/
Navigation
Chronicles
Browse all articles
Explore
Semantic exploration
Research
Entity momentum
Nexus
Correlations & relationships
Story Arc
Topic evolution
Drift Map
Semantic trajectory animation
Posts
Analysis & commentary
Pulse API
Tech news intelligence API
Browse
Entities
Companies, people, products, technologies
Domains
Browse by publication source
Handles
Browse by social media handle
Detection
Concept Search
Semantic similarity search
High Impact Stories
Top coverage by position
Sentiment Analysis
Positive/negative coverage
Anomaly Detection
Unusual coverage patterns
Analysis
Rivalry Report
Compare two entities head-to-head
Semantic Pivots
Narrative discontinuities
Crisis Response
Event recovery patterns
Connected
Search: /
Command: ⌘K
Embeddings: large
TEXXR

Chronicles

The story behind the story

days · browse · Enter similar · o open

PatronView's owner details a year of fighting scrapers: 214:1 bot-to-human page loads, 35K Claude crawls per referred visitor, and Amazon's bot referred none

A year of fighting scrapers on my 1.5 million-page website.  What I tried, how I failed, and what's working now.  August 7, 2026

PatronView Nick Gray

Context & Ripple Effects

PatronView's figures put a concrete workload behind a conflict publishers had already been trying to manage through robots.txt restrictions on AI bots and newer anti-scraping tools. The reported failures matter because a 1.5 million-page site offers crawlers a large surface area to revisit.

The traffic imbalance also sharpens the case for mechanisms such as Cloudflare's Pay per Crawl launch, which paired crawler charging with default blocking for new sites. PatronView reports that Amazon's bot sent no referrals, while Claude generated 35,000 crawls for each referred visitor.

First-order effects

  • PatronView must absorb or block bot activity that it says outnumbers human page loads 214 to 1, making scraper mitigation an operational priority for its owner.
  • Claude's reported crawl-to-referral ratio and Amazon's lack of referrals leave PatronView with little visible audience return for the bot traffic it identifies.

Second-order effects

  • AI crawler operators face stronger incentives to provide publishers with clearer controls and measurable exchange for access as sites compare crawl volume with referred traffic.
  • Anti-scraping and access-control providers gain a more immediate use case from publishers whose robots.txt rules or earlier defensive attempts have not adequately constrained crawlers.

Third-order effects

  • If publisher reports repeatedly show heavy crawling without referral value, access to web content is likely to move toward enforceable publisher controls and paid or negotiated crawler access rather than voluntary bot rules alone.

The trend: Publisher-AI search relations are shifting from permissive crawling toward verifiable access controls tied to the value crawlers return.

Discussion

  • @nickgraynews Nick Gray on x
    I was looking at my website server logs and was shocked to see that 99% of my traffic is from bots Anthropic scrapes my site 35,000 times for every 1 visitor they send I researched it more and wrote up all the Cloudflare and firewall rules I use here https://patronview.com/...
  • @nickgraynews Nick Gray on x
    This went viral on HackerNews today! I'll pull the stats and share after I make a matcha
  • @kaka_ruto Kaka Ruto on x
    Interesting experiment - Nick's website denied my custom web fetch tool access, but my agent decided to use the chrome devtools mcp and it worked There's a crawling pandemic going on and I'm pretty sure someone will come up with a solution for this - and it's not cloudflare [imag…
  • @nickgraynews Nick Gray on x
    I'm working on a new article: 99% of My Website Traffic Is Bots If you're trying to steal my data: start here! I'll tell you exactly what firewall rules I have in place so you can work around them more easily 😆 https://patronview.com/...
  • @bankuei @bankuei on bluesky
    This is a technical view of stuff this guy has to do to keep AI scraper bots from eating up all of his data and slowing his site to unusable.  —  This is a different way AI works to silence voices by putting a cost on sites.  —  patronview.com/news/99-perc...