Listed players
SST2.44▼ -7.58%TIG40.00▲ +3.90%TEAD0.56▲ +3.77%PERI8.50▼ -2.97%TBLA3.23▼ -2.71%INUV0.57▼ -1.74%AV10.06▼ -1.59%GOOGL343.50▲ +1.56%SNAP5.58▼ -1.24%PINS19.26▼ -1.03%MSFT517.53▲ +0.92%PPLI41.28▲ +0.81%IOS32.24▲ +0.44%META728.08▲ +0.30%GDDY97.21▲ +0.24%DV13.49▲ 0.00%MCHX1.29▲ 0.00%
Ticker byClearTrust

Reports & studies

18 studies, standards and rulings that shaped the industry, each summarised in our own words with a link to the original.

Google · 2026 · Policy · Free

AdSense for Search (AFS) policies

The rulebook for anyone showing Google search ads on their own results page. It is short, and almost every feed termination traces back to one of its clauses. It is a living page, so read the current version.

  • Google's policy says search queries must come from real user intent, such as typing in a search box, and must not be pre-populated by the publisher.
  • The policy forbids modifying the AdSense for Search code and limits where search boxes can be placed.
  • It prohibits paying or rewarding users for searching or for viewing ads.
  • The page sets an annual query cap for standard publishers, measured from 1 August to 31 July.
Google Ads Help · 2026 · The advertiser side · Free

Search partner network (SPN) announcements

Google's running log of changes to the Search Partner Network as advertisers see it. It is the primary source for the end of parked domains and for the new transparency tools.

  • Google states that parked domains stopped being an ad surface in the Search Partner Network on 10 February 2026.
  • It announced full placement reporting for partner sites in August 2025, covering Search, Shopping and App campaigns.
  • In April 2025 it introduced a pre-screen option using third-party exclusion lists from DoubleVerify, Integral Ad Science and Zefr.
  • In April 2026 it added an invalid-activity credit report that shows adjusted cost and clicks after credits for invalid traffic.
System1 (filed with the US SEC) · 2026 · Company filings · Free

System1, Inc. Annual Report on Form 10-K for fiscal year 2025

The annual report of the best-known listed search arbitrage company. The business description and risk factors explain the model, and its dependence on Google, more frankly than any marketing material.

  • System1 reports that 67% of its 2025 revenue came from its agreements with Google.
  • The filing says there is substantial doubt about the company's ability to continue as a going concern, with its credit facilities maturing in 2027.
  • It lists chargebacks for fraudulent or low-quality clicks, and clawbacks for bad traffic from Network Partners, as factors that reduce what Google pays.
  • The company says it owns and operates about 40 websites, including Startpage, Info.com, MapQuest, HowStuffWorks and CouponFollow.
  • It reported 2025 revenue of $266.1 million, down 23% on the year before.
Perion Network (filed with the US SEC) · 2026 · Company filings · Free

Perion Network Ltd. Annual Report on Form 20-F for fiscal year 2025

The annual report of a company that lost its main search feed. It documents the end of the Microsoft Bing agreement, the move to Yahoo, and how management now describes the risks of search syndication.

  • Perion states that its agreement with Microsoft Bing expired on 31 December 2024 and that from 2026 its search business operates primarily with Yahoo.
  • It reports that search advertising revenue fell 44%, from $162.7 million in 2024 to $91.0 million in 2025, about 21% of total revenue.
  • The filing warns that search providers can change their policies on their own, and have done so in the past.
  • It names AI chat tools as a risk that could reduce traditional search use faster than the company expects.
  • It discloses a securities class action over its past statements about the Microsoft Bing partnership.
Team Internet Group · 2026 · Company filings · Free

Team Internet Group plc Annual Report 2025

The annual report of the London-listed owner of Tonic, ParkingCrew, Zeropark and Voluum, covering the year Google ended ads on parked domains. The figures below are as reported in the company's results and trade coverage of them.

  • The company reported that gross revenue fell 40% to $481.9 million in 2025 and net revenue fell 27% to $136.2 million.
  • Domain Name Wire's summary of the results says revenue per thousand visits in the monetisation business fell 51%.
  • The company describes a move of its search business to Google's Related Search on Content product.
  • It is running a strategic review and has said it expects its domains division to be valued at more than $160 million.
Inuvo (filed with the US SEC) · 2026 · Company filings · Free

Inuvo, Inc. Annual Report on Form 10-K for fiscal year 2025

A small-company view of the same business: ads supplied by Yahoo and Google, shown on owned and partner sites. Useful for seeing customer concentration stated in plain numbers.

  • Inuvo reports that two customers supplied 64.2% and 19.3% of its 2025 revenue, and that one customer supplied 75.0% in 2024.
  • The filing names Yahoo and Google as long-standing key relationships under multi-year service contracts.
  • It says the company buys keyword advertising, mainly on Google and Facebook, to bring traffic to its own websites.
CHEQ · 2025 · Policy · Free

RAF Is Here: What Google's RSOC Changes Mean for Traffic Quality

A vendor explainer of Google's Restricted Access Features for RSOC, written by a company that sells traffic screening to monetisation platforms. Read it for the mechanics, and remember the author has a product to sell.

  • CHEQ reports that from 25 August 2025 Google reserves advanced RSOC features for publishers that consistently meet quality and compliance standards.
  • It describes the change as a shift from rewarding volume to rewarding traffic quality.
  • Domain Name Wire separately reported that restricted publishers are limited to five suggested terms per block and one related-search block per page.
Domain Name Wire · 2025 · Feeds · Free

2025 in Review: The death of AdSense for Domains

A trade-press account of how Google wound down ads on parked domains during 2025 and what it did to the companies that depended on them. It is the clearest timeline of the event.

