The Google SerpApi Lawsuit: What Happened, and What It Means for SEO Tools
Google sued SerpApi in December 2025 over scraping search results. A judge dismissed the core claims in July 2026 and Google refiled. The full timeline.
Google sued SerpApi on 19 December 2025, alleging it scraped and resold Google search results. A federal judge dismissed the core of the case on 20 July 2026. Google refiled on narrower grounds on 10 August, and SerpApi moved to dismiss again on 25 August. The case is unresolved.
- Who is suing whom: Google filed against SerpApi, an Austin company that sells Google search results as structured data through an API, in the U.S. District Court for the Northern District of California.
- The legal claim: Google sued under the DMCA's anti-circumvention provisions, alleging SerpApi bypassed its anti-bot systems using cloaking, rotating bot identities and large bot networks, and resold licensed content such as Knowledge Panel images.
- SerpApi's defence: SerpApi argues it reads only pages any browser can load, that the DMCA protects copyrighted works rather than advertising revenue, and that Google does not own the content its results describe.
- The July ruling: Chief Judge Yvonne Gonzalez Rogers dismissed Google's claims covering results with no copyrighted content, with prejudice. She gave Google 21 days to amend on the narrower question of licensed content in Knowledge Panels.
- Where it stands now: Google's amended complaint of 10 August 2026 relies on its content licensing agreements, including Reddit terms. SerpApi's second motion to dismiss, filed 25 August, was pending as of 1 September. SerpApi reports no change to its service.
- Why SEO teams should care: Most rank trackers buy SERP data from providers like SerpApi. Google removed the num=100 parameter in September 2025, which raised the cost of collecting 100 results roughly tenfold. Search Console data is unaffected by any of this.
No injunction has been issued and no tool has stopped working. The trend in costs and access, however, has been running one way since early 2025.
This is a report on an unresolved case, with the dates and filings as they stand on 1 September 2026. CrawlRaven works from data Google and your own site provide directly: it joins Search Console, GA4 and a 200-point crawl into one ranked plan, from $49 at launch. Try CrawlRaven free: 1 site, no credit card →
Google sued SerpApi on 19 December 2025, accusing the company of scraping Google search results at scale and reselling them. A federal judge dismissed the core of that case on 20 July 2026.
Google refiled on narrower grounds on 10 August, and SerpApi moved to dismiss again on 25 August. As of 1 September 2026 the case is unresolved, and SerpApi's service continues to operate.
This report covers:
- What Google filed, and the specific conduct it alleges.
- How SerpApi answered, and what the court decided in July.
- What changed in Google's amended complaint, and what SerpApi says is wrong with it.
- The access restrictions that preceded the lawsuit, including the removal of the num=100 parameter.
- What the case means for rank trackers, keyword tools and AI search tools, and for your own reporting.
What Google filed
Google filed suit in the U.S. District Court for the Northern District of California. The defendant, SerpApi, is a company based in Austin, Texas that fetches search results pages and sells them as structured JSON through an API. Its customers are mainly other software companies rather than end users.
The claim was brought under the anti-circumvention provisions of the Digital Millennium Copyright Act. Google announced the action publicly on the day it was filed, describing SerpApi's business model as “parasitic”.
Section 1201 of the DMCA makes it unlawful to bypass a technological measure controlling access to a copyrighted work. It is normally applied to things like DRM on films. Google argued that its anti-bot systems are such a measure and that its results pages are such a work.
What Google alleges SerpApi did
Google's allegations fall into four groups. All of them are contested.
- Bypassing technical protections. Google says SerpApi circumvented its anti-bot systems and ignored standard crawling controls rather than respecting them.
- Cloaking and rotating identities. Google says SerpApi gave its crawlers false and frequently changed names, and spread requests across large networks of bots, so that blocks did not hold.
- Reselling licensed content. Google says SerpApi took material Google licenses from partners, including images in Knowledge Panels and real-time data in Search features, and charged customers for access to it.
- Overriding site directives. Google says the practice disregards the access preferences websites express through crawling controls, leaving those sites, in its words, “no choice at all”.
