Couples asking ChatGPT, Gemini, Perplexity or Google's AI Mode for suppliers get names the AI can find described clearly and consistently: your website saying what you do, where, in what style and from what price; a complete Google Business Profile; reviews describing real weddings; and directory and venue listings. Your site must also be open to AI search crawlers.
What changes from ordinary search is the shape of the answer. Instead of ten links, the couple gets a shortlist with reasons: "known for relaxed documentary photos, packages from about $2,800, works often at barn venues". If your website never states your style, starting price or the venues you work at, the assistant has nothing to repeat, and it names someone whose site does. Nobody can pay for a place in these organic answers, and Google itself says no special optimisation is needed beyond being indexed and useful.
Where the AI assistants get supplier names
When a couple types "who are the best wedding cake makers for a rustic wedding in [town]?", the assistant usually searches the web live, reads a handful of pages, and writes a summary with links. The pages it reads are the same kinds of pages a human would find: supplier websites, directory listings, venue "recommended suppliers" pages, review sites and wedding forums.
- Google's AI Overviews and AI Mode draw on Google's search index. Google says a page must be indexed and eligible to show with a snippet to appear as a supporting link, and that Business Profiles can help your services show up in AI responses as well as ordinary results.
- ChatGPT search uses OpenAI's crawler, OAI-SearchBot, to surface websites. OpenAI states that sites which block OAI-SearchBot won't be shown in ChatGPT search answers.
- Perplexity uses PerplexityBot to surface and link websites in its results, and asks site owners to allow it in robots.txt.
Two consequences follow. First, a wedding business with a thin website and a strong Instagram is nearly invisible, because the assistants mostly read web pages, not social feeds. Second, what other sites say about you matters as much as what you say, because the assistant is looking for agreement between sources. The broader version of this, for any business, is in getting your business recommended by ChatGPT and AI search.
Step 1: find out what AI says about you today
Spend 30 minutes running the questions couples actually ask, in three or four assistants, and write down what comes back. Use a private or logged-out window where possible, since personal history can colour the answers. That caution is not theoretical: an owner who runs the test in the ChatGPT account they use for their own business emails may find their name first in every answer, because the assistant remembers what they've been working on. The flattering result is worthless; a couple's fresh account won't see it.
1. Who are the best [your service] for weddings in [your town]?
2. Recommend a [your service] for a [style] wedding at [a venue you
work at often].
3. [Your service] for a wedding of about [typical guest count] with
a budget of around $[typical budget]?
4. What do couples say about [your business name]?
5. How much does [your business name] charge?Here is an illustrative log for a wedding cake maker:
| Prompt | Assistant | Named? | What it said | Source it cited |
|---|---|---|---|---|
| Best wedding cake makers in [town] | ChatGPT | No | Named three competitors | A directory "top 10" page |
| Cake for a rustic barn wedding at [venue] | Gemini | Yes, second | "Semi-naked cakes, works with local venues" | The venue's recommended suppliers page |
| What do couples say about [business]? | Perplexity | Yes | Summarised six Google reviews | Business Profile, one wedding forum thread |
| How much does [business] charge? | ChatGPT | Yes | "Prices not published" | Own website |
This log is typical: the business is found when a venue or reviews mention it, and missing when the question is general or about price. That tells the owner exactly what to fix: publish a starting price, and get onto the directory pages the assistants cite. If an assistant states something wrong about you, what to do when ChatGPT gets facts about your business wrong covers the fixes.
Step 2: write pages that answer couples' actual questions
Assistants repeat specifics. Vague websites give them nothing. Compare two versions of the same wedding cake maker's services page.
Before:
We create beautiful, bespoke wedding cakes for your special day. Every cake is made with love and attention to detail. Get in touch to discuss your dream cake!
