Resources · Intake

What happens when a guest's AI reads your event inquiry form?

Dexter's co-founder let Meta's new agent, Muse, explore a venue's inquiry form. What it understood, what to spell out, and why an AI inquiry deserves a reply.

October 30, 2026 8 min read
Table of contents

Google’s guide for website owners now describes AI agents that can book a reservation on someone’s behalf. When a guest’s AI assistant lands on your venue’s event inquiry form, it tries to figure out your spaces, your prices, and the kinds of events you host. Meta launched their own agent, Muse, on September 8, 2026, and they say Muse can fill out forms for the people who use it.

So I tried it. I gave Muse one of our customers’ venues and let it explore. It read the venue’s Dexter form, understood the differences between the spaces, caught nuances like serving only beer and wine, and picked up their starting prices. Then I could have a conversation with it about what would work best for my event. It worked better than I expected.

What got me excited was the context it picked up. It knew enough to help me figure out whether a space actually made sense before reaching out. And it could submit the lead for me.

What to put on your form

A guest’s AI assistant or agent needs the basics before it sends anyone your way: which spaces you have, how many each holds, and what kinds of events you host. Where prices start helps too, if you choose to share it. At Dexter, we spend a lot of time helping restaurants and venues explain their offerings clearly. And it turns out that organizing information for a guest makes it easier for someone’s AI to understand, too.

So name each space and say what sets it apart, whether it’s a private dining room, a semi-private area, or a full buyout. Say how many guests each one holds, so nobody has to guess whether a party of 40 fits. List the events you host: corporate, wedding, social, or whatever mix you take, and whether you offer seated, standing, or both. If you’re comfortable sharing it, give a sense of budget, such as a starting price for each space. And spell out the details that can make or break an event, like a heated outdoor space or a room that only works for a seated dinner.

It also helps to read your last 50 inquiries and your first replies to the guest. The questions you keep asking leads by email belong on the form, so your team doesn’t spend every first reply asking them again. Then ask only the questions that move the sale: the date, the budget, the headcount, and any deal breakers. Whether a space fits depends more on budget than on headcount, unless there’s a hard cap, like the fire code or the number of seats for a seated dinner. With all of that on the form, guests know what they’re asking about, and your team has what they need to reply.

Getting all of that onto your form is the job of Dexter Intake. Dexter is an event sales software for restaurants, hotels, and venues to grow event revenue. It has four parts in one product: AI Search, Intake, and Outreach, with Insights included. Intake is the inquiry form, and it sits on your website where your contact form lives, or on a branded page you link to. Each inquiry arrives enriched with the guest’s details. It goes to your event management system, like Tripleseat or Perfect Venue, or to an email inbox, like Gmail or Outlook.

If you use an event management system like Tripleseat, it often comes with its own form. If that form is doing the job and helping you grow your event revenue, don’t change a thing. But if you find yourself asking the same questions over and over, getting bombarded with unqualified leads, and spending hours going back and forth, you likely have a lead qualification problem up front.

An AI shops around faster than a guest does

When a guest goes quiet, they’re usually shopping around. They may have written to five venues and stopped answering four of them. An AI assistant shops around the same way, only faster. So for many guests, what their AI reads on your form is their first impression of your venue. The venues whose forms spell out their spaces and details are the ones the guest ends up talking to.

Just because it’s AI doesn’t mean it’s spam

An inquiry an AI helped write can still come from a real person with a real event to book. They’ve just asked an AI assistant to help with the homework. We want Dexter to make that homework easier for the guest’s AI, and to make sure the inquiry that reaches your team has the details in it.

Make your events page easy for AI to read

  • Put your spaces and capacities on your events page as real text, not in a PDF, image, or slideshow, because AI can sometimes have trouble reading PDFs. Add your minimums too, if you choose to share them.
  • Keep it current. If a room was renovated or a capacity changed, update the page, because AI will confidently repeat stale information.
  • Give each space its own section with the basics: seated and standing capacity, a price range or food and beverage minimum if you choose to share one, and what’s included, like AV or a private bar.
  • Add an FAQ at the bottom that answers what guests actually ask: deposits, outside vendors, lead time, parking, and dietary options.

What’s next for Dexter

We’re exploring a few things:

  • Making our intake forms even easier for search tools and AI agents to find and read, building on Dexter AI Search, which runs with Intake.
  • Giving agents a more direct way to see what a venue offers and ask the right questions, so the inquiry arrives qualified. That matters because when guests search for a venue’s event details through AI today, the information they get back is often wrong.
  • Helping your team recognize when an inquiry, or a reply, comes from an AI working on someone’s behalf.

Try Muse on your own form

Muse is rolling out in the US on iOS, Android, and muse.ai, and Meta says it’s free for most of what people need. To test your own form, ask it about your venue the way a guest would. Tell Muse that you’re planning a work dinner for 20, and ask where prices start.

FAQ

Can an AI assistant read my event inquiry form?

Some assistants can. Meta describes what Muse does once a person shares a goal with it: “It can open a browser, fill out forms, and negotiate on their behalf.” Google’s guide says browser agents may visit a website to gather what they need, and that they may look at screenshots and at how the page is built. In my test, Muse read a customer’s Dexter form and found each space’s starting price. ChatGPT and Claude can also visit websites for the people who use them, and Anthropic says Claude in Chrome can click buttons and fill forms.

How should my team answer an inquiry an AI helped write?

A venue’s team should answer it like any other inquiry, based on what it says. If it has a date, a budget, and a headcount, your team has what they need to reply with something specific. An inquiry with none of those details is vague, no matter who typed it. Answer it fast, too, because the guest’s AI may have written to several venues at once.

Should I show prices on my event inquiry form?

Starting prices, yes, if you’re comfortable sharing them. They help a guest, or their AI, find the right fit before they reach out. Guests whose budget doesn’t match can move on early, and the ones who do reach out already know what to expect. Your team spends their time on inquiries that are ready to book. With Dexter Intake, starting prices can adjust to what the guest is planning. A guest who picks a weekday in January for 20 people in your private dining room sees a different starting price than one asking about a Saturday in December for 80 in the main room. That way the number they see fits their event. Minimums are a touchy subject, and some venues like to keep them private. Dexter Intake shows them only if you choose to display them.

Do I need special files or markup so AI can read my venue’s pages?

Venues don’t need special files or markup for Google. Their guide says you don’t need new machine-readable files, AI text files, markup, or Markdown to show up in Google Search, and that includes its AI features. In Google’s words, “You don’t need to write in a specific way just for generative AI search.” For ChatGPT, Claude, and Perplexity, the step that matters is letting their search crawlers reach your pages: OAI-SearchBot, Claude-SearchBot, and PerplexityBot. OpenAI and Perplexity both recommend allowing theirs in your website’s robots.txt file, which tells crawlers what they can visit. If someone else builds your website, send them web.dev’s guide, Build agent-friendly websites, which is the one Google links to.

Sources

Book a Demo

See it with your own form

Give the demo a whirl, or book 15 minutes and we will show you a Dexter form flowing into Tripleseat, another event management system or your inbox.

Book a Demo