Google also stated that this activity had increased over the past year. It did not publish figures for the volume of scraping it attributes to SerpApi.
How SerpApi responded
SerpApi's general counsel said the company provides “the same information any person can see in their browser”. The company filed a motion to dismiss on 20 February 2026.
That motion made three main arguments:
- The DMCA protects works, not revenue. SerpApi argued that Google's anti-bot systems protect its advertising business rather than any specific copyrighted work, which it said is outside what the statute covers.
- Google does not own most of what it shows. Search results describe other people's web pages, so much of a results page contains nothing Google holds copyright in.
- Public pages are not access-controlled. Citing the Ninth Circuit's decision in hiQ v. LinkedIn, SerpApi argued that reading a page any browser can load is not circumvention.
SerpApi also calculated that statutory damages on Google's theory could reach $7.06 trillion, which exceeds U.S. GDP. It offered the figure as evidence that the interpretation proves too much.
The July 2026 ruling
On 20 July 2026, Chief Judge Yvonne Gonzalez Rogers granted the motion. The court held that Google had not stated a valid DMCA claim over search results containing no copyrighted content.
That portion was dismissed with prejudice, meaning Google cannot refile it. The court allowed Google 21 days to amend a narrower claim covering results that do contain a copyrighted element, such as licensed text and images in Knowledge Panels.
Google v. SerpApi: the case timeline
December 2025 filing to the second motion to dismiss
Google Search stops returning results without JavaScript enabled. Scrapers using simple HTTP requests have to move to full browser rendering, which costs more to run.
The parameter returning 100 results in one request stops working. Collecting the top 100 now takes ten requests. Search Console impressions fall across many sites at the same time.
Reddit alleges its content was scraped from Google search results and resold. The suit names SerpApi, Oxylabs, AWMProxy and Perplexity.
Filed in the Northern District of California under the DMCA. Google alleges cloaking, rotating bot identities and large bot networks, and calls the business model parasitic.
SerpApi argues the DMCA protects copyrighted works rather than advertising revenue, and that reading a page any browser can load is not circumvention. It cites hiQ v. LinkedIn.
Chief Judge Yvonne Gonzalez Rogers dismisses with prejudice the claims covering results with no copyrighted content. Google is given 21 days to amend a narrower claim.
The amended complaint rests on content licensing agreements, including Reddit terms barring third parties from extracting licensed content. Shopping and Maps claims are dropped.
SerpApi says Google has not filed the agreements it cites, that a privacy policy is not a copyright licence, and that the only two named images date from 1921 and 1951.
Dates from the parties' public filings and statements, checked 1 September 2026. The second motion to dismiss was pending at publication.
Google's amended complaint
Google filed an amended complaint on 10 August 2026. It drops the claims relating to Google Shopping and Maps and rests the remaining case on Google's content licensing agreements.
Google says those agreements oblige it to protect licensed material. It cites terms from its Reddit deal directing Google not to let third parties extract and commercialize the licensed content, and a second partner agreement, in versions dating to 2017, requiring commercially reasonable efforts to protect licensed material from outside access.
SerpApi's second motion to dismiss
SerpApi filed a second motion to dismiss on 25 August 2026, raising three objections:
- The agreements are not in the record. Google cites four sources, including two unnamed licensing agreements, but has not filed them with the court or quoted terms showing its anti-bot systems are authorised access controls.
- A privacy policy is not a licence. One of the four cited sources is Google's own privacy policy, which SerpApi says does not address automated access.
- The named images are old. The only two specific images Google identified are photographs of Babe Ruth from 1921 and Willie Mays from 1951, which SerpApi says are in the public domain or no longer protected.
The motion was pending as of 1 September 2026. SerpApi has said its service is unchanged and its customers are unaffected.
The first complaint asked whether scraping public search results is unlawful in itself. The court said no, and closed that route permanently.
The amended complaint asks a narrower question: whether Google's contracts with content licensors oblige it to stop third parties extracting that content. If that argument works, it does not depend on Google owning the results page, and any licensor could supply the same basis against any provider.
What happened before the lawsuit
The case follows a series of changes that made large-scale collection of Google results harder and more expensive. Each is documented and dated.
