Woody Wu · Findability for Hospitality · REF WW-AR-01
Findability
Your next guest asks a machine where to eat. I check what it tells them before they arrive.
Remote first · Canada and the United States
THE PROBLEM THIS SOLVES
Guests used to find a restaurant through a friend, a guidebook or a phone call. Now most of them look it up: on a map, in a list of the best places for something specific, and increasingly by asking ChatGPT, Gemini, Perplexity or Meta's Muse where to go on Friday.
Those answers are built from whatever text a machine can read about you. A wine list that exists only as a photograph, hours that changed last spring, a description written for a brochure: every one of them is a reason to be left out of the answer, or to be described wrongly in it. Nobody sells a fix for this, because there is no ad slot to mark up.
WHAT I CHECK
1. Found.
When a guest asks for the kind of place you are, in your neighbourhood, are you named?
2. Described correctly.
Hours, open or closed, cuisine, chef, the things a guest decides on. Wrong answers are worse than none.
3. Bookable.
Can an agent reach a real booking path for you? I stop before confirming. A test booking is a no-show.
4. Readable.
Shown your wine list, does the machine read it, or invent wines that are not on it?
5. Price-true.
Does it quote your prices exactly? This is checked word for word against your list, not by eye.
6. The guest who is not drinking.
Asked what a non-drinker would have, does it name your real zero-proof list, or admit it does not know?
WHAT YOU GET
A one-page scorecard.
Six checks across four assistants, pass, partial or fail, dated. The assistants change weekly, so the date matters.
The three failures that cost you money.
Ranked, each with the actual answer a guest would have been given.
The fix list.
Most of it is not AI work and your own team can do it: a text version of your list, correct hours everywhere, a description that says what you actually are, the zero-proof list published, prices on the same line as the wine.
The lists worth being in.
The articles and roundups the assistants lean on when they answer questions about places like yours, and what it takes to earn a place in them. This is the part of PR that fills a dining room now.
What your reviews say about discovery.
Your public reviews, read for how guests say they found you and what they mention, with the counts.
The error rate of every instrument.
Every count comes with how often the tool that produced it was wrong when checked by hand, and in which direction.
THEN THE FIX
Most venues do not have the hours to work through the fix list, so I do it for you. The Agent-Ready Fix: your menu and wine list published as real web pages with every price on the same line as its wine, structured data on your site, a plain summary for AI systems, the same name, hours, description and menu link on every profile and platform guests check, a booking path an agent can actually reach, and one description written for the question guests really ask.
Thirty days later I run the check again, and you get the before and after on one page. I cannot promise any assistant will name you, and nobody honest can. What I promise is that they can read you correctly, and I measure it.
WHAT I DO NOT DO
I do not sell advertising, paid placement, review schemes or monthly SEO packages, and I never buy or incentivise reviews. I do not pitch journalists on retainer. I have nothing to mark up, so I have no reason to recommend spending.
I never make a real booking in your name without your written go-ahead, and I never automate the consumer assistants. Every answer in the report is the one a guest would actually see.
If you are a restaurant in my own city, I will tell you so and point you toward someone good.
THE FEE
One venue: $750 CAD, fixed.
The Agent-Ready Fix: $5,000 CAD per venue, done for you, with the before and after included. The $750 check is credited toward it if you commission the Fix within sixty days.
The monthly re-check: $250 CAD a month. The findable, bookable and price checks rerun on all four assistants, with a one-page change report.
WHY ME
I run dining rooms, and I build the AI myself. I built The House Somm, a live AI sommelier that holds a real wine list without misquoting a price, and I test it with the same checks I will run on you. I also built the layer that lets machines read this site: structured data, a plain-text summary for AI crawlers, prices a machine can parse. I know what the machine needs because I have had to give it one.
THE RECORD SO FAR
The record starts with the first clients. Anonymised results will be published here as they come in, including what the assistants got right. No client is ever named.
TO COMMISSION A CHECK
Send the venue name and a link to your wine list. That is enough to start.
DIRECT ANSWERS
How do I get my restaurant recommended by ChatGPT and other AI assistants?
AI assistants recommend what they can read and trust: a wine list and menu published as text rather than images, consistent hours and descriptions everywhere, reviews that mention what you are known for, and a place in the articles and roundups they cite. Findability checks what four assistants currently say about you, and gives you the fix list. From $750 CAD, fixed.
Can AI agents like Meta's Muse book a table at my restaurant?
Some can reach booking platforms, and more will. Whether an agent can book you depends on how you take reservations and what it can find about you. Findability tests whether an agent reaches a real booking path for your venue, stopping before it confirms, so no test booking becomes a no-show.