Fixing the AdSense "Low Value Content" Rejection

"Low value content" means Google's reviewers looked at your site and decided it does not yet offer enough original, useful material to carry ads. It is a verdict about the site as a whole, not about one page, and it is fixable. Most sites that clear it do so by cutting pages, rewriting what is left and then waiting for a fresh review.

There is no word count that gets you out of it and no published list of pages to add. What Google does publish is a set of policies and a short list of causes, and between them they point at a fairly clear repair job. This guide goes through what the message means, what usually triggers it, how to audit your own content in an afternoon, and when to ask for a second look. If you have not applied yet, the AdSense approval checklist covers the groundwork; this one picks up after the rejection.

What the message actually says

In your AdSense account the site shows up under Sites with a status of "Needs attention", and the reason given is low value content. Google's own help page on what to do when your site isn't ready to show ads lists four kinds of problem: the ad code is missing or incomplete, the site cannot be reached, there are policy violations, or there is not enough unique content or the user experience is poor. Low value content sits in the third and fourth group, depending on what the reviewer saw.

The rule behind it is in the publisher policies, under inventory value. Google does not allow its ads on screens "without publisher-content or with low-value content", on screens that are under construction, or on screens used for alerts and navigation. The policy page on ads on screens without publisher content gives dead-end pages such as thank-you, exit and error pages as examples. It also names automatically generated content that has had no manual review or curation.

Two related rules matter just as much. One bans ads on pages with replicated content, meaning material copied from elsewhere with no added commentary or curation. The other bans pages where ads and paid promotion outnumber the publisher's own content. We come back to both below.

Google also asks you to confirm the ad code is on the site and to use Request review only after you have fixed things. Its help page warns that submitting before that is likely to delay approval. That one line is worth taking seriously, because every pointless resubmission costs you time rather than saving it.

What reviewers are probably looking at

Google does not publish a scoring sheet, so anything beyond its own wording is inference. What the help pages do say is consistent, though.

The rejection help page says reviewers may check all of your pages, and the sign-up page is just one of them. It describes the low-value problem as a site that lacks "original, rich content that would be of value to users". And it separates this from insufficient content, which is a site with too little text or one that looks unfinished.

Put together, a reviewer is asking something like: if a visitor landed here from a search, would they find an answer they could not get from the next ten results? Is it obvious who runs the site and why it exists? Does it look maintained?

That framing lines up with Google's guidance on creating helpful content, which is written for search but asks the same questions. Does the page offer original information or analysis? Is it a substantial, complete treatment of the topic? Is the author clear, with background on who they are? Would you bookmark it or recommend it to someone? Run your own pages against that list and be honest.

The usual causes

After reading the official pages and looking at what gets flagged, the causes sort into a handful of patterns. Most rejected sites have two or three at once.

Thin pages. Posts of a few hundred words that restate what any search result already says. There is no magic word count (our own rule of thumb for triage is to look hard at anything under about 300 words, but that is a sorting aid, not a Google rule). A 250-word page that answers a narrow question completely can be fine. A 1,500-word page padded with filler is not.

Copied or lightly reworded content. Google's replicated content policy lists mirroring, framing and scraping, but also sites that copy content and "lightly modify it", for instance by swapping in synonyms or using automated techniques. Running another site's article through a paraphrasing tool is the classic example. Sites made mostly of embedded videos or images from elsewhere, with nothing added, fall under the same rule.

Mass-produced text. Google's spam policies define scaled content abuse as many pages generated mainly to manipulate rankings, and one of their listed examples is using generative AI tools to produce many pages without adding value. The wording is method-neutral: it applies to unoriginal content "no matter how it's created". Google's page on generative AI content does not ban AI assistance. It says the problem is scale without value, and that anything published should be fact-checked and reviewed by a person, because generative models predict likely words rather than retrieve facts.

