AdSense Readiness Checker
For site owners who are about to apply for Google AdSense, or who were told their site has low-value content. Paste your website address and the checker counts your URLs, finds thin and copied pages, checks About, Contact and Privacy, ads.txt and empty ad boxes, then prints one row per URL with the exact words, FAQ or fix to add. It is not Google and it cannot approve a site.
How to use this tool
What to paste. Your home page address, or the page you plan to submit. The checker stays on that host, reads robots.txt, sitemap.xml and ads.txt, finds your About, Contact and Privacy pages, then samples up to 40 URLs from the sitemap — legal pages first, then a spread across folders. On a bigger site, press “Check the next URLs” in the report to add 40 more at a time, up to 200, in the same report.
What is counted. For every fetched URL: HTTP status, sitemap and robots status, title, H1, visible main-text words, unique words inside a template family, H2 headings, the primary keyword and how often it appears, FAQ questions and answer length, five article blocks, a visible author and internal links.
How to read the score. The 100 points come from eleven checks shown in the breakdown. 80–100 means the listed checks pass; 60–79 means fix the listed rows first; under 60 means do not apply yet. Read the rows marked Add content, Merge or Noindex before anything else.
The URL rule. The tool tells you how many URLs you have and which band that falls into. It never asks you to add pages to reach a number. The goal is more strong URLs and fewer thin or duplicate ones.
One filled sample row
| URL | Words | Keyword | FAQ | Action |
|---|---|---|---|---|
| /guides/resume-tips | 640 / floor 900 | “resume tips” × 2 · missing in H1 | 0 questions | Add content: about 260 words, 4 FAQ answers, one worked example. |
| /resume-templates/ember | 1,100 total, 180 unique | same title pattern as 45 siblings | FAQ copied | Noindex. Do not rewrite all 46 in one pass. |
Limits
The check reads the HTML your server sends, at most 45 requests and 40 full pages per round (a report can grow to five rounds, 200 pages), with a 12-second limit per request. JavaScript-only text, pages behind a login and anything outside your host are not measured. Sites that block unknown crawlers return “We could not read this site. This is not a verdict.”
Manual checklist before you click “Apply”
Ticks stay on this page only. 0 of 9 done
Read the full walkthrough: How to apply for Google AdSense without a low-value content rejection →
Frequently asked questions
How this check works
1. Purpose
This is a pre-apply review for site owners. It looks for the patterns that most often sit behind an AdSense “low-value content” or “needs attention” message: thin pages, families of near-identical template pages, duplicate titles, missing policy pages, empty ad placeholders and public build notes. It cannot approve an account, it does not talk to Google, and it is not affiliated with Google. A reviewer at Google makes the decision.
2. What is fetched
The address you type, the home page if it is different, robots.txt, sitemap.xml (and one level of sitemap index), ads.txt, and your About, Contact and Privacy pages. About, Contact and Privacy are found from the links on your home page, then from the sitemap, and only then by trying one common path each. After that, the checker samples up to 40 URLs in total: legal pages first, then a spread across your folders. The “Check the next URLs” button reads the rest of the sitemap in rounds of 40, up to 200 URLs, and adds them to the same table and score. Every request stays on the same host, follows robots.txt, has a 12-second limit and a 2 MB size limit, and the whole check stops at 45 requests. Private networks, IP addresses and local hosts are refused.
3. How a word is counted
We take the HTML the server returns, remove scripts, styles, navigation, header, footer, aside blocks, forms and cookie or consent banners, then count the words in the main content area (the main element, else the article, else the body). A word is any run of letters or digits, so numbers count as words and non-Latin scripts are counted too. Text that only appears after JavaScript runs is not seen.
4. How a keyword is counted
The primary phrase is the part of the title before the brand separator (|, - or —). If the first part is your brand name, the next part is used, and long phrases are cut to their first five words. We count the phrase in the title, the H1 and the body text. The verdict is missing when the phrase is not in the H1, or appears only once on a page whose floor is 800 words or more; stuffed when a phrase of two or more words appears more than eight times on a page under 800 words (legal and listing pages excepted) or its density is above about 3 percent, when a single-word phrase passes about 5 percent, or when the page shows a visible “keywords:” list; otherwise natural. Density is shown for context. It is not a target to hit.
5. How URLs are counted
Total is the number of loc entries in your sitemap; without a sitemap, only fetched pages are counted and no URLs are invented. Indexable means in the sitemap, fetched, and not noindex. Noindex comes from the meta robots tag or the X-Robots-Tag header, and a noindex URL that is still in the sitemap is a clash. Thin and duplicate-title counts come from the measured rows; URLs over the cap are listed as not measured. The total then places the site in a band — 1–7, 8–25, 26–60, 61–120 or over 120 — with the sentence: you have N indexable URLs, M strong, K thin and D with a duplicate title; the target is not more URLs.
6. How a thin page is decided
A page is thin when its words are under the floor for its type: home 500, tool 800, guide or article 900, About 400, Contact 120, Privacy and Terms 500, a blog or listing index 250, and other pages 300. When three or more pages share over half of their five-word phrases, they form a template family. Inside a family, unique words are the words a page does not share with its siblings, and a copy with less than 40 percent unique text is thin even if its total is high. These floors are SocialKit review rules, not a Google quota.
7. How FAQ and article are detected
An FAQ is a visible list of questions: details and summary elements, accordion buttons, or headings ending in a question mark with an answer below. Schema alone does not count. Home and guide pages need four to six questions, tool pages five, with answers of about 40 to 80 words, and the same questions repeated across URLs are flagged as copied. The five article blocks — who it is for, steps, a worked example, mistakes, and method — are each marked yes or no from headings, ordered lists and the opening text.
8. What the score means
The 100 points come from eleven checks: HTTPS and the page loads (5), home page substance (10), About, Contact and Privacy (15), sitemap (5), URL band (15), sitemap versus noindex (10), titles (10), FAQ (10), author (5), ads.txt (5) and no empty ad box (10). A missing ads.txt still scores, because it is normal before approval. 80–100 reads “looks ready to apply — still not a guarantee”; 60–79 “fix the listed rows, then apply”; under 60 “do not apply yet”. Ready is not approved. Each fix is tagged as a proven blocker (seen directly), an inferred risk (a pattern) or advisory (good practice), and confidence depends on how much of the sitemap was measured.
9. What we cannot see
Your AdSense account history, policy strikes, invalid clicks, real traffic and its sources, domain age, pages behind a login, content added by JavaScript, and the reviewer’s decision. None of these are guessed.
10. What to do next
Fix the rows marked Add content first, using the exact sentence in the “Add this” column. Noindex only the template copies and language mirrors, and remove them from the sitemap. Remove empty ad boxes and public notes. Then rescan. If a review is already running in AdSense, do not send another request; wait for it to finish.
11. Sources
Read the rules at the source: AdSense Program policies, Google Search Essentials and Google spam policies. This page summarises; it does not quote those documents.