Google Refiled Its Scraping Lawsuit on August 10. The Data Pipe It Wants Shut Feeds Your Dealership's Rank Reports.

On August 10, Google filed a 15-page amended complaint in federal court in Oakland against a company most dealers have never heard of. It landed on the last day of a 21-day window the judge had given Google to fix its case, and PPC Land read the filing line by line the next morning.

The defendant is SerpApi, a company that sells programmatic access to Google search results. The case is Google LLC v. SerpApi LLC, No. 4:25-cv-10826-YGR, in the Northern District of California, before Chief Judge Yvonne Gonzalez Rogers.

This looks like a plumbing dispute between two tech companies. It is closer to a fight over who is allowed to know where your dealership ranks, and whether the vendors selling you that answer can keep buying the data they build it from.

What Google is actually claiming

In January 2025, Google deployed an anti-scraping system called SearchGuard. Per the complaint, it throws JavaScript challenges at queries coming from unrecognized sources and requires the browser to hand back specific device and user information before it will serve results.

Google alleges SerpApi built workarounds within weeks. The filing quotes SerpApi CEO Julien Khaleghy describing the approach as "creating fake browsers using a multitude of IP addresses that Google sees as normal users." Google puts SerpApi's volume at hundreds of millions of artificial search requests each day, up by as much as 25,000% over two years, against annual revenue the complaint pegs at a few million dollars.

The legal hook is the DMCA's anti-circumvention provisions, 17 U.S.C. 1201(a)(1)(A) and 1201(a)(2), with statutory damages running $200 to $2,500 for each act of circumvention. Google alleges billions of them. You can do that arithmetic yourself against a few million dollars of revenue.

Google lost the first round. On July 20, Judge Gonzalez Rogers granted SerpApi's motion to dismiss. Claims covering search results with no copyrighted content were killed permanently, because the DMCA does not protect uncopyrighted work. Claims covering copyrighted content were dismissed with permission to refile, because Google had not shown that SearchGuard operated "with the authority of the copyright owner." The judge did reject SerpApi's broader arguments, finding the DMCA "reaches any person injured by a violation, not only copyright owners."

So Google had one hole to fill: prove somebody who owns copyrights told Google to lock the door.

The August 10 amendment fills it with contracts. Four new paragraphs describe licensing partners: an unnamed major licensee holding copyrights to millions of works, with agreement versions dating back to 2017, who the complaint says "insisted that Google do so"; a second unnamed provider that "requires that Google ensure the content it licenses will not be available for download by third parties"; and Reddit, whose agreement "directs Google not to enable third parties to extract and independently commercialize the licensed content."

Reddit is the only licensor Google named. SerpApi has not conceded anything. General Counsel Chad Anson said on August 11: "We remain confident in the strength of SerpApi's position and look forward to responding."

Why this is the second front, not the first

Ten days before the amended complaint, the same defendant lost a bigger argument in New York.

Reddit sued SerpApi, Perplexity AI, Oxylabs and AWMProxy in the Southern District of New York in October 2025. Its theory is that Perplexity did not scrape Reddit directly. It routed collection through Google search results, with SerpApi as the technical intermediary, in what Reddit calls industrial-scale harvesting of hundreds of billions of search results. On July 31, U.S. District Judge Paul A. Engelmayer let Reddit's main DMCA claims proceed, finding Reddit had plausibly pleaded a conspiracy between SerpApi and Perplexity.

Hold onto that defendant list, because one name on it turns up again in a minute.

Two federal courts, two plaintiffs, one defendant in common, and a shared idea forming underneath both cases: a page of Google search results is starting to look like a licensed transaction rather than a public bulletin board.

Dealers have watched this movie before

Every dealer who has ever fought to get a third-party tool connected to their DMS recognizes the shape of this. A platform owner controls the pipe. Third parties build integrations that the platform owner never authorized. The platform owner locks the pipe and calls it security. The third parties build workarounds and call it access. Then lawyers spend years deciding who was right, while the dealership sitting at the end of the pipe keeps paying for reports and hopes they keep arriving.