Affiliate-heavy pages. Google's guidance for rejected applicants says affiliate content "should form only a minor part of the content of your site" unless it adds real features. A site of product roundups copied from merchant descriptions looks exactly like what that sentence describes.

A site that is mostly ads. The more-ads-than-content policy counts images, video, games, article text and user-generated content as publisher content. Whitespace, headers, footers and links to other pages on your site do not count.

Unfinished or empty areas. Placeholder posts, theme demo pages, empty categories, "coming soon" pages. Nobody planned for them to hurt, but a reviewer clicking through will land on them.

Weak navigation. Google's rejection help lists redirects, pages behind a login, broken links, excessive pop-ups and unfinished pages under navigation problems.

A content audit you can finish in an afternoon

The goal is a spreadsheet with one row per published URL and one decision per row. Keep it plain.

1. List every URL. Export from your CMS, or pull them from your XML sitemap. Include pages, categories and tag archives as well as posts.
2. Add four columns: word count, any search clicks in the last 12 months or so (Search Console's Performance report will give you this), whether the page is mostly original, and a decision.
3. Read the pages, not just the numbers. Open each one in a private window and ask whether it would help a stranger. This is slow and it is the only step that really matters.
4. Mark each page as keep, improve, merge or remove.
5. Make the changes in one batch rather than trickling them out across weeks, so the site a reviewer sees is the cleaned one.
6. Fix the plumbing afterwards: update the menu, redirect merged and removed URLs where it makes sense, and make sure no internal link points at a page you deleted.

Here is how the four decisions usually play out.

Removing pages feels wrong because it makes the site smaller. For this problem it is often the right move. A site of 30 pages that all earn their place reads better than 80 pages where half are filler.

One more option exists for pages you want to keep but cannot fix quickly: keep them off the review. Google's rejection help suggests removing the ad code from auto-generated pages and pages with little original content. That helps once ads are running on a site. For an unapproved site, the review looks at the whole thing, so cleaning or removing is the safer route.

Making the pages you keep worth reading

"Add value" is vague advice, so here is what it means in practice.

Replace summaries of other people's work with something only you can give. A recipe site can include the failed attempts and what changed. A software blog can show real screenshots, real settings and the error messages you actually hit. A local site can cover places the national sites ignore. Where you do rely on outside facts, name the source and link it, as the helpful content guidance suggests.

Tidy the structure. One clear topic per page, a title that matches what the page delivers, subheadings that help someone skim, and an intro that gets to the answer. Skip the throat-clearing paragraph that restates the title.

Put a person behind it. A real About page, a visible author name on posts and a working contact route are small things, but they answer the "who runs this" question quickly. Google's guidance warns specifically against fabricated author profiles, such as invented names or AI-generated headshots.

If you used AI to draft anything, treat the draft as a rough first pass. Check every claim, rewrite in your own voice, add what the model could not know, and delete anything generic. A page you could not defend line by line should not go live.

Finally, deal with the experience around the content. Slow, jumpy pages with pop-ups over the text read as low quality whatever the words say. The guide on ads and Core Web Vitals covers the speed side, and if you have visitors in Europe or the UK, set up your consent message before you request a review.

Navigation and structure fixes

Navigation is a smaller item on Google's list than content, but it is the quickest to repair.

• Make sure every page is reachable from the menu or a category within two or three clicks.
• Remove empty categories and tag pages that list a single post.
• Check header and footer links for 404s and redirect loops.
• Check the site on a phone. Overlays that cannot be closed and menus that do not open count against you.
• Take down anything that sends visitors to a login before they can read.
• Make sure the site loads without a password. Google's help page says AdSense must be able to reach it, and that a login can be removed temporarily during review.

A site that is easy to move around also tends to look finished, which reviewers notice.

A worked example

This is a hypothetical site, with the numbers chosen to be simple. A hobby blog has 48 published posts and was rejected for low value content. The audit turns up this split:

• 14 posts are under 300 words and have no search clicks. Decision: remove.
• 9 posts cover the same three topics from slightly different angles. Decision: merge into 3 stronger pages.
• 11 posts are on-topic but generic. Decision: improve with original examples and sources.
• 14 posts are already solid. Decision: keep.

