A menu and hours audit for Wellington restaurants
Start with the information a diner needs tonight: the menu, the location, today's hours and a working reservation or contact path. Make those facts agree before trying to influence an AI answer.
By Limor · Published
Build a single public reference
Choose the website page your team treats as the current menu and hours reference. A readable text menu helps people on phones and assistive technology; a designed image or PDF can be an additional format. Keep dishes, prices and availability accurate rather than letting an old upload become the default.
For a Wellington restaurant, useful location context may include genuine nearby landmarks and arrival details. Do not imply that the restaurant is in another community to reach more searches. Describe outdoor seating, accessibility and dietary options only where those statements are true and appropriately maintained.
Compare the versions diners can find
Use a simple audit with one row per public listing: website, Google, Apple Maps, Yelp, Tripadvisor and any other channel your guests actually use. Record the displayed hours, phone, address and menu URL, then compare them with the approved reference.
Prioritize contradictions that could spoil a visit, such as an old closing time or a menu link that no longer works. Assign an owner and a correction date. If a third-party platform is beyond your control, record the limitation and use its supported correction process rather than claiming the information is fixed.
Add a change trigger
A monthly review is useful, but changes in service hours, menus or holiday plans should trigger an earlier check. Name who updates the website and who checks the other profiles. A kitchen or operations change is not complete from the guest's perspective until the public information catches up.
Test the mobile journey: find a dish, confirm the service time, locate the entrance and use the booking or call link. Ask staff which questions still arrive repeatedly. Those are candidates for clearer information, not a reason to generate dozens of thin pages.
Keep feedback and claims honest
Invite every guest neutrally to share honest feedback, without incentives or suggested wording. Reviews should reflect the diner's own experience. Avoid promises that a particular review count or menu format will cause ChatGPT or another assistant to recommend the restaurant.
When checking AI answers, save the question, date, location and conditions, and repeat the test. Use factual mistakes to guide maintenance. The most dependable outcome of this audit is accurate information diners can act on, regardless of where they discovered you.
Source and further reading
This Pretty Limor edition adapts Limor’s original BeautyAI guide on AIrecommend, with a focused practical workflow for this site. The original source and this edition may be updated separately.