The Cult of “First-Party Data”
In SEO, there is one unwritten rule everyone repeats:
“Don’t trust third-party tools like Semrush or Ahrefs for traffic estimations—only trust Google Search Console. It’s first-party data directly from the source.” (me included 😄)
We used to treat GSC like an infallible source of truth. If Google says a query generated 10 clicks, we assume 10 human searchers clicked our link.
Except they didn’t.
This case study demonstrates why blind faith in GSC can completely mislead you, how automated bot behavior pollutes deep SERP data, and why “first-party” does not mean “clean.”
A holy grail traffic page
To understand the anomaly, let me share info about a page:
- The Page: A neglected on-page SEO service page.
- The Content: Thin, low-effort, with mixed search intent (half commercial pitch for a €460 audit, half elementary explanation of title tags).
- The Ranking: Solidly parked in search purgatory—averaging positions 40 to 60 (Pages 4 through 6 of Google) across almost all tracked queries.
In a normal search ecosystem, this page should be invisible. Nobody browses page 5 of Google to click a thin service page.
The Anomaly
Suddenly, GSC reported a surprising surge in clicks over a 7-day period:
- Position 47.5: 8 clicks (4% CTR)
- Position 45.0: 5 clicks (11.6% CTR on an informational query)
- Position 51.3: 4 clicks
In organic search benchmarks, a page ranking in the top 3 spots might get an 11% CTR(once upon a time 😄). A result on Page 5 getting an 11.6% CTR is mathematically and behaviorally absurd.
If an SEO looked only at this top-level GSC view, they might conclude:
- “Our meta titles are insanely magnetic!” or
- “Users really want what this page is offering—let’s optimize it immediately!”
Both assumptions would be completely wrong.
Drilling into the Data
Step 1: The Geographic Breakdown
When you isolate the data by country, the numbers become even more surreal:
- Georgia: 10 clicks on 12 impressions at Position 48.5 (83.3% CTR)
- Belarus: 7 clicks on 30 impressions at Position 41.0 (23.3% CTR)
- Kazakhstan: 6 clicks on 29 impressions at Position 35.8 (20.7% CTR)
- US & Tier-1 Markets: Hundreds of impressions, 0 clicks (the expected baseline for position ~48).
Searchers in Tbilisi are not scouring page 5 of Google to click a niche European SEO agency’s service page 83% of the time.
Step 2: The GA4 Reality Check (Ghost Traffic)
Cross-referencing this URL with Google Analytics 4 exposed the phantom nature of the traffic:
- GSC reported: Over 30 clicks across these regions.
- GA4 recorded: Exactly 5 sessions over a 28-day window.
- Average Engagement Time: 0s.
The clicks were registered on Google’s SERP redirect (google.com/url?q=…), but the visitors never actually engaged with the site. They were headless bots that dropped the request before JavaScript rendered, hit an anti-bot challenge, or simply scraped the raw HTML and left in milliseconds.
The Root Cause: SERP Poisoning & Deep-SERP Scraping
Why would automated scripts browse to Page 5 and click obscure results?
A. Google’s SERP Degradation Forces Scrapers Deeper
Google’s first page has become an obstacle course: AI Overviews, Reddit forums, sponsored shopping carousels, video packs, and People Also Ask modules.
For scraper networks, automated competitor trackers, and AI training crawlers seeking clean, indexable target URLs, Page 1 is increasingly polluted and heavy to parse. Crawlers and scraping tools are frequently configured to paginate 4 to 6 pages deep, programmatically interacting with and following the SERP links.
B. Proxy Hubs & CTR Spoofing Networks
The geographic concentration (Georgia, Belarus, Kazakhstan) aligns directly with known regional proxy routing pools. Whether it’s:
- Automated tools verifying rankings through residential proxy networks, or
- Black-hat CTR manipulation scripts attempting to simulate “human” search behavior by clicking down the SERP,
…the result is that bot noise is actively contaminating Google Search Console performance reports.
The Takeaways for Modern SEOs
- GSC Clicks ≠ Site Visitors: A click in GSC simply means someone (or some script) triggered Google’s outbound redirect link. It does not mean a browser loaded your tracking scripts or that a human was on the other end.
- Beware the “Average Position” Trap: A position of 45 doesn’t mean your page is ranking evenly at 45. It could mean #2 in a small proxy region generating bot clicks, and #85 everywhere else. Always segment by country.
- Triangulate Your Data: Never report GSC performance to a client or stakeholder without verifying GA4 engaged sessions and server log / CDN (Cloudflare) traffic.
- First-Party Doesn’t Mean Clean: Google’s anti-bot detection filters out billions of automated queries, but it is far from airtight. Deep SERPs remain a wild west of crawler activity that GSC dutifully logs as “organic traffic.”
Leave a Reply