Every visitor you see is a person
Search engines, AI crawlers, uptime monitors, scrapers and data-center traffic are identified with 197+ rules in 5 layers and moved to their own reports. Bounce rate, conversion and every chart describe people.
- 197+ rules
- 5 detection layers
- Bots reported, not deleted
| Metric | Mixed | Humans only |
|---|---|---|
| Bounce rate | 61% | 44% |
| Conversion | 2.6% | 3.8% |
| Avg. visit | 0m 52s | 1m 37s |
Bots water down every number you report
Rates look worse than they are
Crawlers never convert. Every bot visit lowers conversion and raises bounce rate, so a good page looks broken.
Spikes that aren’t real
A scraper or a monitoring service can double a day’s traffic. Teams celebrate, then chase a growth that never happened.
Mixed in, so invisible
When bots are silently filtered or silently included, you can’t see who is crawling you or how much.
5 layers, from user agent to behaviour
Each request is checked layer by layer. A match moves it to the bot reports; everything else counts as human.
- 1
User agent
Known crawlers, tools and headless browsers matched against a maintained rule list.
- 2
Request headers
Missing or contradictory headers that real browsers always send.
- 3
Client signals
Automation flags and signals a real browser exposes when the page runs.
- 4
Network (ASN)
Data-center and cloud networks labelled, so server traffic is not mistaken for visitors.
- 5
Frequency
Request patterns no person produces — too fast, too regular, too many.
Bots get reported, not hidden
Filtering bots out is only half the job. PathClue shows you who they are.
Human and bot traffic, side by side
Total traffic split into humans, search engines, AI crawlers, monitors and data-center traffic — with how much each one changes bounce rate, conversion and visit length.
- Bot share by category
- Metrics with and without bots
- Top bots this month
Crawlers checked against your rules
Verified and suspected-fake crawlers, the pages they fetch, and whether they respect robots.txt — with evidence and a snippet to block them on your server.
- Verified vs suspected-fake crawlers
- Disallow and Crawl-delay violations
- Robots.txt and Nginx snippets
| Crawler | Requests | robots.txt | Status |
|---|---|---|---|
| Googlebot | 6,410 | Follows rules | OK |
| GPTBot | 3,120 | Follows rules | OK |
| Bytespider | 2,650 | Ignores Disallow | Flagged |
| CCBot | 890 | Follows rules | OK |
User-agent: Bytespider Disallow: /
Network type on every session
Each session is labelled as residential, mobile or data-center network, so you can spot traffic coming from servers rather than people.
- Residential, mobile and data-center labels
- Cloud provider and network names
- Filter any report by network type
- 1
Visit 1 · Google · “web analytics”New//blog/ai-search2m 41s
- 2
Visit 2 · NewsletterReturning/pricing/case-studies4m 02s
- 3
Visit 3 · DirectSigned up/pricing/register1m 18s
Bot detection questions
Anything else — open the chat in the corner and ask.