After the batch of changes the site has 11 + 3 + 14 = 28 pages, every one of them with a reason to exist. That is a smaller site that is considerably easier to approve.

Is it worth the effort? Our benchmark estimates put AdSense page RPM in a general niche at about $7.00 per 1,000 pageviews from Tier 1 readers, and general content carries a 0.8 niche multiplier. A site with 10,000 monthly pageviews from Tier 1 works out at 10 x $7.00 x 0.8, roughly $56 a month. The 10,000 monthly pageviews page breaks the same sum down by network, and the AdSense revenue calculator lets you put in your own traffic and countries. Small numbers, yes. But the content work also helps search traffic, which is what grows the figure.

When to request a review, and how long it takes

Request a review when the cleaned-up site is live, not before. In your account go to Sites, open the site, confirm the AdSense code is in place, click Next, then Request review. Finish the required account setup tasks first. Google's page says that if you skip them the site is not reviewed and stays in "Getting ready". It also advises against removing and re-adding the site, because that can delay the process.

Google says the review "usually takes a few days, but in some cases can take 2-4 weeks". That is the only timeline it gives, and it is for the review itself.

There is no published cooling-off period before you may ask again. What follows is our opinion: give Google time to recrawl the pages you changed before you resubmit, which in practice means a few weeks, not a few days. If you can check in Search Console that your main pages have been crawled since the edits, so much the better. Resubmitting the same site with cosmetic changes just produces the same answer.

And do not set up a second AdSense account to get around a rejection. Google's help page says its policies allow one account per publisher.

If it fails again

A second rejection is information. Read the new message carefully, because the reason may have changed. If it is still low value content, the changes were not big enough, and the honest fix is usually more original material rather than more tweaking. Look at the five or six pages a visitor is most likely to see first and ask whether they would impress anyone who did not write them.

It can also be worth asking whether AdSense should be the first network at all. Some networks, such as Ezoic and Media.net, publish no fixed traffic minimum, though each has its own review. Our guide to ad networks for small websites lists the options, the AdSense vs Ezoic comparison shows the estimated difference, and AdSense alternatives covers when a switch makes sense. Affiliate links and sponsorships are another path while you build; see how to monetize a blog.

Remember too that approval is not the end of the policy story. The same inventory rules apply to approved sites, and pages can be flagged individually later. Our guide to common AdSense policy violations covers those, and why ad accounts get limited explains the early restrictions new accounts sometimes see.

Common questions

How many pages or words do I need?

Google gives no number. Its guidance talks about unique, valuable content and a site that feels complete, not about counts. Aim for a set of substantial articles on a clear topic and judge each one by whether it helps a reader.

Is AI-written content automatically rejected?

No rule says so. Google's published position is that mass-producing pages without adding value is the problem, however the text was made. Anything you publish should be checked by a person and carry something original.

Should I delete posts or add more?

Usually both, in that order. Remove or merge the weak pages first, so the next reviewer meets a cleaner site, then add new work that meets the bar you have just set.

Can I just remove the ad code from the weak pages?

For a site that has not been approved, treat the review as covering the whole site. Remove or fix the weak pages themselves rather than counting on a partial removal of ad code.

Your next seven days

Start with the export. List every URL and read at least the ten that get the most traffic, because those are the pages a visitor and a reviewer are most likely to see. Make your keep, improve, merge and remove calls, and apply them in a single session.

Then check the basics: an About page with a real name, a working contact page, a privacy policy, and a menu with no dead ends. If you want to set expectations for revenue before you start, the answer on AdSense pay per 1,000 pageviews and the general-niche estimates for AdSense show the range. Once the changes have been live and crawled for a few weeks, open Sites in your account and request the review.