All terms

Tooling

Bot detection

Also known as: bot management, automated traffic detection

Bot detection is the practice of identifying automated traffic to a website or API and separating it from human visitors. It distinguishes wanted automation, such as search engine and AI crawlers, from unwanted activity like scraping, credential stuffing and click fraud. Detection results feed decisions about what to allow, challenge, rate limit or block.

What it is

Bot detection combines identity signals with behavioural analysis to classify each request. Identity signals include the declared user agent, IP address range and reverse DNS verification, while behavioural signals cover request rate, navigation patterns, header consistency and browser fingerprint. Most implementations sit in a CDN, WAF or dedicated bot management product.

Why it matters

AI assistants, agents and training crawlers now generate a meaningful share of requests, so teams need to know which automation is visiting before they can decide access policy or read their analytics correctly. Misclassification cuts both ways: blocking a legitimate AI crawler can remove you from answer engines, while counting bots as humans inflates sessions and distorts conversion rates. Accurate classification is the foundation for both visibility and reporting.

How it works

Practitioners verify declared crawlers against published IP ranges or reverse DNS rather than trusting the user agent, then apply graded responses such as allow, rate limit, challenge or block by category. Server logs and CDN reports are segmented by agent so crawl volume from search and AI systems can be tracked separately from human sessions. Rules are reviewed regularly as new agents appear and as spoofing patterns change.

When it applies

It applies whenever you are setting crawler access policy, investigating unexplained traffic spikes, protecting login or checkout endpoints, or reconciling analytics with server logs.

Examples

  • A publisher verifies requests claiming to be an AI crawler against the vendor's published IP list and blocks the spoofed ones.
  • An ecommerce site rate limits scrapers hitting product pages every few seconds while leaving verified search crawlers untouched.
  • An analytics team excludes identified bot traffic from a report after a session spike with a near-zero conversion rate.

How it is measured

  • Share of requests classified as human, verified bot and unverified or suspected bot
  • Verified versus spoofed request volume per declared crawler
  • False positive rate, measured through challenge pass rates and user complaints
  • Crawl requests and bytes served by agent, tracked month on month

Related terms in Tooling

Primary research · August 2026

How ChatGPT Shortlists Software Brands

An audit across 10 categories and 60 buying questions. I recorded what ChatGPT reads, throws away and links to when a buyer asks it which software to buy, and what that decides.

60
Questions asked
10
Software markets
2,680
Results read
367
Links shown
Free35 pages · PDF · 536 KBDiscovery Digest every Friday

Free download

Get the full report

35 pages · PDF · 536 KB. Enter your details and it downloads straight away.

How ChatGPT Shortlists Software Brands downloads straight away. No spam, unsubscribe anytime.