- January 2025: JavaScript required. Google Search stopped returning results without JavaScript enabled. Scrapers using simple HTTP requests had to move to full browser rendering, which costs substantially more to operate.
- September 2025: num=100 removed. Between roughly the 8th and the 10th, the URL parameter that returned 100 results in one request stopped working. Collecting the top 100 now takes ten requests.
- October 2025: Reddit sued four companies. Reddit filed against SerpApi, Oxylabs, AWMProxy and Perplexity, alleging its content was scraped from Google search results and resold.
- December 2025: Google filed. The lawsuit came after those technical restrictions were already in place.
The num=100 removal had a second, widely reported effect. Search Console impression counts fell sharply across many sites while clicks stayed flat and average position improved.
The common explanation is that automated result checks had been registering as impressions. On that reading, part of the impression growth many sites recorded through 2025 reflected bot activity rather than human searches.
Google is not alone in restricting data access. Google Trends changed its comparison limits in August 2026, and Microsoft retired parts of the Bing Webmaster Tools API this year.
How SERP data reaches your tools
The case matters to SEO teams because of how ranking data is sourced. Few tools that display a position collected it themselves.
How SERP data reaches your rank tracker
Four layers. The case targets the second.
Where the ranking exists. Protected by anti-bot systems, a JavaScript requirement since January 2025, and no bulk num=100 fetches since September 2025.
Each restriction added here raises costs downstream.
SERP API providers and proxy networks fetch results at scale and resell them as structured data. This is the layer Google is suing.
Both the Google and Reddit lawsuits target this layer.
Rank trackers, keyword tools, volatility trackers and AI answer engines. Some run their own collection; many buy from stage two and inherit its costs.
Since num=100 was removed, tools have cut depth or raised prices.
The position chart in your report. Three layers sit between it and the results page, none of them under your control.
Upstream changes arrive as price rises or gaps in the data.
Search Console data does not pass through this chain. Google delivers it to verified site owners directly.
SerpApi is one provider among several. DataForSEO, Oxylabs and Bright Data operate in the same layer, and the larger SEO suites run their own collection infrastructure.
The legal question is not specific to one vendor. A ruling on whether anti-bot systems are DMCA access controls, or whether licensing agreements support claims against scrapers, would apply across the layer. We reviewed one of these providers in our review of the DataForSEO MCP server, which sells the same category of data by the API call.
What it means for each kind of tool
Exposure varies by where a tool sits and whether it buys or builds its data collection.
| Tool category | What it uses scraped data for | Exposure |
|---|---|---|
| Rank trackers | Daily positions for a set keyword list | Higher cost per keyword, reduced depth and refresh rate. Already observed since September 2025. |
| SERP API providers | The product itself | Direct legal exposure. This is the layer the case targets. |
| Keyword research tools | Volume estimates and SERP feature data | Costlier sampling, and more reliance on modelled rather than freshly collected figures. |
| Volatility trackers | Daily reshuffle scores across sampled results | Smaller samples if collection costs rise, which reduces reliability. |
| AI answer engines | Current sources for generated answers | Covered by the same licensing argument. Reddit's suit already names Perplexity. |
Tools built on data the platforms supply directly are not affected. That includes the Search Console API, the GA4 export, your server logs and a crawl of your own site.
The possible outcomes
The case could end in three ways. These are scenarios for planning, not predictions.
- SerpApi prevails again. The second dismissal is granted and the DMCA route closes. The technical restrictions remain in place regardless, since no ruling requires Google to restore num=100 or relax its bot defences.
- Google prevails on the licensing argument. Content licensors gain a basis for claims against data providers. Likely effects are consolidation among providers, higher API prices, and AI companies moving to direct licensing.
- The parties settle. Terms would probably be confidential, leaving no public precedent while providers reprice their own legal risk.
Costs to buyers rise in all three. The direction was set by the 2025 technical changes, which no verdict reverses.
What this means for your own reporting
Search Console is not part of this dispute. Google delivers it to verified site owners through the interface, the API and a free BigQuery export, with no third party in between.
