How to Track AI Crawlers in WordPress (GPTBot, ClaudeBot and More)

AI crawlers are visiting your WordPress site right now. You just cannot see them.The problem is that the place most people look, Google Analytics, is completely blind to bots. So the very traffic that decides whether ChatGPT or Perplexity will ever quote you does not show up in your analytics at all.This guide shows you how to actually see those crawlers, how to tell a real GPTBot from an impostor, and how to read what the visits mean for your AI visibility.

Why Google Analytics never shows AI crawlers

GA4 measures humans. It fires when a browser loads your page and runs its JavaScript tracker. AI crawlers request your raw HTML and leave; there is no browser and no script execution, so GA4 records nothing. You could be crawled thousands of times a month by OpenAI, Anthropic and Perplexity and see a flat zero in your analytics. The only place these bots reliably appear is in your server's own request logs, which is where any real AI crawler tracking has to look.

The AI crawlers worth knowing

You do not need to memorize every bot, but a handful matter most for AI citations.

The fuller list and what each one does is in the Crawl Academy lesson on GPTBot, ClaudeBot and PerplexityBot.

Step one, see who is crawling you

Start by reading your server access log, where every bot leaves a row. You can do this manually by filtering the log for known user-agents, but that is slow and error-prone. The free AI Crawler Log Analyzer does it for you: paste a log file and it shows which AI bots visited, how often, and which pages they read, all in your browser with nothing uploaded. It is the fastest way to get a first picture of your AI crawl activity.

Step two, verify the bot is real

Here is the trap that ruins most bot tracking. Anyone can set their user-agent to "GPTBot," and scrapers do it constantly to look legitimate. If you count by name alone, your numbers are inflated by impostors, and a fake GPTBot hitting your login page can look like AI interest when it is just an attack. Real verification uses reverse-DNS confirmation and published IP ranges to prove a request actually comes from OpenAI, Anthropic or Perplexity. Insist on it, or treat raw counts with suspicion. The full verification method takes two DNS lookups and catches every impostor.

Step three, read the intent behind the visit

Not all crawls mean the same thing. A bot fetching your page to train a model is different from a bot fetching it live to answer someone's question right now. The second kind, like ChatGPT-User, is a leading indicator that you are appearing in an AI answer, because a real person triggered it. Tracking that separates a flat hit count from a real visibility signal. A spike in live-answer fetches on a page is one of the earliest hints that you are being cited.

Step four, watch for crawl waste and blocks

Once you can see the crawlers, two patterns are worth watching. The first is crawl waste, AI bots spending their visits on thin archives or filtered URLs instead of your money pages; trimming that focuses their attention where it counts. The second is accidental blocks. If a bot you want suddenly stops appearing, check your robots.txt and host rules, because a silent block is the most common self-inflicted reason citations dry up. The free AI Crawlability Checker confirms which bots your rules currently allow.

How CiteTrack tracks crawlers continuously

A log analyzer gives you a snapshot; continuous tracking gives you the picture over time. CiteTrack AI's crawler analytics runs inside WordPress, logging 50+ AI bots as they arrive, verifying each one with reverse DNS, flagging spoofs, classifying intent, and showing your most-crawled pages and how often each engine returns. The data stays in your own database with hashed IPs, so nothing leaves your site. If you want to connect crawler data to citation tracking and revenue attribution, the full WordPress analytics stack shows how the three layers fit together.

Frequently asked questions

Where to go from here

Crawler tracking is one layer. To see whether those crawls actually turn into citations, add citation tracking on WordPress so you can close the loop. The full WordPress analytics stack connects crawlers, citations, and revenue attribution in one view. And if you want the foundational concepts, the Crawl Academy covers what AI crawlers are, the specific bots you will see, and how to verify they are real.

Guides · AI Crawlers, ClaudeBot, GPTBot, How-To, WordPress