You won, and the case still ranks, because the page that ranks was built from the filing and not from the outcome. Docket republishers copy the caption and the complaint early, then rarely come back. Google ranks what is on the page, and the page never changed.
Why does a case you won still show up on Google?
The result on page one is almost never the court’s own website. It is a republisher page: UniCourt, Trellis, PacerMonitor, CourtListener, Justia Dockets or DocketBird, each carrying its own copy of the same docket. Those pages are the ones optimized for a name search, and those are the ones that rank.
That is also why clearing one case is rarely one link. A single matter can generate six to twelve indexed URLs across republishers, plus a law firm write-up and a newsletter that syndicated the filing. Every one was written from the complaint, and most were never rewritten at all. Same pattern, more detail, in how to remove a civil lawsuit from Google search.
Why do republishers index the complaint and not the disposition?
Because filings are pushed to them and outcomes have to be pulled. Filing is the loudest moment in a case’s life: court RSS feeds broadcast new entries, CM/ECF notices go to everyone on the service list, and case metadata is available immediately. An ending is quieter. It is one text line on a docket sheet, and the order behind it often sits behind a fee.
PACER charges $0.10 per page with a $3.00 cap per document, with fees waived below $30 in a quarter. That is trivial for one lookup and meaningful at the scale of a republisher tracking millions of dockets. The Free Law Project describes the RECAP Archive as built substantially from what its users contribute: roughly 30,000 people running the RECAP extension, documents captured from ECF notification emails, a nightly pull of documents clerks have marked as free opinions, and ongoing RSS crawling. Nobody buys the dismissal order in a case nobody is watching. So the archive fills in at the front of a case and thins out at the back.
What is the crawl-timing problem?
There are two clocks between your win and your search results, the republisher’s and Google’s, and both slow down in the same direction. The republisher stops re-pulling a closed case with little traffic. Google crawls a page that never changes less and less often. Neither system is broken. The combined effect is that the accusation outlives the outcome.
| Stage | What moves it | What happens when your case ends |
|---|---|---|
| Court docket (CM/ECF or state case portal) | The clerk enters the disposition | Accurate the same day. This layer is fine. |
| Republisher ingest | RSS feeds, ECF notices, user uploads, scheduled re-pulls | May never re-pull a closed, low-traffic case |
| Republisher page | Rewritten only when ingest runs again | Still leads with the caption and the complaint |
| Googlebot crawl | Google says crawl demand turns on staleness and popularity | An unchanged page earns fewer recrawls over time |
| The search result | The last crawl of an unchanged page | Still the filing, indefinitely |
Why doesn’t Google’s Refresh Outdated Content tool fix it?
Because that tool does not judge whether a result is fair. It only compares Google’s stored copy against the live page. If the republisher page still says what it said, the request is denied. Google publishes the denial reasons, and the one that applies here is “Content still on page”: “The text or image that you described in your refresh outdated content request is still present on the live page.”
That single line explains why people burn weeks on Google forms and get nowhere. The work has to happen at the publisher. Google is downstream of a page it did not write.
Which republishers will act, and what does each one require?
They differ more than people expect. Three major operators publish a documented route. Two publish none, which does not mean they never act, only that there is no stated standard to hold them to. Note that most of what is on offer is de-indexing rather than deletion. That is still a real outcome: a page blocked from search engines stops costing you the deal.
| Operator | Documented route | What it requires | What it does |
|---|---|---|---|
| CourtListener / RECAP (Free Law Project) | Yes, a published content removal policy | Links to the specific pages. A court order to delete a document. | Will “generally block search engines from indexing pages on our site by using the robots HTML meta tag and/or the x-robots-tag HTTP header” |
| Justia Dockets | Yes, its “Why Is My Information Online?” page | A written request to block. A sealing or removal order to delete. | Blocks pages from search results using robots.txt. Warns it can only block its own database. |
| UniCourt | Yes, a Public Records Redaction Request | A party, guardian, fiduciary or attorney for a party. Documentation “such as court orders and police reports.” | Redacts or seals the record. States URLs are “generally de-indexed by the internet search engines within about 30 days.” |
| PacerMonitor | None stated in its published privacy policy | Not published | Handled case by case through direct contact |
| Trellis | None stated in its published privacy policy | Not published | Handled case by case through direct contact |
| Not the publisher | Its own removal policies, or a valid legal order | Refreshes a result only after the live page actually changes |
Does the shape of your win change what you can send?
