Hidden or mislabelled ad unitsPolicyAdSense does not allow ad code to be changed: hiding units with display:none is forbidden, except in responsive units. Ads may only be labelled “Advertisements” or “Sponsored Links”. Headings like “Useful links” or “Download” trick visitors into clicking.
• Ads on pages without contentPolicyGoogle does not allow ads on screens without publisher content: error, thank-you, exit, login and empty pages. Reviewers judge the whole site, and ad code on these pages is a policy issue once ads run.
• Page reloads itselfPolicyAdSense does not allow a page or part of a page to refresh without the visitor requesting it. Automatic reloads create ad impressions nobody asked for, and they are common on news sites with a refresh meta tag.
• Google name in the domainPolicyAdSense does not allow ads on pages whose content or URL could make people think they are linked to Google through misuse of its trademarks. YouTube downloaders also break copyright and YouTube's terms. A reviewer may decide on the domain alone.
• Back button manipulationPolicyGoogle's spam policies list back button hijacking, and Chrome's abusive experiences list includes pages that keep the visitor from going back. Scripts that do this are usually “recommended content” or ad widgets that show a page of ads when the visitor presses back.
• Selling linksPolicySelling links that pass ranking credit is link spam under Google's spam policies. Sites that sell dofollow posts usually fill up with unrelated sponsored articles, which is also the site reputation abuse pattern. Reviewers see an advertising page for links as a sign that the site exists to sell placements.
• Misleading claimsPolicyGoogle's policies forbid misleading content and harmful health claims that go against scientific consensus. Exaggerated claims also lower reader trust.
• Many pages on a new domainProgrammatic SEO & duplicationThousands of pages in a few months is a typical sign of mass-produced content. Google's spam policies call this scaled content abuse when the pages add little value. This is our estimate, not proof.
• Thin page groupsProgrammatic SEO & duplicationA large group of short, similar pages is what Google means by scaled, low-value content. One weak group can drag down how the whole site is judged.
• Sources on sensitive topicsPage qualityOn health, money and legal topics, readers need to know where the information comes from. Links to official bodies or research show your claims are based on evidence.
• Heading structurePage qualityA clear heading structure helps readers scan an article and helps search engines understand it. Messy headings usually come from the theme or page builder.
• SubheadingsPage qualityLong articles without sections are hard to read. Good subheadings show the reader, and the reviewer, that the article is organised.
• Template-filled articlesPage qualityPages that fill one template with different keywords are a typical pattern of scaled, low-value content. Google's spam policies and its replicated-content policy cover this.
• Honest titlesPage qualityGoogle's content guidance asks whether titles avoid exaggerating or shocking. Clickbait titles also make a site look spammy to reviewers.
• Cookie security flagsSecurityA cookie without Secure can also be sent over an unencrypted http:// connection, where it can be read. A session cookie without HttpOnly can be stolen by an injected script, which hands the attacker the logged-in session. We only list cookie names, never their values. Cookies that scripts must read (consent, preferences) do not need HttpOnly.
• HSTS (Strict-Transport-Security)SecurityHSTS tells browsers to open your site only over https:// for a set time. Without it, a visit that starts at http:// (a typed address, an old link) can be intercepted on public Wi-Fi before the redirect to https:// happens. It is one line in the server settings once https:// works everywhere.
• Canonical points to a bad pageTechnical SEOA canonical should point to a working page that Google can index. If it points to an error page, a redirect or a noindex page, Google gets conflicting signals and may ignore the tag or drop the page.
• Canonical uses http://Technical SEOThis tells Google that the insecure http:// version is the main one. Google usually prefers HTTPS anyway, but mixed signals can slow things down and send some visitors through an extra redirect.
• More than one canonical tagTechnical SEOWhen a page has several canonical tags, Google may ignore every one of them. This usually happens when the theme and an SEO plugin, or two plugins, each add their own.
• Empty main headingTechnical SEOAn empty H1 is usually a theme bug, such as a logo wrapped in a heading without text, or a title field that was never filled in. Readers and screen readers get no heading, and Google gets no hint about the subject from it.
