AI referral tracking, built in

See the guests AI assistants send you

Guests ask AI where to stay. See which answers become visits.

How it works

No extra setup. If the reporte.ai snippet is on your site, the report is already filling up.

1. A guest asks an assistant

Someone asks ChatGPT, Perplexity, Gemini or Copilot for a hotel like yours. The assistant answers and links to websites it trusts, including yours.

2. They click through to your site

The visit arrives with a referrer such as chatgpt.com or perplexity.ai. Your tracking snippet records it like any other session, cookieless and privacy friendly.

3. reporte.ai classifies it

The panel matches the referrer against 16 known assistants, groups subdomains under one brand, and shows AI sessions next to search engine sessions.

The shift, in numbers

Independent industry data. AI Analytics measures this same trend on your own hotel website.

+3,500%

year over year growth of traffic from generative AI sources to U.S. travel sites in July 2025, according to Adobe Analytics.

-44%

bounce rate of AI-referred visitors versus other channels. They also browse longer and view more pages per visit (Adobe Analytics).

88%

of travelers who planned a trip with AI say it improved their booking and travel experience (Adobe survey, 2025).

If an AI assistant cannot read your hotel website, it cannot recommend your hotel. Measuring AI referrals is the first step to winning them.
Source of the figures: Adobe Analytics, traffic to U.S. travel sites, July 2025.

What you see in the panel

Three views, one question answered: is the AI ecosystem sending you guests?

Screenshot: AI sources ranking

Every assistant, named and ranked

A table lists each assistant by name with its sessions and events over 7, 28 or 90 days. ChatGPT at the top with 41 sessions, Perplexity with 18, Gemini with 9, and so on down the list. A headline KPI compares total AI sessions against search engine sessions so you see the shift happening.

Screenshot: daily trend

Your AI visibility over time

A day by day chart of AI referred sessions. Flat at zero means assistants do not cite you yet. A rising line means your content changes are working. This is the curve to watch after you apply the recommendations below.

Screenshot: landing pages

The pages assistants trust

The exact URLs where AI visitors land: your rooms page, your FAQ, a blog post about the area. These are the pages assistants quote, so they are the ones worth polishing first. Multi property accounts also get a per website breakdown.

Key recommendations to win AI traffic

A working checklist for hotels. Apply it, then watch the trend chart move.

Write answer-first content

Open every page with the direct answer: where the hotel is, who it is for and the price range, all in the first paragraph. Assistants quote pages that answer cleanly, not pages that warm up for three paragraphs.

Keep prices visible in plain HTML

Assistants cannot read booking engine widgets or images. Publish "from" rates as real text on rooms and offers pages so an assistant can cite a concrete number next to your name.

Add structured data

Mark up your site with Hotel, Offer and FAQPage schema. Structured data is the most machine readable version of your facts and it feeds both Google and AI answer engines.

Publish real FAQs

Answer the questions guests actually ask assistants: cancellation policy, parking, check-in time, pets, breakfast. One clear question and one clear answer per block.

Allow AI bots in robots.txt

Let GPTBot, ClaudeBot, PerplexityBot and Google-Extended crawl your site. If they cannot read you, they cannot recommend you. Blocking them saves nothing and costs citations.

Serve an llms.txt file

Publish a short llms.txt at your root describing the property and linking your key pages. It is a cheap, standard way to hand assistants a curated map of your site.

Keep facts consistent everywhere

Name, address, amenities and policies must match across your site, Google Business Profile and the OTAs. Assistants cross check sources and drop hotels with conflicting data.

Measure, then iterate

Use this report to see which pages AI visitors land on. Strengthen those pages, add the missing answers, and check the daily trend after each change.

Frequently asked questions

Any session whose referrer is a known AI assistant domain, such as chatgpt.com, perplexity.ai, gemini.google.com or copilot.microsoft.com. We match 16 assistants, group their subdomains under one brand, and keep classic search engines like Google and Bing in a separate comparison metric.

No. The standard reporte.ai tracking snippet already records the referrer of every session. The AI Analytics report classifies those referrers on the server, so existing installations get the report with zero extra setup.

No tool can measure it perfectly. Many AI answers are read without any click, and some assistant apps strip the referrer. What you see in the panel is the measurable floor of your AI visibility, which is exactly the number you can act on and grow over time.

Generic analytics tools bury assistants inside long referral lists. reporte.ai names each assistant, shows the daily trend, compares AI sessions against search engine sessions, and lists the exact landing pages guests arrive on. It is also cookieless and privacy friendly by default.

Publish answer-first pages with visible prices, add Hotel and FAQPage structured data, answer real guest questions on your site, allow AI crawlers such as GPTBot, ClaudeBot and PerplexityBot in robots.txt, and serve an llms.txt file. Then use this report to see which pages assistants send guests to and strengthen them.
 

Your first AI guest may already be in the data

Sign in and open the AI Analytics report. It works on every site you already track.