How to get your business recommended by ChatGPT

How does ChatGPT decide which businesses to recommend?

When you ask ChatGPT for a business, it usually searches the web first, reads what it finds and writes an answer with links to its sources. OpenAI runs a crawler for this, OAI-SearchBot, and its crawler documentation says sites that opt out of it “will not be shown in ChatGPT search answers.” Everything else follows from that. An AI app can only recommend what it can find, read and trust.

Where ChatGPT's results come from is still debated. A 2025 Seer Interactive study found that 87% of ChatGPT search citations matched Bing's top results for the same question, on a sample its authors call limited and directional. Later that year Profound reported that ChatGPT's citations were lining up more with Google and less with Bing. And an Ahrefs study of 15,000 prompts found that only 12% of the URLs cited by ChatGPT, Gemini and Copilot ranked in Google's top 10 for the same prompt.

The studies measure different things, so there is no side to pick. The practical reading is simple: be easy to crawl, be indexed and ranking in both Google and Bing, and be described the same way everywhere. The five steps below do exactly that, in order.

Step 1. Let the search crawlers in

Each AI company runs separate bots for search and for training, and documents them separately. Blocking a training bot is a decision about training. Blocking a search bot takes you out of that app's answers. OpenAI says it plainly: each of its crawler settings “is independent of the others.”

Check two places. The first is your robots.txt file, which lives at yourdomain.com/robots.txt. The second is your firewall or CDN, which can block crawlers on its own settings without anyone on your side deciding to. If you want to stay out of training but stay in the answers, the file can say exactly that:

# Let AI search crawlers in
User-agent: OAI-SearchBot
Allow: /

User-agent: Claude-SearchBot
Allow: /

User-agent: PerplexityBot
Allow: /

# Keep your pages out of model training
User-agent: GPTBot
Disallow: /

User-agent: ClaudeBot
Disallow: /

Think twice before you add Google-Extended to that list. Google says it does not affect inclusion or ranking in Google Search, but it does govern whether Gemini can use your pages, including for grounding answers. Blocking it is a trade-off, not a free opt-out. OpenAI says robots.txt changes take about 24 hours to reach its search systems.

Step 2. Be indexed in Google and Bing

An AI answer is built from search results, so a page that is not indexed cannot be cited. For Google's AI features this is written down: Google's guide to AI features says a page must be “indexed and eligible to be shown in Google Search with a snippet,” and that there are no additional requirements.

Verify your site in Google Search Console and in Bing Webmaster Tools, which can import your Search Console setup in a few clicks, then submit your sitemap to both. Bing also supports IndexNow, an open protocol your site can use to tell participating search engines the moment a page is added or changed.

There is no paid fast lane into Google. Its Indexing API only accepts job postings and livestream video pages, and Google's own guide to hiring an SEO tells you to beware of anyone who advertises a “priority submit.” To check a page, paste its address into the URL Inspection tool in Search Console. If it is not indexed, find out why before you write anything new.

Step 3. Answer the question on your own site

Write the pages a customer's question would land on, and put the answer in the first two sentences. Use the questions people already search for you, which you can see in Search Console's Performance report, as your headings. Then state the facts only your business can give:

• the towns you serve, and how far you travel
• how fast you respond, and your hours
• your licences, insurance and warranty
• what a typical job involves and, if you choose to publish them, your prices
• real photos of your own work

The 2024 research paper that coined the term GEO, by Aggarwal and others, found that adding sources, statistics and quotations could make content up to 40% more visible in AI-generated answers on its test set. That is a lab benchmark, not a promise, but the direction holds: clear, specific, sourced pages are easier to quote.

You do not need a special format for AI. Google says “You don't need to create new machine readable files, AI text files, or markup to appear in these features.” A well-written page that answers the question is the format.

Step 4. Be described well beyond your website

AI apps read more than your site. Listings, directories, review sites and articles that mention you all describe your business, and an app that recommends local businesses can draw on any of them. Keep your name, address, phone number, hours and categories identical everywhere you are listed, starting with your Google Business Profile.

Reviews belong here too, collected honestly. Ask every customer, not only the happy ones, and never offer anything in exchange. Google's Business Profile policy prohibits incentives for reviews and selectively asking for positive ones, and the FTC's rule on reviews bans fake reviews, including AI-generated ones.

Skip any service that posts about you on Reddit or in forums with bots. Platforms treat that as spam, and a business caught at it has a harder problem than being left out of an answer.

Step 5. Measure how often you are named, not a “rank”

AI answers change from one run to the next. In a SparkToro and Gumshoe study of 2,961 runs, ChatGPT and Google's AI returned the same list of brands less than once in a hundred runs of the same prompt. A screenshot of one answer tells you very little.

Measure a rate instead. Pick 10 to 20 questions your customers actually ask, run each one many times a week across several AI apps, and record how often you are named and how often your site is linked. Average it over a few weeks before you read anything into it, and always keep the number of runs beside the number.

Two free signals help. Bing Webmaster Tools now has an AI Performance report, in public preview since February 2026, that shows how often your pages are cited in Copilot and Bing's AI answers and which searches led to them. And in Google Analytics, visits from ChatGPT arrive with chatgpt.com as their source, so you can see whether the mentions turn into visits.

What does not work?

• An llms.txt file as a ranking trick. Google says you do not need “AI text files” to appear in its AI features, and no major AI company has confirmed reading llms.txt. It costs nothing to publish one. Do not pay anyone for it.
• Schema markup as a shortcut. Google says there is “no special schema.org structured data that you need to add,” and when Ahrefs tracked 1,885 pages that added schema, AI citations barely moved. Keep your schema accurate because it is good hygiene, not because it is a trick.
• A guaranteed AI ranking. Answers vary on every run, so nobody can guarantee one. Google's own guide adds that “No one can guarantee a #1 ranking on Google.”
• Mass AI pages. Google's spam policies name “using generative AI tools or other similar tools to generate many pages without adding value for users” as scaled content abuse.

How does Settr Search do this every week?

Settr Search runs these five steps as a team of AI agents. The Answers agent checks crawler access and runs your customers' questions across AI apps every week. When it finds a gap, it hands the work to the agent that owns it: Content drafts an answer-first page from your facts, Links finds a listing you are missing, and Local runs your review requests. You approve anything that goes public, and every change is logged with a before and after.

What is GEO explains how AI apps choose their sources and how we measure mentions. What is AEO covers Google's AI Overviews and AI Mode.