• hreflang points to bad pagesTechnical SEOEvery address in an hreflang set must be a working, indexable page that returns 200. Links to errors, redirects or noindex pages break the set, and Google may ignore it.
• Valid hreflang codesTechnical SEOAn hreflang code must be a language code from ISO 639-1 (such as en, tr or de), optionally followed by a region code from ISO 3166-1 (such as en-GB or pt-BR), or x-default. Google ignores invalid codes. A common mistake is “uk” for the United Kingdom: uk means Ukrainian, so use en-GB.
• hreflang without return linksTechnical SEOhreflang only works in pairs. If page A lists page B as its translation, page B must also list page A. Google ignores one-way hreflang links.
• Relative hreflang addressesTechnical SEOGoogle requires fully qualified URLs in hreflang, including https:// and the domain. Relative paths such as /en/page can be misread or ignored.
• Images much larger than shownTechnical SEOEvery visitor downloads the full file, even when it is shown in a small box. That wastes data and slows the page, especially on phones. This is an estimate: we compare the file's real pixel size with the img tag's width attribute, and CSS can show an image larger than that attribute. We only check a few of the largest images.
• Meta refresh redirectsTechnical SEOA meta refresh reloads or redirects the page after it has loaded. Google understands instant meta refresh redirects, but a server-side 301 is faster and clearer. A delayed refresh that moves readers on without asking annoys them and can look like a sneaky redirect.
• More than one meta descriptionTechnical SEOWith several meta descriptions on one page, Google cannot tell which one you mean. This usually happens when the theme and an SEO plugin both add one.
• More than one title tagTechnical SEOA page should have exactly one title tag. With several, Google has to guess which one you mean. This usually happens when the theme and an SEO plugin both add a title.
• Redirect chainsTechnical SEOChains slow pages down and waste crawling. Googlebot follows up to 10 redirects in a row, but each extra step adds risk, and other crawlers may give up sooner.
• robots.txt blocks CSS or JavaScriptTechnical SEOGoogle renders pages the way a browser does. If it cannot load your CSS and scripts, it may see a broken or empty page, think it is not mobile-friendly, or miss content.
• Incomplete structured dataTechnical SEOStructured data (schema.org JSON-LD) helps Google understand a page and can make it eligible for rich results. Items with missing properties may not qualify. Structured data is not needed for AdSense approval, and it must describe content that is visible on the page.
• Sitemap entries that should not be thereTechnical SEOA sitemap should list only the final, indexable address of each page. Error pages, redirects, noindex pages and non-canonical copies waste crawling and send Google mixed signals about which pages matter.
• Same page with and without a slashTechnical SEOGoogle treats /page and /page/ as different URLs. If both show the same content with no redirect or canonical, you have duplicate pages that split your signals.
• Broken imagesSite structureBroken images make pages look unfinished, and reviewers notice. They often appear after a site move or a deleted media library.
• Links between articlesSite structureLinks between related articles help readers go deeper and help crawlers find every page. A site where every post is a dead end looks thin.
• Links between articlesSite structureLinks inside your text help readers find more on the same subject and help Google understand how your pages relate. Articles that no other article links to are easy to miss. Menu and footer links do not count here. This is our estimate from the pages we crawled.
• XML sitemapSite structureA sitemap helps Google find all your articles, including ones that are hard to reach by links. It is not required, but it makes your site easier to crawl and review.
• Dead URLs in sitemapSite structureA sitemap should list only working pages. Dead entries waste crawl time and look careless.
• Sitemap errorsSite structureA broken sitemap is worse than none, because crawlers waste time on it and may miss your pages.
• Canonical points to another pageTechnical SEOThis tells Google to ignore these articles and use another page instead. It is usually a theme or plugin mistake, and it hides your content.
• Mixed contentTechnical SEOBrowsers block or warn about insecure files on secure pages. Images may disappear and the padlock may break, which hurts trust.
• Good pages marked noindexTechnical SEOThese look like real articles with enough text, or your homepage, but they tell Google not to index them. They get no search visitors, and a reviewer may see too little content. This is often a setting left over from when the site was being built.