The difference is what is in the pipe this time. Last time it was your service history. This time it is the record of what the internet says about your store.

What this means for your dealership

Start with a question almost nobody asks a vendor: where does your rank data come from?

SerpApi's own homepage advertises structured access to Google Search, Maps, Local, Shopping and News, plus Yelp and TripAdvisor, and it lists its customers by name. Among them are Ahrefs and BrightLocal, two platforms whose output ends up in dealer SEO reports every month, and Perplexity, an answer engine your shoppers may be using right now. Published pricing tops out at $275 a month for 30,000 searches before you get to custom enterprise deals.

That is the supply chain. Your agency runs a local rank report. The platform behind that report buys SERP data wholesale. The wholesaler is currently defending itself in two federal courts against the two largest owners of the content it resells.

There is a second layer, and we documented it last week. When RESONEO captured ChatGPT's internal result_source field before OpenAI switched it off on July 21, they found ChatGPT fetching web results through commercial scraping vendors alongside its own index. On a paid account in thinking mode, 74.6% of results came through the Bright Data pipe. The four pipeline names in that data were labrador, bright, serp, and oxylabs.

That last one is the name to hold onto. Oxylabs is a co-defendant in Reddit's New York case. The AI answer a shopper reads about your dealership has commercially scraped search results somewhere in its bloodstream, and at least one of the companies supplying them is already in court.

Now put the two together. If Google's licensing theory works, the cost and legal risk of commercially reselling Google's results goes up for everybody in that layer. Three things follow for a franchised store:

Your rank reports get more expensive, thinner, or both. Nobody is going to send you an email announcing that their data cost tripled. It shows up as a price increase at renewal, or as a report that quietly covers fewer keywords, fewer competitors, fewer ZIP codes than last quarter.

Third-party measurement becomes less trustworthy at exactly the wrong moment. Dealers are being sold AI-visibility scores right now. We have already written about how statistically noisy those numbers are. A legal squeeze on the underlying data supply does not make them steadier.

Google's first-party data becomes the only source nobody can sue away. Search Console, Google Business Profile insights, GA4, and your own CRM are not scraped from anybody. Whatever happens in Oakland, that data still shows up Monday morning.

What to do about it

1. Ask every SEO, GEO, and AI-visibility vendor in your building, in writing, where their SERP data originates and what happens to your reporting if that source becomes unavailable. A vendor who cannot answer is reselling a dependency they have not examined.
2. Do not sign a multi-year rank-tracking or AI-visibility contract this quarter on the assumption that the data feed behind it is permanent. Shorter terms, or an exit clause tied to reporting coverage.
3. Make sure Search Console is verified and connected for every rooftop domain, and that somebody actually reads it. It is free, it is first-party, and it is the fallback if third-party rank data gets scarce.
4. Grade your marketing on outcomes you own. Leads, calls, appointments, VDP engagement, and closed deals do not depend on anyone's ability to scrape a search page.
5. Keep the GBP and inventory data clean anyway. Every scenario in this case, whoever wins, still routes shoppers through local records and structured inventory.

The bottom line

Nobody knows how this ends. SerpApi has beaten Google once already and says it will respond. Meredith Rose of Public Knowledge told Search Engine Journal that Google may have "talked themselves into a little bit of a corner" by disclosing how much copyrighted material sits inside its own Knowledge Panels. Discovery is frozen pending the next round of motions. This will take years.

The useful takeaway does not require predicting the verdict. Two federal judges are now treating Google's search results page as protected, licensed property, and a meaningful share of what your vendors know about your dealership's visibility is bought from a company being sued over exactly that. Reports built on borrowed data are worth what the loan is worth.

Build your scoreboard on the data you actually own, and put the real money into being the dealership that both Google and the answer engines want to surface in the first place. That part survives any ruling.

Want a clear read on how your store currently shows up in AI answers and search, using first-party data instead of a scraped dashboard? Book a walkthrough and we will go through it with you.