For measuring your own site, it also has advantages over scraped positions:
- It reflects real searches. A tracker records the rank a datacenter browser sees for a chosen keyword. Search Console averages your position across actual impressions, on real devices and locations.
- It covers every query you appeared for. A tracked keyword list is a selection made in advance. The query report includes terms nobody thought to track.
- It is free at any volume. There is no per-keyword cost, so no incentive to narrow what you measure.
- It is unaffected by the case. No outcome changes how the data is delivered.
Scraped SERP data and Search Console data
What each source can and cannot tell you
Contested in federal court
Delivered by Google to verified site owners
Rising. Fetching 100 results has taken ten requests since September 2025
Free, including the API and the BigQuery export
Yes. The whole results page, including competitors
No. Only your own site's queries and pages
Only the keywords chosen in advance
Every query the site appeared for, including untracked ones
Rank seen by a datacenter browser at check time
Average position across real impressions
Subject to rulings, blocks and price changes
16 months by default, permanent once exported
Competitor visibility is the one thing scraped data provides that Search Console does not. The other rows favour the data Google supplies directly.
The main limitation to plan around is retention. Search Console keeps 16 months of history by default and drops older days permanently.
Our guides to extending Search Console data retention and setting up the BigQuery bulk export cover how to preserve it. The GSC regex generator splits the query report into branded, non-branded and question segments.
What Search Console does not show
Search Console reports only your own site. It does not show a competitor's ranking, a results page you do not appear on, or which page outranks you.
Those are legitimate questions, and scraped data is currently the only way to answer them. The practical division is:
- Use owned data for your own performance. Clicks, impressions and position trends come from the source that measures real searches.
- Use scraped data for competitive questions. Size the keyword list to the questions you actually ask rather than tracking by habit.
- Reconcile toward Search Console. Where a tracker and Search Console disagree about your own performance, the tracker is sampling and Search Console is reporting what happened.
What to do now
None of this depends on the ruling. Each step is worth taking either way.
- Enable the BigQuery export. It collects data only from the day you turn it on, so delay costs history permanently.
- Review your tracked keyword list. Lists built when collection was cheap are where price increases will be felt first.
- Ask your vendors where their SERP data comes from. Whether they collect it themselves or buy it determines how exposed your reporting is to one supplier.
- Base recurring reports on owned data. Position from Search Console, sessions from GA4, technical findings from your own crawl, with scraped competitive data labelled as such.
- Watch impressions for the next change. The num=100 removal showed up first as a drop in impressions. That report is also where striking distance keywords come from.
Key Takeaways
- →The filing: Google sued SerpApi on 19 December 2025 in the Northern District of California under the DMCA's anti-circumvention provisions, alleging cloaking, rotating bot identities, large bot networks and the resale of licensed content.
- →The ruling: On 20 July 2026 Chief Judge Yvonne Gonzalez Rogers dismissed with prejudice the claims covering results with no copyrighted content, and gave Google 21 days to amend on licensed material in Knowledge Panels.
- →Current status: Google's amended complaint of 10 August 2026 relies on content licensing agreements including Reddit's. SerpApi's second motion to dismiss, filed 25 August 2026, was pending on 1 September. No injunction has been issued.
- →The wider context: Google required JavaScript for Search in January 2025 and removed the num=100 parameter in September 2025, which raised the cost of collecting 100 results roughly tenfold. Reddit sued SerpApi and three others in October 2025.
- →Effect on tools: Rank trackers, keyword tools, volatility trackers and AI answer engines all draw on scraped results. Costs to buyers rise whether Google wins, loses or settles, because the 2025 technical changes are unaffected by the verdict.
- →Effect on your data: Search Console is delivered directly to verified site owners and is outside the dispute. It reports every query your site appeared for, measured from real impressions, but never shows competitor rankings.
Sources used in this post
Dates and filing details come from the parties' public statements and the trade press reports listed below, checked on 1 September 2026. The second motion to dismiss was pending at publication and this post will be updated when the court rules.
