AI Traffic Tracking is the analytics tool inside your SiteStakes dashboard that tells you two things: which AI crawlers are reading your site, and which AI assistants are sending visitors to it. ChatGPT, Claude, Perplexity, and Bing Copilot all crawl websites to learn from them — and then send people to sites they cite as answers.

Every SiteStakes website ships fully wired for AI traffic. Your robots.txt already allows the major AI bots, the indexing tracker is already logging visits, and AI referrer attribution runs without any setup. The dashboard is where you read the results.

AI is replacing Google for a growing share of searches. When a buyer asks ChatGPT "who buys houses fast in Atlanta," the answer is built from sites the AI has crawled. If you are not in those answers, you are invisible to that buyer. AI referral traffic is becoming a real channel — small for most sites today, growing fast, and most agency platforms have no way to measure it.

Tracking AI traffic answers questions you cannot answer any other way:

• Is AI even reading my site? If GPTBot and ClaudeBot have never visited, your content is not getting absorbed.
• Which pages does AI find useful? Your most-crawled pages are the ones AI has decided are worth learning from.
• Is AI sending me real people? Crawls are one thing — clicks from "according to your-site.com" citations are another.
• Do those visitors become leads? AI Traffic Tracking ties the AI referral all the way to leads and closed deals in your CRM.

Open your SiteStakes dashboard and go to Analytics → AI Traffic in the left sidebar. The page splits into two halves: the top is the indexing report (which bots have crawled your site), and the bottom is the attribution report (clicks, leads, and deals from AI referrals). It gives you AI bot visibility on every plan, with no setup needed.

This page is read-only — it shows you what is happening. To change which bots are allowed to crawl your site, you use a separate page covered further down.

AI crawler tracking runs every 10 minutes and scans for the 27 known AI bots — GPTBot (OpenAI), ClaudeBot (Anthropic), PerplexityBot (Perplexity), Google-Extended (Google's AI training crawler), Applebot-Extended, Bytespider (ByteDance), CCBot (Common Crawl), and 20 others.

Every match logs to the ai_bot_visits table with the bot name, the page visited, the timestamp, and the IP.

The dashboard rolls those visits up into the 27-bot inventory. Each bot gets a checkmark once it has visited your site at least once. This is your at-a-glance answer to "is AI reading my site?"

The Most Crawled Pages panel ranks your URLs by how many AI bot visits they have received. Your top pages are your most AI-absorbed content. Use this to spot which topics are working and which deserve more depth.

SiteStakes generates your robots.txt dynamically from a managed allow-list. Every major AI bot is allowed by default — search bots that bring referral traffic (GPTBot, ChatGPT-User, PerplexityBot, OAI-SearchBot, Bing's AI crawlers) and training bots that absorb content into the underlying models (Google-Extended, Applebot-Extended, ClaudeBot, Bytespider, CCBot). Training bots also get a 5-second crawl-delay so they cannot overload your server.

The same wiring applies to admin paths: your robots.txt blocks crawl access to back-office routes, API endpoints, and tracking pixels. AI bots only see your public content, which is what you want them to see.

Go to Back Office → SEO → AI Traffic → Indexing tab. Every toggle on this page autosaves the moment you change it, and your live robots.txt updates within 5 minutes.

The page header shows a preset badge — "Recommended" when your settings match the SiteStakes default, "Custom" when you have changed something. A "Reset to default" button rolls back any custom changes in one click. Below the badge, a status stripe confirms your SSL, sitemap, and recent AI crawl activity.

Three layers of control sit beneath that. The master switch at the top toggles indexing for the whole site — flipping it off makes your robots.txt block every crawler. Two sub-master toggles let you switch entire categories at once: AI search and retrieval bots (the ones that send referral traffic) and AI training crawlers (the ones that absorb your content into models).

For fine control, the per-bot grid below the masters has a card for every one of the 27 tracked bots, sorted by activity. Each card shows that bot's hits, pages crawled, and first/last seen dates. Toggle any individual card to override its category default — useful for turning off a single noisy crawler while keeping the rest of its category active.

A hidden paths editor lets you exclude specific URLs from indexing (like /staging or /draft). A live preview on the right shows your robots.txt exactly as bots will see it, refreshing on every save. An audit strip at the bottom records your last 10 changes so you can verify or roll back.

When an AI assistant cites your site as a source and a user clicks through, the click arrives with a referrer header from the AI's domain — chat.openai.com, claude.ai, perplexity.ai, copilot.microsoft.com, and 10 others. The SourceClassifier reads the referrer on every page load and tags the visit as AI traffic.

If that visitor submits a form, the AI source attaches to the new lead — the same way utm_source attaches for paid ad traffic. You see "Source: ChatGPT" or "Source: Perplexity" directly on the lead card in your CRM. If the lead becomes a closed deal, the AI referral attribution stays with it, so you can measure revenue from AI traffic the same way you measure revenue from Google Ads.

The attribution panel shows totals — visits, leads, deals — grouped by AI assistant. You see which assistants are sending you business and which ones are crawling but not citing.

The data tells you where to invest. Three concrete moves:

• No crawls yet? Your robots.txt is already allowing AI bots, so the issue is content. Publish pages that answer specific questions buyers ask — AI bots prioritize content with clear answer structure, headings, lists, and direct responses.
• Crawls but no referrals? Your content is being absorbed but not cited. This usually means your pages need more specific information — local data, named neighborhoods, exact processes — that AI assistants will quote as authoritative.
• Referrals but no leads? AI is sending you traffic but your pages are not converting it. Treat AI visitors the same way you treat paid ad traffic: clear CTAs, prominent forms, and pages that match the question the visitor asked.

To find what AI users ask about real estate in your market, see the Keyword Discovery guide. To produce content AI bots will pick up, see the AI Blog Factory guide.