• Articles marked noindexTechnical SEONoindex hides pages from Google Search, so they get no search visitors. If many articles are noindex, a reviewer may see too little content.
• Missing page titlesTechnical SEOThe title tag names the page in browser tabs and search results. A missing title often points to a broken theme or template.
• Affiliate disclosureTrust & required pagesGoogle's policy forbids hiding the purpose of your content. Readers should know when you earn money from a link. Many countries also require this by law.
• Consent banner (CMP)Trust & required pagesTo show personalised ads to visitors in the EEA, the UK and Switzerland, Google requires a certified CMP that supports the IAB TCF. Without one, those visitors may get limited ads or none. It does not block approval, but it affects earnings.
• Contact informationTrust & required pagesNo Google rule names a contact page, but a site with no way to reach the owner looks anonymous. Reviewers and Product Experts treat this as a trust problem.
• Cookie consent for EU visitorsTrust & required pagesUnder the GDPR in Europe, a site needs the visitor's consent before it sets ad or analytics cookies. Google's EU User Consent Policy requires publishers to get that consent, and AdSense requires a Google-certified consent tool (CMP) to show ads to visitors in the EEA, the UK and Switzerland. Turkey's KVKK has similar rules for visitors from Turkey. We load your homepage from a server in the EU, so we see what an EU visitor sees.
• Cookies set before consentTrust & required pagesConsent has to come first. Under the GDPR and Google's EU User Consent Policy, ad and analytics cookies may only be set after the visitor agrees. A banner that appears while trackers are already running does not count as consent. We loaded your homepage as a visitor from the EU and did not click anything.
• Disclaimer for sensitive topicsTrust & required pagesOn health, money, law and safety topics, readers need to know who is giving advice and on what basis. Google's content guidance asks for this kind of transparency. It is not a written AdSense rule.
• Required privacy statementsTrust & required pagesGoogle lists what your privacy policy must say: that third-party vendors, including Google, use cookies to serve ads, and how users can opt out. A generic template often leaves these out.
• Real-user speed (Core Web Vitals)User experienceThese numbers come from the Chrome UX Report (CrUX): real visits by Chrome users over the last 28 days, measured at the 75th percentile. Good means LCP (loading) of 2.5 s or less, INP (responsiveness) of 200 ms or less and CLS (visual stability) of 0.1 or less. Poor means more than 4 s, 500 ms or 0.25. Core Web Vitals are part of Google Search's page experience signals. They are not an AdSense approval rule, but slow pages lose readers and earn less.
• Pop-up covering the pageUser experienceThe Better Ads Standards, which Google's policies follow, count large pop-ups and full-screen overlays as bad experiences. Reviewers want to see your content first.
• Content needs JavaScriptUser experienceGoogle can run JavaScript, but not always quickly or fully. If your text appears only after scripts run, crawlers may see an almost empty page.
• Fits on a phone screenUser experiencePages that scroll sideways are hard to read on a phone, and most visitors come from phones. Google asks for an easy, readable experience.
• Pop-up windowsUser experienceAdSense does not allow sites that interfere with navigation through pop-ups or pop-unders, and Google ads may not appear in windows the visitor did not open. Pop-up windows usually come from aggressive ad networks.
• Loading speed on mobileUser experienceSlow pages lose readers, especially on mobile data. Under 2500 ms is a good target. Speed matters for visitors more than for approval.
• Google line in ads.txtVerification & ads.txtGoogle gives one exact line to use. If your site has an ads.txt file that does not list your publisher ID correctly, Google may not buy your ad space.
• ads.txt formatVerification & ads.txtCrawlers skip lines they cannot read. Invisible characters, rich-text quotes or extra commas are common causes. If your Google line is broken, it does not count.
• AdSense code problemsVerification & ads.txtTo connect your site, Google asks for the code with your own publisher ID in the head section of your pages. A wrong ID, or someone else's ID left in a theme or plugin, makes verification fail or sends earnings to another account.

