Scraping via Googlebot – How is it possible?

3 points by devx_ 3 hours ago

Hi,

I run a website that recently experienced unusually high traffic from what appeared to be legitimate Googlebot. After investigating the access patterns, I was able to identify the source through some creative analysis.

Background

Someone has been scraping my website extensively using what appears to be authentic Googlebot. I traced the activity back to the person responsible, and they revealed they're using a commercial API service that can trigger real Googlebot crawls on-demand.

Technical Details

I tested the service myself to verify their claims, and confirmed it does indeed dispatch legitimate Googlebot to any URL within 1–2 seconds.

Verified Googlebot IPs (via reverse DNS):

- 66.249.76.65 → crawl-66-249-76-65.googlebot.com

- 192.178.4.87 → crawl-192-178-4-87.googlebot.com

- 2001:4860:4801:002d::0006 → crawl-2001-4860-4801-002d...googlebot.com

- Additional IPs from 34.96.x.x range → googleusercontent.com

Request Headers:

- User-Agent: Mozilla/5.0 (Linux; Android 6.0.1; Nexus 5X Build/MMB29P) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/117.0.0.0 Mobile Safari/537.36 (compatible; Googlebot/2.1; +http://www.google.com/bot.html)

- From: googlebot(at)googlebot.com

- Referer: https://www.google.com/

What Makes This Unusual:

- The service returns scraped HTML within 1–2 seconds

- It works for completely fresh URLs that have never been crawled

- All reverse DNS lookups confirm legitimate Google infrastructure

- The requests are triggered on-demand via API call

Verification Offer

I'm happy to validate these claims by having the service trigger a crawl to a unique test URL, so you can verify in your internal logs that it's genuinely Googlebot being dispatched.

Any insights into how this is technically possible?

Thanks!

cmckn 3 hours ago

The search console for domains allows you to put in a URL and test-scrape it to see how things look to the bot. Could be some reverse-engineering/abuse of that API.

  • devx_ 2 hours ago

    Correct me if I'm wrong, but I believe you referring to the Rich Results Test. Fetching through that embeds `Google-InspectionTool` in the user agent, which isn't the case here.

DaveZale 3 hours ago

there are blockers for webcrawlers. A few dozen were supplied by my neocities.org account, but I had to uncomment them

  • devx_ an hour ago

    Not sure how this is relevant.