Does ChatGPT use schema markup?

AI assistants draw on the same crawled, parsed web that search engines index, and structured data is the most reliable way to tell a machine what a page is about. Markup on its own does not get a company cited. Markup on a page with real, indexable text about a real product in a real market is what does.

How do I check whether AI assistants mention my business?

Ask them the questions your buyers ask, in your buyers' words, and write down the answers. Do it monthly against the major assistants and keep the record. There is no dashboard that does this reliably for you yet; the panel of real questions is the measurement.

What is an llms.txt file?

A plain-text file at the root of your site, like robots.txt, that summarises who you are, what you offer and where, in a form written for AI systems to read. It takes an afternoon to write and most competitors do not have one. OpsPoint publishes its own at opspoint.ai/llms.txt.

How to get your rental catalog cited by ChatGPT and Google AI

October 26, 2026 · Updated October 27, 2026 · 8 min read · OpsPoint

In short

When a planner asks an AI assistant who rents lounge furniture in a city, it returns a few names drawn from pages it could read, parse and trust. Rental catalogs lose because they are photographs without text, carry no structured data, and have no page for the markets they serve. Indexable text per product line, Product and LocalBusiness markup, a page per market, an llms.txt file, and a site that stays up are what put a catalog on that short list.

When a planner asks ChatGPT, Claude or Gemini “who rents lounge furniture in Atlanta?”, the assistant doesn't return ten blue links — it returns a few names and moves on. Those names are assembled from pages the assistant could read, parse and trust. Most rental catalogs, for all their photography, give it nothing to work with.

Why beautiful catalogs lose

Photos without text. A gallery of product images with a name and a “add to quote” button is nearly invisible to a machine. No description, no dimensions, no context — nothing quotable.
No structured data.Without Product, Organization and LocalBusiness markup, the machine is guessing what the page is about, and machines don't cite guesses.
No market pages. If you serve eleven cities but your site only says so in a footer, there is no page for “event furniture rental in Houston” to land on — for Google or for an assistant.

A worked example: the same catalog, two ways

Two rental companies in Atlanta both carry a white leather lounge collection. The first has a gallery page: twelve photographs, each titled “Lounge Set 4”, with an “Add to quote” button. The second has a page for the collection with four paragraphs: what the set is, its dimensions and seating capacity, what it pairs with, the kinds of events it has been used for, and the delivery area. Underneath that sit a Product markup block and a LocalBusiness block with the company's name, address and phone matching the ones on Google.

A planner asks an assistant for white lounge furniture rental in Atlanta for a 150-guest gala. The assistant names the second company, cites the collection page, and mentions the seating capacity. The first company does not appear. Its photographs were better. The machine could not read them.

What has to be on the page

Indexable text per product line: what it is, dimensions, capacity, typical uses, what it pairs with — real sentences a machine can lift and cite.
Product markup (JSON-LD) on catalog templates, and LocalBusiness markup per market with consistent name, address and phone everywhere they appear.
A page per market you serve, generated from the catalog data you already have — inventory highlights, delivery area, venues you know.
An llms.txt file: a plain-text summary of who you are, what you rent and where, written for AI systems. An afternoon's work; most competitors don't have one.
A site machines can reach: fast, crawlable, and up — repeated outages teach crawlers to stop coming.

How to measure it

Write down the twenty questions your buyers actually ask, in their words, per market: “who rents farm tables in Nashville”, “lounge furniture rental for a wedding in Charleston”. Run them against ChatGPT, Gemini, Claude and Google's AI results once a month and record who is named and which page is cited. The first run is your baseline. The only honest scorecard is movement against it. And check your own pages the way a machine does: view the source. If the facts are not in the HTML, they do not exist.

This is exactly the work of our standalone AI Search Optimization offer, and it pays back fastest for rental operators with deep catalogs.