Worth doing. On its own it has a small effect.

• Buttons have a nameUser experienceIcon-only buttons need a text alternative so that screen-reader users know what they do.
• Text contrastUser experienceWCAG AA asks for a contrast ratio of at least 4.5:1 for normal text and 3:1 for large text. Low-contrast text is hard to read, on a phone outdoors even more so. axe-core computes the actual colours; unclear cases (text over images or gradients, text that is still animating) are skipped.
• Page language declaredUser experienceScreen readers use the lang attribute to pick the right pronunciation, and browsers use it for translation and hyphenation.
• Images have alt textUser experienceScreen readers and Google Images rely on alt text, and axe-core rates a missing alt attribute as critical. Decorative images are fine with an empty alt="": only images without any alt attribute are counted.
• Form fields have labelsUser experienceText next to a field is not enough: it has to be connected to the field (a label element whose for attribute points to the field, a label that wraps the field, or an aria-label). Otherwise screen-reader users hear only “text field” and don't know what to type.
• Links have a nameUser experienceIcon links and image links without text are announced as just “link”, or as the web address.
• robots.txt formatAccess & availabilityCrawlers expect plain text at /robots.txt. A web page there means your rules are ignored, and it often means a theme or plugin is answering every address.
• Duplicate meta descriptionsContentMeta descriptions are the snippet people see in search results. Repeated ones look careless and do not help anyone choose the right page. This is a small issue on its own.
• Update dates that do not changeContentGoogle asks you not to change page dates to make content look fresh when it has not changed much. Sitemap lastmod values are only used when they are accurate. Dates that change on every build or every visit are ignored and can look manipulative.
• Recent contentContentA site with no new posts for months looks abandoned to a reviewer. Google does not reward fake date changes, so publish or update for real.
• Image alt textContentAlt text describes an image for screen readers and for search engines. Missing alt text is a small issue on its own, but it is part of a well-built page.
• Generic filler phrasesContentAI or template tools are allowed, but Google warns against mass-producing pages without adding value. Many stock phrases are a sign of unedited generated text. This is our estimate, not proof.
• Mixed languagesContentGoogle checks that content is mainly in a supported language. A random mix of languages also makes the site harder to review and to read.
• Long articles without subheadingsContentLong text without subheadings is hard to scan, and readers leave. Clear structure is one of the signs of effort that reviewers look for.
• Domain expiryDomainIf the domain expires, the site goes offline and someone else can register it. A site that goes down during review is rejected.
• SubdomainDomainIn AdSense you add the main domain as a site, not a subdomain. The reviewer may look at the whole domain, not only this part.
• Topic changed over timeDomainA topic change is normal for many sites. It only matters if the old reputation is used to rank new, unrelated content. This is our estimate from public archives.
• Broken imagesUser experienceBroken images look neglected to readers and to reviewers. We count an image only when the browser shows it as broken and its file request failed, on all three loads.
• Text cut off on phonesUser experienceWhen a box hides what does not fit (overflow: hidden) and no ellipsis is meant, readers lose part of a label or heading and may not understand what a button does. We list a case only when the box really cuts through its own text, on both phone sizes and on a reload. Ellipsis (…), line clamps, screen-reader-only text, tickers and carousels are skipped.
• Buttons that seem to do nothingUser experienceWe asked the browser's developer tools which event handlers are attached to each button and to every element around it, and checked that hover and focus styles do not reveal anything. Pages whose scripts listen at the top level (React, Vue, jQuery delegation, plugins that delay JavaScript until the first tap) are skipped, because we cannot tell there. Please tap the button yourself before changing anything.
• No repeated JavaScript errorsUser experienceAn error that repeats on every load usually means a broken feature: a menu, gallery, form or ad slot that never starts. We count only uncaught errors from scripts on your own domain (not ads, analytics or network hiccups), and only when they appear on all three loads.