  • The review says Google stopped opting new advertiser accounts into parked domains in September 2024 and announced in February 2025 that it would opt out all existing advertisers.
  • It reports that a final wave in September 2025 cut parking revenue to a small fraction of earlier levels.
  • It notes that Sedo's revenue fell 66% in the third quarter and that both Sedo and Team Internet businesses were put up for sale.
  • According to the review, companies moved to Google RSOC, Yahoo's ad feed and zero-click redirects.
Infoblox Threat Intel · 2025 · Fraud & invalid traffic · Free

Parked Domains Become Weapons with Direct Search Advertising

Security research, published in December 2025, into what visitors to parked domains actually receive now that many are monetised by direct redirect and not by search ads.

  • Infoblox said that in its tests visitors to parked domains were sent to illegal content, scams, scareware or malware more than 90% of the time.
  • The researchers found that the redirects profiled each visitor, and often showed harmless pages to VPN and data-centre addresses.
  • They described operators holding thousands of look-alike domains, including a misspelling of a major email service set up to receive misdirected mail.
  • The report argues that Google's 2025 move to make parked-domain ads opt-in pushed domain owners towards riskier direct-search redirects.
Pew Research Center · 2025 · AI search · Free

Google users are less likely to click on links when an AI summary appears in the results

A study of the real browsing of 900 US adults in March 2025, comparing Google searches with and without an AI summary. Search arbitrage resells clicks, so fewer clicks per search matters directly.

  • Pew found that users clicked a traditional result on 8% of visits to pages with an AI summary, against 15% on pages without one.
  • Users clicked a link inside the AI summary itself on about 1% of visits.
  • People ended their browsing session after 26% of pages with an AI summary, against 16% without.
  • About 18% of the Google searches in the study produced an AI summary, and longer or question-style searches produced one far more often.
Ahrefs · 2025 · AI search · Free

AI Overviews Reduce Clicks by 34.5%

An SEO-industry analysis of 300,000 keywords comparing click-through rates before and after AI Overviews. It measures organic clicks, not ads, but it is widely cited as a size estimate for the effect.

  • Ahrefs estimated that the presence of an AI Overview was associated with a 34.5% lower click-through rate for the top-ranking page.
  • The study compared 150,000 keywords that showed an AI Overview with 150,000 that did not, using March 2024 and March 2025 data.
  • The authors compare AI Overviews to featured snippets, which also answer the query on the results page.
Adalytics · 2023 · The advertiser side · Free

Does a lack of transparency create brand safety concerns for search advertisers?

The most detailed outside study of the Google Search Partner Network, the programme that pays arbitrage and partner sites. Published in November 2023, it catalogued where search ads were appearing beyond Google itself. Google disputed its conclusions.

  • Adalytics said it identified tens of thousands of websites carrying Google search partner ads, where Google's documentation referred to hundreds of sites.
  • The report said search ads from major brands and US government bodies had appeared on pornographic, piracy and sanctioned websites.
  • It found that a large share of the partner sites it identified were parked domains, some of which sent visitors to pre-filled search queries with ads.
  • It noted that Google Ads did not then tell advertisers which partner websites their search ads had appeared on.
  • AdExchanger reported the firm's estimate that the partner network carried roughly $10.5 billion of ad spend a year.
Media Rating Council · 2020 · Fraud & invalid traffic · Free

Invalid Traffic Detection and Filtration Guidelines and Standards Addendum

The industry's reference definition of invalid traffic, first issued in October 2015 and updated in June 2020. It is where the terms general and sophisticated invalid traffic come from.

  • The MRC guidelines divide invalid traffic into general invalid traffic, which routine list-based checks can catch, and sophisticated invalid traffic, which needs advanced analysis.
  • They set out what a measurement company must do to detect and remove invalid traffic in order to be accredited.
  • The MRC issued a further interim update memo in April 2024, listed alongside the guidelines on its standards page.
US Federal Trade Commission · 2015 · Policy · Free

Native Advertising: A Guide for Businesses

The FTC's plain-language guide to when an ad that looks like content is deceptive. It applies to the native ads that buy arbitrage traffic and to landing pages that dress sponsored links as articles.

  • The FTC says an ad must not suggest or imply that it is anything other than an ad.
  • The guide says the closer an ad looks to the surrounding editorial content, the more a clear disclosure is needed.
  • It advises placing disclosures near the headline or focal point, and before the message in video or audio.
  • The FTC regards terms such as "Ad" or "Advertisement" as clear, and warns that terms such as "Promoted" may not be understood.
US Federal Trade Commission · 2013 · Policy · Free

Updated guidance to the search engine industry on distinguishing paid results

A letter from FTC staff to search engines, updating guidance first given in 2002. It is the regulatory basis for labelling sponsored listings, including on partner and arbitrage results pages.

  • FTC staff said search engines must clearly and prominently distinguish paid results and ads from natural search results.
  • The letter said the distinction had become less noticeable to consumers over time and asked the industry to correct this.
  • It said the same principle holds whatever form search takes, naming specialised search, social media, mobile apps and voice assistants.
Tyler Moore and Benjamin Edelman · 2010 · Fraud & invalid traffic · Free

Measuring the Perpetrators and Funders of Typosquatting

An academic study of misspelled domain names and who paid for them. It is old, but it remains the classic evidence that pay-per-click feeds funded typosquatting.

  • The authors identified at least 938,000 typosquatting domains aimed at the 3,264 most popular .com sites.
  • They found that about 80% of the typo domains they examined were supported by pay-per-click ads.
  • The study found that 63% of typo domains showing Google ads used one of just five advertising IDs.
  • It found typosquatting was more common in website categories where pay-per-click prices were higher.