After:
Wedding cakes for 50 to 250 guests, delivered and set up at venues within about an hour of [town]. Styles we're known for: semi-naked and buttercream cakes with fresh flowers, and simple three-tier iced cakes. Prices start at $450 for a two-tier cake serving 60, and most couples spend $650 to $1,100. We deliver regularly to [venue], [venue] and [venue]. Tastings on Saturdays, booked online. Gluten-free and vegan cakes available; we're not a nut-free kitchen.
The second version answers five questions a couple might put to an assistant: can they cater my size, do they do my style, what does it cost, do they work at my venue, can they do dietary options. Each answer is something the assistant can quote. Add a short FAQ page with the questions couples email you most often, answered plainly, and a page per venue you work at regularly if you have genuine photos and notes from weddings there.
The FAQ page works the same way: specific answers the assistant can lift whole. Three entries, filled in for the illustrative cake maker:
How far ahead should we book?
Most couples book 6 to 12 months ahead. For May to September
Saturdays, book at least 8 months ahead; we take four weddings a weekend.
Do you deliver on Sundays and public holidays?
Yes, for a $60 surcharge. Friday weddings have no surcharge.
Can you match the cake to our flowers?
Yes. Send us your florist's recipe or a photo, and we'll either use
the same flowers (supplied by your florist) or match the colours in
sugar flowers, from $80 extra.Each answer contains a number or a condition, which is what makes it quotable. "We're flexible, just ask!" is the FAQ equivalent of the "before" services page.
Two cautions. Only name venues where you've actually worked, and don't create dozens of near-identical "wedding cakes in [town]" pages; that pattern looks like spam to search engines and adds nothing for couples.
Step 3: reviews that describe real weddings
When an assistant says a supplier is "praised for being calm on the day" or "great with dietary requirements", it is summarising reviews. A review saying "Amazing, thank you!" gives it nothing to summarise. You can't write reviews for couples, and you mustn't offer incentives for them. Google's rules have also tightened: since April 2026 you may not ask customers to mention specific things in a review (a staff member's name, for example), and you may not ask only the couples you expect to be happy. So the request itself has to be neutral. A filled-in version, sent a week after the wedding while the day is fresh:
Hi both, I hope married life is treating you well. If you have two
minutes, an honest review on Google would help other couples who are
choosing a cake maker. Here's the link: [link]. Thank you again for
letting us be part of your day. [first name]Couples who write in their own words a week after the wedding tend to mention the things that mattered to them anyway: the venue, the flavour, the delivery. Send the same request to every couple. Google also notes that seeking inauthentic mentions across the web isn't as helpful as it might seem, because its systems focus on high-quality content and filter spam.
Where you can add detail is in your reply, which is public and is read alongside the review. A couple writes "Beautiful cake and it tasted even better. Thank you!" The stock reply ("Thank you so much, it was a pleasure!") adds nothing. A reply with the specifics of the job adds a lot: "Thank you both. It was a pleasure setting up your three-tier lemon and elderflower cake in the orangery at [venue]; the garden roses from your florist were perfect on it. Congratulations again." That sentence tells the next couple, and any assistant summarising your reviews, your style, a venue you work at and that you coordinate with florists. Keep it true to the wedding, and don't repeat the couple's full names in public. Google says owner replies can take up to 30 days to appear, so don't panic if one doesn't show at once.
Step 4: directories, venue lists and your Business Profile
In the illustrative log above, the business was named when a venue's recommended suppliers page mentioned it. That's common. Three places deserve an hour each:
- Venue recommended-supplier lists. Ask the venues you work with most to include you, with a sentence about what you do. These pages are exactly what an assistant reads when a couple names their venue.
- Wedding directories. Make sure your listing on the main directories couples in your market use has the same business name, the same description of your style and the same starting price as your website. Inconsistency makes assistants hedge, and sometimes it quotes the wrong figure. A realistic case: the website says cakes start at $450, but a directory listing written three years ago still says "from $350". An assistant asked about prices answers "from about $350 to $450", and the first sign is a couple's enquiry that opens "we saw you start at $350". Keep a list of every listing with the price and description it shows, and check it each time your prices change.