• Main content shows quickly (our phone test)User experienceThis is our own lab measurement of Largest Contentful Paint from our server, without speed throttling. A page is listed only when it is over 4 s on all three loads and the page's first byte or the main element's own download took 2.5 s or more, so a busy test machine cannot cause it. Visitors on mobile networks are usually slower. Google's real-user data (CrUX) is shown separately in this report. Good is 2.5 s or less, poor is above 4 s.
• Mobile menu opensUser experienceIf the menu does not open on phones, visitors and reviewers cannot reach your other pages, including About, Contact and the privacy policy. We tapped the button once, waited three seconds, clicked it once more and looked for new links on the screen. Leaving the page was blocked during the test. If your menu opens in a way we could not see (inside an iframe or after a long animation), ignore this.
• Buttons or links covering each otherUser experienceWe found two tappable elements on top of each other and checked in the browser that a tap on the centre of the lower one hits the upper one. Visitors cannot reach the covered one. Closed menus, carousel arrows at the edge of a slide, sticky headers and two links to the same page are not counted.
• Buttons and links big enough to tapUser experienceWCAG 2.2 (level AA, 2.5.8) asks for tap targets of at least 24×24 CSS pixels, or enough space that a 24px circle around each one does not touch another target. Links inside sentences are exempt. We report only clear cases (small in both directions, or under 16px), confirmed by axe-core and by our own measurement. For comfortable tapping Google recommends about 48×48px.
• Readable text size on phonesUser experienceGoogle's mobile guidance asks for legible font sizes, and Lighthouse flags pages where much of the text is under 12px. Small print is hard to read without zooming, especially for older readers.
• Words split in the middleUser experienceCSS such as word-break: break-all or overflow-wrap: anywhere lets the browser cut a word at any letter. In narrow buttons and headings this turns “Standard” into “Stan / dard”, which looks broken and is hard to read. We measure where each letter lands, so only words that really break across two lines are listed.
• Pinch-zoom allowedUser experienceuser-scalable=no, or a maximum-scale below 2, stops people from zooming in on many phones. WCAG asks that text can be resized to 200% (1.4.4). This is an accessibility note, not an AdSense rule.
• Core subtopicsTopic & nicheA site that covers the main questions of its topic looks complete and useful. Gaps make it look thin. This list is our own suggestion, not a Google rule.
• Shallow subtopicsTopic & nicheOne article per subtopic rarely gives a reader the full picture. A few deeper groups of related articles show more expertise than many scattered posts.
• Bulk publishingProgrammatic SEO & duplicationHundreds of pages on one day often means automatic publishing. Google does not allow ads on automatically generated content without manual review. A site move or bulk update can also cause this, so check before you act.
• Author names on articlesPage qualityGoogle's content guidance asks whether it is clear who created the content. It matters most on health, money and legal topics.
• Heading matches the titlePage qualityGoogle's content guidance asks whether the main heading and title give a clear, honest summary of the page. When they disagree, readers from search may feel misled.
• Images, tables and listsPage qualityLong plain text with no visuals or lists looks low-effort and is harder to follow. Your own photos and tables show original work.
• ReadabilityPage qualityHard-to-read text makes visitors leave. Readability scores are rough, but very low scores usually mean long sentences and heavy words.
• Stock photosPage qualityStock photos are allowed, but they add no original value. Your own photos and screenshots show first-hand experience.
• Title lengthPage qualityVery short titles do not explain the page. Very long ones get cut off in search results. Aim for about 30–65 characters.
• Walls of textPage qualityHuge paragraphs are tiring, especially on a phone. Short paragraphs help readers stay.
• Protection against clickjackingSecurityIn a clickjacking attack another site loads your page in an invisible frame and tricks visitors into clicking buttons on it (sign in, buy, subscribe). X-Frame-Options or the frame-ancestors rule of a Content-Security-Policy tells browsers who may frame your pages. AdSense ads run in their own frames, so this does not affect them.
• Content-Security-PolicySecurityA Content-Security-Policy tells the browser which scripts, frames and other files a page may load. If an attacker manages to inject a script (through a vulnerable plugin, a comment field or a hacked third-party file), the policy can stop it from running. Sites with ads must allow Google's ad domains, so build the policy in Report-Only mode first and tighten it step by step. This is good practice, not an AdSense requirement, so we show it as a notice.