- Google: why we're taking legal action against SerpApi
- Search Engine Land: Google sues SerpApi over scraping and reselling Search data
- Search Engine Land: SerpApi moves to dismiss Google scraping lawsuit
- SerpApi: the court granted our motion to dismiss
- Search Engine Roundtable: Google amends lawsuit against SerpApi
- SerpApi: moving to dismiss the claims a second time
- PPC Land: SerpApi asks judge to end Google's scraping case over three unseen contracts
Related reading on CrawlRaven
- GSC BigQuery Export: Setup, the Three Tables, and Your First Queries
- How to Keep Google Search Console Data Past the 16-Month Limit
- How to Find Striking Distance Keywords in GSC
- DataForSEO MCP Review: Maximum Data per Dollar
- Google Trends Now Compares 400 Terms
- Google August 2026 Spam Update: Timeline, Targets and Recovery
Frequently asked questions
What is the Google v. SerpApi lawsuit about?
Google sued SerpApi on 19 December 2025 in the U.S. District Court for the Northern District of California. Google alleges that SerpApi circumvented its anti-bot protections to scrape Google search results at scale and resell them, in violation of the DMCA's anti-circumvention provisions. Google also alleges SerpApi resold licensed content, such as images and real-time data shown in Search features. SerpApi says it accesses only publicly visible pages.
Has Google won the SerpApi lawsuit?
No. On 20 July 2026, Chief Judge Yvonne Gonzalez Rogers granted SerpApi's motion to dismiss. She dismissed Google's claims with prejudice to the extent they rested on search results containing no copyrighted content, meaning that portion cannot be refiled. She allowed Google 21 days to amend a narrower claim about results containing licensed material. Google filed an amended complaint on 10 August 2026 and SerpApi moved to dismiss again on 25 August 2026.
Is scraping Google search results illegal?
No court has ruled that it is, and the July 2026 decision in this case went the other way on the DMCA question. The court rejected Google's attempt to use the DMCA to control access to search results that contain no copyrighted content. Scraping can still create liability on other legal theories, including breach of contract and the licensing claims Google is now pursuing, so the question remains unsettled.
Will my rank tracker stop working because of this lawsuit?
There is no indication that it will. No injunction has been issued, and SerpApi says its service is running normally. The more likely effect is on cost rather than availability. Collecting the top 100 results already requires ten requests instead of one following Google's removal of the num=100 parameter in September 2025, and vendors have responded with shallower tracking depth, lower refresh rates and higher prices.
What is SerpApi?
SerpApi is a company based in Austin, Texas that fetches search engine results pages and returns them as structured JSON through an API. Its customers include rank trackers, keyword tools, researchers and AI companies that would otherwise need to build and maintain scraping infrastructure. In October 2025, Reddit sued SerpApi, Oxylabs, AWMProxy and Perplexity, alleging Reddit content was scraped from Google's results and resold.
Why did Google remove the num=100 parameter?
Google has not given a public reason. Between roughly 8 and 10 September 2025, adding num=100 to a search URL stopped returning 100 results, so collecting the top 100 now requires ten paginated requests. Two effects followed. The cost of large-scale rank tracking rose sharply, and Search Console impression counts fell across many sites while clicks stayed flat, which is consistent with automated bot views no longer being counted as impressions.
Is Search Console data affected by the lawsuit?
No. Search Console data is delivered by Google directly to verified site owners through its interface, its API and a free BigQuery export. It does not pass through any scraping provider, so no outcome in this case changes how you receive it. The limitation of Search Console is scope rather than legal risk: it reports only your own site's queries, impressions, clicks and positions, and never shows competitor rankings.
How does this case affect AI search tools?
AI answer engines are significant buyers of scraped search data, because a results page is an efficient way to ground a generated answer in current sources. Reddit's October 2025 lawsuit made that link explicit by naming both SerpApi and Perplexity. If Google's licensing argument succeeds, the reasoning would apply to any provider reselling Google results to AI companies, a larger market than rank tracking.
15+ years of growing SaaS websites through SEO | Author, 200-Point Audit Checklist
Aditi has spent 15+ years helping SaaS companies scale organic traffic through technical SEO and content strategy. She is the author of the CrawlRaven 200-Point Audit checklist used by agencies and in-house teams to systematically improve search performance.