Yes, and it surprises people. A clean win can leave you with weaker paper than a messy one, because operators want a document and some endings do not produce one.
Under Federal Rule of Civil Procedure 41, a plaintiff “may dismiss an action without a court order by filing: (i) a notice of dismissal before the opposing party serves either an answer or a motion for summary judgment.” That is the fastest possible end to a federal case and it generates no order at all, only a notice. Later in the case, Rule 41(a)(2) applies instead, and the action “may be dismissed at the plaintiff’s request only by court order,” which does give you a signed document. An involuntary dismissal under Rule 41(b) generally “operates as an adjudication on the merits,” which is a defense win with an order attached. A vacatur or a sealing order is stronger still, because it changes the record itself rather than describing an outcome.
State practice varies, and several states handle dismissal, sealing and expungement differently from the federal rules and from each other. The point holds everywhere: find out what document your ending produced before writing to anyone. If it produced none, say so plainly and send the certified docket sheet showing the case closed. That is what civil litigation removal work usually turns on.
What does it cost to clear a case you already won?
Removal is priced per link, typically $1,000 to $2,000 per URL, quoted after an eligibility check rather than before. We publish the number because almost nobody in this category does, and because the alternative is a quote that hides how many links are actually in play.
The eligibility check matters more than the price. One resolved case with a clean order and six republisher pages is straightforward. The same case ended by a Rule 41 notice, with a wire-service pickup in the mix, is not. More on the model is in what it actually costs to remove a link from Google and on the link removal page. Suppression is quoted per engagement rather than per link and costs more than a removal, which is the practical reason to test removal first.
What this does not do
Removing a link from Google does not delete the source page, and it does not touch the court record. The docket still exists, the clerk’s file still exists, and anyone who goes to the courthouse or pays for PACER will find it.
- It does not defeat a background check, a tenant screening report or a litigation search that pulls from courts and primary sources rather than from Google. Those are separate systems. Where a consumer report is inaccurate, the FCRA dispute process is its own route with its own rights.
- Credit reporting of judgments and liens is governed separately from search results. Clearing the search result does nothing to a tradeline or a public-record entry on a report.
- Any disclosure obligation you have survives a removal entirely. If an employer, lender, licensing board, regulator, counterparty or court is entitled to ask, the answer does not change because a link is gone. Nothing here is a way around that, and we will not take work framed that way.
- Removed content can reappear through scrapers, mirrors and syndication. Monitoring is the only recurring line we sell, and it exists for exactly this reason.
- Accurate news reporting, valid unexpunged convictions, genuine reviews and sex offender registry listings cannot be removed. Anyone telling you otherwise is selling something that does not exist.
We never guarantee an outcome. Removal depends on a decision by a publisher, an operator or Google, and none of them work for us.
Frequently asked questions
How long after a case closes should I wait before doing anything?
Waiting is the common mistake. No scheduled event will update a republisher page for you. If the page has not changed within a few weeks of your disposition being entered, it is unlikely to change on its own: the republisher has moved on, and Google has no reason to recrawl a static page.
Can I just ask Google to remove the result?
Not on these facts. Google’s refresh tool denies requests where the content “is still present on the live page,” and a docket page about a real case does not fall under its personal information removal policies. The route runs through the republisher first.
If one operator de-indexes the page, do the others follow?
No. Each operator holds its own copy and decides independently, which is why a single case is a multi-link job. Justia says this directly: it can only block webpages from its own database. The same logic applies across court record removal generally.
Does a sealing order force a private site to take the page down?
It is the strongest document you can hold, and it is still a request to a private operator rather than an automatic result. CourtListener states it will not remove a public document without a court order requiring it. What an order does is make the ask easy to grant.
Where to start
The first useful step is knowing how many links your case actually produced and which of them are realistically removable. That is what the Search Risk Audit is for: we map every indexed URL tied to your name and the case, sort them by removal route, and tell you what we think will and will not come down before you commit to anything.
Researched and reviewed by the Search Zapper editorial team. Published September 11, 2026. We describe mechanics, not legal advice.