• Folder listingSecurityA folder listing lets anyone browse every uploaded file, including drafts, private documents and backups that are not linked anywhere. Attackers use it to find files to download or to learn which plugins you run.
• X-Content-Type-OptionsSecurityWithout this header some browsers guess a file's type from its content, so an uploaded file that looks like a script can run as one. It is one line and has no side effects on a correctly configured site.
• Referrer-PolicySecurityWhen a visitor follows a link or a page loads a file from another site, the browser tells that site where the visitor came from. The unsafe-url and no-referrer-when-downgrade policies send the full address, which can contain search terms, e-mail addresses or tokens.
• Subresource Integrity on CDN scriptsSecurityWhen you load a fixed library version (jQuery 3.7.1, Bootstrap 5.3.0…) from a public CDN, the integrity attribute pins the exact file: if the CDN is ever hacked or the file changes, the browser refuses to run it. Scripts that change on purpose (analytics, ads) cannot be pinned and are not listed.
• Software versions on displaySecurityAttack bots look for exact version numbers to pick sites with known holes. Hiding the number does not fix an outdated program, but it stops your site from being an easy match. Keeping the software updated matters far more; the “Outdated or vulnerable software” check compares the versions we can see with security advisories.
• WordPress readme.html is publicSecurityThe file tells bots that the site runs WordPress and, on older versions, which version. It is not dangerous on its own, but it is an easy fingerprint to remove.
• WordPress usernames are publicSecurityKnowing the login name is half of a password guess. Bots read /wp-json/wp/v2/users and /?author=1 to collect usernames, then try common passwords on wp-login.php. We only count the accounts and never store their names.
• Render-blocking scriptsTechnical SEOScripts in the head without async or defer stop the browser from showing the page until each one has downloaded and run. This delays the first view of your content, especially on phones.
• Broken outbound linksTechnical SEOLinks to dead pages frustrate readers and make content look neglected, which reviewers notice on sites that cite sources. It does not directly affect approval, but it weakens trust.
• Canonical tag missingTechnical SEOA canonical tag tells Google which address is the main one for a page. Without it, copies of the same page (with tracking parameters, with or without www, with or without a trailing slash) can compete with each other. It is not required, but it prevents mix-ups.
• Crawl-delay in robots.txtTechnical SEOGoogle ignores Crawl-delay, so it does not slow down Googlebot or the AdSense crawler. Some other search engines do follow it, and a high value can make them visit your pages much less often.
• Long meta descriptionsTechnical SEOGoogle has no length limit for meta descriptions, but it shortens snippets to fit the screen, usually after about 155–160 characters. Google also often writes its own snippet from the page text, so your description is a suggestion, not a guarantee.
• Short meta descriptionsTechnical SEOA meta description is your chance to tell searchers what they will find. A very short one gives Google little to work with, so it is more likely to pick other text from the page instead.
• Links without textTechnical SEOA link with no text, such as an icon without a label, tells Google and screen readers nothing about where it goes.
• Vague link textTechnical SEOGoogle uses link text to understand the page you link to, and readers scan links to decide where to go next. “Click here” tells neither of them anything.
• Same main heading on several articlesTechnical SEOWhen several articles share the same H1, readers and Google cannot tell them apart. It can also be a sign of templated or copied pages.
• Long main headingsTechnical SEOThe H1 is the first thing a reader sees on the page. A very long heading is hard to take in at a glance, and Google may use your heading when it writes a title link.
• hreflang list without the page itselfTechnical SEOEach language version should list all versions, including itself. Google recommends this so that the set is the same on every page.
• No x-default in hreflangTechnical SEOx-default tells Google which page to show when none of your languages matches the searcher, for example a language picker or your main version. It is optional, but recommended.
• Very large HTMLTechnical SEOLarge HTML takes longer to download and process, especially on phones. It is usually caused by images embedded as base64, large inline scripts or styles, or a very long page. Googlebot reads only the first 2 MB of an HTML file, so a page this large is close to the point where content gets cut off, and it is slow for readers.