- Google Business Profile. Complete every field: categories, services with prices, service area, opening hours for tastings or viewings, and recent photos. Optimising your Google Business Profile with AI walks through each field.
Step 5: check the technical basics
Many wedding businesses are invisible to one assistant or another for a technical reason nobody noticed. Check three things:
- Your robots.txt file (visit yourwebsite.com/robots.txt). It shouldn't block the search crawlers you want. OpenAI separates training (GPTBot) from search (OAI-SearchBot), so you can refuse training and still appear in ChatGPT search.
- Hosting and security settings. Some services offer one-click blocking of AI bots. Cloudflare, for example, has a block-AI-bots setting and asks new domains whether to allow AI crawlers. If your web designer switched on a blanket block, you may have opted out of AI search without knowing.
- Indexing. Search for your business name in Google. If your pricing page or FAQ doesn't appear, check it isn't set to "noindex" in your website builder's SEO settings. A common way this happens: the pricing page was built as a hidden draft while prices were being decided, and the "hide from search engines" switch was never turned off after launch. Search Google for
site:yourwebsite.com pricing; if nothing comes back, that switch is the first thing to check.
The robots.txt mistake to look for is a single leftover line. Some sites are built with "User-agent: *" followed by "Disallow: /" to keep a half-finished site out of search, and the rule survives the launch. That blocks every crawler that obeys robots.txt, AI search included, and it explains a business that appears nowhere despite good reviews. A robots.txt that allows search crawlers but refuses AI training might look like this (ask your web designer to confirm it suits your site before changing anything):
User-agent: GPTBot
Disallow: /
User-agent: OAI-SearchBot
Allow: /
User-agent: PerplexityBot
Allow: /
User-agent: *
Allow: /What doesn't work, however it's sold
- Special AI files and markup for Google. Google states you don't need machine-readable AI files, llms.txt or special schema markup to appear in its AI features, and that its Search doesn't use such files. Structured data is still useful for ordinary rich results, but it isn't an AI ranking trick.
- Chopping pages into tiny chunks. Google also says there's no need to break content into small pieces for AI to understand it.
- Fake or bought reviews. They break platform rules and are increasingly filtered out. They also don't contain the specific detail that real reviews do.
- Mass-produced AI pages. Hundreds of thin pages about "wedding photographers in" every nearby town rarely help and can drag the whole site down.
Venues: capacity and logistics come first
A wedding venue's version of the same work puts capacity and logistics first. Couples ask assistants things like "barn wedding venue for 120 with on-site accommodation" or "venue that allows our own caterer". The venue that gets named has a page stating ceremony and dining capacities for each space, whether outside caterers are allowed, curfew and noise-limit times, accommodation numbers, and a price guide by season. A block of plain facts near the top of the page does most of the work (illustrative):
The Barn: ceremonies up to 140; dinner up to 120 seated; evening 180.
The Garden: outdoor ceremonies up to 100, April to September.
Catering: our in-house team, or your own caterer from our approved
list of five (a $750 kitchen fee applies).
Music: amplified music until 11.30pm; sound limiter in the barn.
Accommodation: 8 bedrooms on site, sleeping 18.
Prices: Friday and Sunday from $6,500; Saturday from $9,500 (May-Sept).Each line answers a question a couple would otherwise put to an assistant, or to you in an email. Its recommended-supplier page, in turn, helps the photographers, florists and cake makers it lists. Every wedding business benefits when the others publish specifics.
Tracking whether couples are finding you this way
You won't get a neat report, but three signals together give a fair picture:
- Referrals in your analytics. Look for visits from chatgpt.com, perplexity.ai and gemini.google.com. Links clicked in ChatGPT often arrive tagged utm_source=chatgpt.com. Some AI-driven visits show up as direct traffic, so treat the figures as a floor.
- Ask on your enquiry form. Add "an AI assistant such as ChatGPT" to the "how did you hear about us?" options. It's the most direct signal you'll get.
