Someone Scraped Your Site: How to Detect It and Protect Your SEO

You search for a sentence you wrote and it shows up on another domain, with a more recent date than yours. That is copied content, and it is not a ghost: it is scraping. Before you assume a penalty, it pays to know what Google actually says and which levers work.

What Content Scraping Is and Why It Ends Up on Another Domain

Scraping is the automated extraction of content from a site to republish it somewhere else. It is not an attack aimed at you: it is a business model. Scrapers run hundreds of domains from the same template, monetized with ads or affiliate links, and they need volume: they get it by copying whoever already researched, wrote and ranked. Your article is one more line in their inventory.

The Three Ways Your Content Shows Up Off-Site

• Verbatim copy: the whole text as written, sometimes with your images hotlinked from your server.
• Rewritten copy: synonyms swapped, paragraphs reordered or machine translated. The hardest to catch by exact phrase.
• Republished feed: if you publish your full RSS, you are handing over the article ready to copy.

What You Actually Lose (and Do Not)

You lose traffic, attribution and control: the reader lands on another site and never sees your brand. What you do not lose is your ranking simply because someone copied you. Duplication inside your own domain is a different matter, already covered in the impact of duplicate content on SEO.

What Google Actually Says About Copied Content

Copying Is Spam: The Scraped Content Policy

The Google Search spam policies, updated in September 2026, include a section on scraped content. It defines scraping as taking content from other sites, often automatically, and hosting it to manipulate rankings. Its examples: republishing someone else's content without added value or attribution, copying it with synonym swaps or automated techniques, and reproducing full feeds. The closing line is the key one: if you host that content, exclude it from Search. The policy speaks to the copier, not to the copied.

Being Copied Is Not a Penalty Against You

Google made that clear in a classic blog post, Demystifying the duplicate content penalty: duplication does not call for action unless the apparent intent is to deceive. It is a 2008 text, not this year's guidance, and it is still published as a reference. There is no manual action waiting for your domain because someone copied it. What does exist is competition: if the scraper has more authority or indexed first, it can outrank you. That is a problem of signals, not punishment.

The Documented Lever: Copyright Removals and Demotion

This is the part almost nobody explains. On the same page, under practices that can lead to lower rankings, Google describes content removals for legal reasons: when it receives a significant volume of valid copyright removal requests about a site, it may use that fact to demote other content on that same site. A single complaint protects nobody; a sustained pattern of valid removals against a serial scraper is a signal Google documents.

How to Find Out You've Been Scraped

Exact Phrase and the site: Operator in Fifteen Minutes

Pick a distinctive sentence from your article, put it in quotes and search Google with -site:yourdomain.com to exclude your own pages. Then try site:otherdomain.com "your sentence" to see whether the copy is indexed. Repeat with a second sentence: anyone rewriting with synonyms keeps some of the original wording, and one surviving sentence gives them away. The commands are explained in Google search operators.

Alerts, Brand Monitoring, and the Feed You Give Away

Set alerts for your brand and for two or three distinctive sentences from your most visited articles. Cut your feed down to an excerpt: the scraper will keep copying, but it will have to crawl the page and leave a trace.

Server Logs: The Bot's Trail

Logs show which user agents request hundreds of URLs in minutes from the same IP: that is the scraping signature. This server log analysis guide explains how to separate legitimate crawling from abuse. A log entry is a clue, not proof: the user-agent string can be forged.

And do not expect help from Search Console: it reports on your own site, has no report on who copied your content and never warns you when another domain publishes your text.

When the Scraper Outranks You

Why Disavowing Links Doesn't Help Here

Disavowing links works on inbound links to your domain, the ones pointing at you that you want to neutralize. A copy hosted on another domain is not a link to you: it is another page, and the tool never touches it. If you are cleaning up a toxic backlink profile, the method is in how to audit your backlinks and when to disavow links.

Strengthening Authorship, the Date and the Brand

The scraper's date is usually later, though sometimes it is faked. Your advantage compounds: indexing history, links pointing at you, brand mentions and identifiable authorship. They can take the text; they cannot take your history. The approach is entity SEO.

The Mistake of Blocking Google "to Protect Yourself"

Blocking Googlebot in robots.txt or adding noindex does not stop the scraper: it stops your visibility. The bot that copies you ignores robots.txt and keeps going while you disappear from the results. If you already did it, the reversal is in how to remove or deindex content.

How to File a Copyright Removal Request With Google

What Can Be Claimed and What Can't

You claim the substantial reproduction of your own work: text, images, video, design. You do not claim an idea, a topic or the fact that someone covers the same subject. The copyright removal FAQ lists rejected requests, including a driving school that asked Google to remove a competitor's homepage.

URL by URL, With Evidence

The official form is the Search content removal form. You submit one request per infringing URL, naming the original and the copy. Evidence of authorship helps: dates in your CMS history, timestamped screenshots or a registration if you have one. A bulk, careless submission gets rejected; ten precise ones work.

What Happens Next (and What Google Won't Do)

Google removes the search result for the reported URL, not the scraper's page: their site stays up, it just stops appearing in results. The site owner can file a counter-notice and the link can be restored if the case meets DMCA requirements. There is no official response time anyone can promise. If the case involves derivative work, licensing or syndicated content you have permission to use, talk to a lawyer before filing.

How to Prevent Scraping Without Losing Rankings

robots.txt: It Stops the Polite, Not the Rest

A well-written robots.txt keeps out scrapers that identify themselves and follow rules; the rest ignore it. It is a cheap filter, not a defense. Rules and common mistakes are in how to create a correct robots.txt in 2026.

Google-Extended and AI Crawlers

The list of Google's common crawlers explains that Google-Extended controls whether crawled content can be used to train future Gemini models, and states that it does not affect Search inclusion or ranking. Other AI providers publish their own user agents and compliance is voluntary: Google does not control them. More context in making your website agent-ready.

Rate Limits and What Doesn't Work

Per-IP rate limits, temporary blocks on request bursts and CDN bot protection: server-side measures that do not rely on the bot's good faith. Text obfuscation, on the other hand, is stripped in seconds and breaks accessibility, IP blocking punishes users behind shared addresses, and anti-copy scripts fail in reader mode.

Checklist: From Suspicion to Action in 8 Steps

1. Confirm the copy with an exact phrase and the site: operator.
2. Record the infringing URLs and the real date of your original publication.
3. Check whether the scraper already outranks you for your main queries.
4. Gather authorship evidence: CMS history, screenshots, a signed byline.
5. File one removal request per URL, not a generic complaint.
6. Log every submission and repeat it for each new copy.
7. Tighten robots.txt, rate limits and your feed to an excerpt.
8. Strengthen authorship, entity signals and your own links: what a scraper cannot replicate.

Conclusion: The Best Defense Is Being the Source

Copied content hurts because of the traffic you lose, not because Google condemns you. The scraped content policy punishes the copier, copyright removals strip the infringing result and, accumulated against one domain, are a signal Google documents as a cause of demotion. None of that replaces what cannot be transferred: being the recognizable original, with clear authorship, history and brand.