• Heavy imagesTechnical SEOLarge image files slow pages down, especially on mobile data, and they are a common cause of a slow Largest Contentful Paint (LCP). Most article images look fine at well under 200 KB.
• Older image formatsTechnical SEOWebP and AVIF files are usually much smaller than JPEG or PNG at the same quality, so pages load faster. All current browsers and Google Images support them. This is a speed improvement, not an approval rule.
• Images without width and heightTechnical SEOWithout width and height, the browser does not know how much space to keep for an image, so the text jumps when it loads. This layout shift (CLS) annoys readers and can cause misclicks, especially near ads.
• Links to http:// pagesTechnical SEOLinks to http:// either go through a redirect or send readers to an insecure page. Links to your own pages over http:// often point to old addresses from before you moved to HTTPS.
• Internal links that redirectTechnical SEOEach redirect adds a delay for readers and an extra request for crawlers. It usually happens after a permalink change, or when links use http:// or the wrong www version. Google follows redirects, so this is a small issue, but it is easy to fix.
• Page language labelTechnical SEOGoogle works out a page's language from its visible text and does not use the lang attribute for that. Browsers, translation tools and screen readers do use it, so a wrong label can offer readers a translation they do not need or read the text aloud with the wrong voice. This is our estimate from the text.
• More than one viewport tagTechnical SEOThe viewport tag tells phones how to size the page. With two of them, the browser uses only one and may pick the wrong settings, so the page can look zoomed out or too wide.
• Missing doctypeTechnical SEOWithout the HTML5 doctype line, browsers show the page in “quirks mode”, an old compatibility mode that can break layout and sizes. It is usually a theme or caching bug.
• Nofollow on internal linksTechnical SEOnofollow is meant for links you do not vouch for, such as ads or user comments. On links to your own pages it only makes it harder for Google to find your content and see how it connects.
• Conflicting robots instructionsTechnical SEOWhen the meta tag and the HTTP header disagree, Google follows the most restrictive one, so the page is treated as noindex. This is usually a server or plugin setting that someone forgot.
• Security headersTechnical SEOStrict-Transport-Security (HSTS) tells browsers to always use HTTPS, and X-Content-Type-Options: nosniff stops browsers from guessing file types. Neither is required by AdSense or Google Search, but both are cheap protection for your readers.
• Sitemap lastmod shows the build date, not real changesTechnical SEOGoogle uses lastmod only if it is consistently and verifiably accurate. When every URL carries the same recent date, Google learns to ignore your lastmod, so real updates are picked up more slowly. It is not a penalty, but it wastes a useful signal and can look like fake freshness.
• Temporary redirectsTechnical SEOA temporary redirect tells Google the move is not permanent, so it may keep the old address as the main one. For pages that have moved for good, a permanent redirect (301 or 308) sends a clearer signal.
• Long titlesTechnical SEOGoogle has no fixed length limit for titles. It shortens title links to fit the screen, so very long titles often end in “…” and lose their last words. A clear, shorter title is easier to read.
• Short titlesTechnical SEOA very short title, such as “Home” or a single word, rarely tells searchers what the page is about. Google may then write its own title link from your headings or other text on the page.
• Messy URLsTechnical SEOGoogle recommends simple, readable URLs with hyphens between words. URLs are case-sensitive, so capital letters can create duplicate addresses, and very long URLs are hard to share. This is our estimate, and changing existing URLs is only worth it with proper redirects.
• BreadcrumbsSite structureBreadcrumbs (Home › Category › Article) show readers where they are and how to go back. They are optional, but they help navigation.
• Pages buried deep in the siteSite structureReaders rarely click four or more times to find a page, and Google finds and revisits pages mainly by following links. Pages buried deep get fewer visits and may be crawled less often. This is our estimate from the shortest link path we found. We do not follow tag, category and archive pages, so the real depth can be lower.
• Site footerSite structureThe footer is where people expect links to the privacy policy, About and Contact. A missing footer makes those pages harder to find.