- Re-run the test monthly. Use the same five prompts and record the same columns. Over a season you'll see whether you're named more often and what the assistants say.
Put the three signals on one line a month so the trend is visible. An illustrative run for the cake maker after publishing prices and the FAQ: "April: named in 1 of 5 prompts, 3 visits from chatgpt.com, 0 enquiries saying AI. June: named in 3 of 5, including the price question (now quoted correctly), 11 visits, 2 enquiries ticking 'an AI assistant'." Small numbers, but moving the right way, and the price answer changing from "not published" to the real figure is the clearest sign that the page is being read.
Tracking whether AI search is sending you customers goes into the analytics set-up in more detail, and how AI Overviews are changing local search covers what's happening to ordinary search results alongside.
Wedding suppliers and AI search: more questions
Can I pay to be recommended by ChatGPT or Gemini?
No mainstream AI assistant sells places in its organic answers, and anyone promising guaranteed AI recommendations should be treated with suspicion. What you can do is make it easy for the assistants to find accurate, specific information about you across your own site, your Business Profile, reviews and directories. Paid ads are separate and labelled.
Should I add an llms.txt file to my website?
Google says its Search features, including AI Overviews and AI Mode, don't use llms.txt or similar AI-specific files, so it won't help there. Other AI companies haven't committed to using it either. It does no harm, but your time is far better spent on a clear pricing page and more detailed reviews.
If I block AI training bots, will I disappear from AI search?
Not necessarily. OpenAI uses separate crawlers for training (GPTBot) and for surfacing sites in ChatGPT search (OAI-SearchBot), so you can block one and allow the other. Perplexity's search crawler is PerplexityBot. Check your robots.txt and any hosting or security settings that block AI bots, and allow the search crawlers you want.
How long before changes show up in AI answers?
It varies. Assistants that search the web live can pick up a new page once it's crawled and indexed, which can take days to weeks. Reviews and directory listings build up over months. Re-run your test prompts monthly and track the trend over a season rather than expecting a change in a week.
Further reads
- How to Choose an SEO Agency That Understands AI Search — If you'd rather hire help, what to ask an agency.
- Does Google Penalise AI-Written Content on a Business Website? — Whether AI-written pages hurt your site.
- What Is Review Gating and Can It Get Your Business Penalised? — The review-request practice that can get you penalised.
- How Wedding Suppliers Can Reply to Enquiries in Minutes With AI — Reply quickly once AI search sends couples your way.
- Can AI Qualify Wedding Leads by Date, Budget and Guest Count? — Sort the extra enquiries by date, budget and guests.
- How to Do Keyword Research With AI for a Small Business Website — Find the questions couples actually search for.
- Is AI Culling Worth Paying for as a Wedding Photographer? — What AI culling costs a wedding photographer, the break-even by weddings per year, where it misjudges wedding frames, and a two-wedding trial to settle it.
- How Wedding Venues Use AI for Viewings, Enquiries and Follow-Ups — Use AI at three points in a venue's sales pipeline: the first reply, the notes after each viewing, and follow-ups timed around provisional holds.
- How to Write Wedding Flower Proposals With AI — A florist's method for AI-assisted wedding proposals: consultation brief, recipe costing, prompts for the words, and the seasonality check AI can't do.
- AI Mistakes That Put Couples Off Your Wedding Business — Eight ways AI makes a wedding supplier look careless or fake to couples, each with an example from real-looking enquiries and the fix.
- What Is Generative Engine Optimisation and Does Your Business Need It? — Where the GEO idea came from, how AI answer engines choose sources, a decision table by business type, and five checks that cover most of it.
- AI Tools and AI Development: The Complete 2026 Guide — the AI hub, including every tutorial in the AI-for-business series.
Sources: Google Search Central pages on AI features and optimising for generative AI in Search; Google's review content policies; OpenAI's crawler documentation; Perplexity's crawler documentation; Cloudflare's documentation on blocking AI bots.