• Topics without a hub pageSite structureWhen the articles on one subject are scattered, readers cannot find the rest, and Google sees no clear structure. A hub page, such as a category page with an introduction or a guide that links to every article on the subject, ties them together. This is our estimate from the pages we crawled.
• Archive and utility pages in the indexSite structureTag pages that are only lists of links, page 2, 3 and 4 of lists, internal search results and login or cart pages give searchers little. When many of them are indexed, they can outnumber your articles and make the site look thin. This is our estimate from the pages we crawled.
• Orphan articlesSite structurePages that only exist in the sitemap are hard for readers to find. Many orphan pages can also be a sign of bulk-published content. This is our estimate from the pages we crawled.
• Sitemap full of list pagesSite structureList pages have no content of their own. When they fill the sitemap, a reviewer or crawler sees mostly thin pages.
• Sitemap listed in robots.txtSite structureA “Sitemap:” line in robots.txt lets every crawler find your sitemap without guessing. It is a small, easy fix.
• CompressionTechnical SEOCompression makes pages download several times faster. It is usually one switch in your hosting or CDN settings.
• FaviconTechnical SEOThe favicon is the small icon in browser tabs and in Google results. A missing one is a small detail, but it makes a site look less finished.
• Language attributeTechnical SEOThe lang attribute tells browsers, screen readers and search engines which language the page uses. It is a small fix.
• Meta descriptionsTechnical SEOA meta description is the summary that can appear in search results. It does not affect AdSense approval much, but it helps people choose your page.
• Social sharing tagsTechnical SEOOpen Graph tags control the title and image shown when your page is shared on social media. They do not affect AdSense, but they help your links look trustworthy.
• Structured data errorsTechnical SEOBroken JSON-LD is ignored by search engines. It does not affect approval much, but it means your markup is not doing its job.
• About pageTrust & required pagesNo Google rule names an About page. But Google's policy forbids hiding who runs a site, and Product Experts often point to missing About pages in rejected sites. It also builds readers' trust.
• No easy way to refuse cookiesTrust & required pagesUnder the GDPR, refusing must be as easy as accepting. Several European data protection authorities have fined sites whose first screen offered only “Accept” or hid refusal behind the settings. Google-certified CMPs can show a reject option on the first screen.
• Publish datesTrust & required pagesDates tell readers whether information is current. Without them, a reviewer cannot tell an active site from an abandoned one.
• Site identity in structured dataTrust & required pagesOrganization or Person structured data tells Google who is behind the site. It is optional and has a small effect, but it supports transparency.
• Privacy policy linkTrust & required pagesVisitors and reviewers expect to find the privacy policy from any page, usually in the footer. A hidden policy does little for transparency.
• Terms of useTrust & required pagesGoogle does not require terms of use. Many bloggers add one because it is part of a complete, professional-looking site. Its effect on approval is small.
• Media plays with soundUser experienceAuto-playing video with sound is on the Better Ads Standards list of ad experiences that annoy users. Google Publisher Policies do not allow Google ads on screens that break these standards.
• JavaScript errorsUser experienceScript errors can break menus, forms and layouts. They often come from outdated plugins or broken ad code.
• Page sizeUser experienceHeavy pages are slow on mobile data and can cost visitors money. Large images are usually the cause.
• Browser checkUser experienceThese checks need a full browser. A failure here can be temporary, so it does not lower your score.
• Small text on mobileUser experienceText that is too small makes people zoom or leave. Around 16px body text is a good minimum on phones.
• ads.txt content typeVerification & ads.txtThe ads.txt standard asks for text/plain. Most crawlers still read the file, but a wrong type often means the server returns a web page instead of your file.
• ads.txt fileVerification & ads.txtGoogle says ads.txt is not required but highly recommended. It is also one of the ways to verify your site. Once you show ads, it tells buyers you are allowed to sell your ad space.
• ads.txt redirectVerification & ads.txtThe ads.txt standard expects redirects to stay inside your own domain. Google accepts one redirect to another domain, but other ad buyers may not.