How do I get my small business recommended by ChatGPT?
By Shaeel Iqbal · Published October 13, 2026 · Last updated October 13, 2026
The short answer
There is no setting to switch on and no way to guarantee a recommendation. What you can do is make your business easy for ChatGPT to find, understand and trust: check what it says now, let its search crawler read your site, state what you do plainly, keep those facts consistent, and earn real mentions on other sites.
1. Ask ChatGPT the questions your customers ask
Open a new chat with search on and ask the way a customer would: "best [what you sell] in [your city]" or "who can help me with [problem]?" Write down whether you are named, which businesses are, and which sites it cites.
Ask each question three or four times. A SparkToro study of 2,961 runs across ChatGPT, Claude and Google's AI found that for ChatGPT and Google's AI the chance of getting the same list of brands twice was under 1 in 100 (Claude was only slightly more likely to repeat), although brands that appear often kept appearing.[1] One answer tells you little. How often you show up across several runs tells you more. (The study is not peer reviewed and was run with Gumshoe.ai, an AI-tracking company, which the author disclosed.)
Our free AI Visibility Test gives you a first look (no account required; it asks for an email address).
2. Let ChatGPT's crawler read your site
OpenAI uses a crawler called OAI-SearchBot to surface sites in ChatGPT's search features, and recommends allowing it in your robots.txt file if you want to appear in those results. It is a separate setting from GPTBot, which OpenAI uses to gather training data, so you can allow one and block the other. Changes take about 24 hours to apply.[2]
Check yours: open yourdomain.com/robots.txt and look for a blanket Disallow: / or a rule that names OAI-SearchBot. Two other basics: the text on your key pages should appear without scripts having to run, and your site should be indexed by Google and Bing. Both are free to check with Google Search Console and Bing Webmaster Tools.
3. Say plainly what you do, where, and for whom
AI answers are built from text a model can read and quote. Put a direct statement near the top of each key page. An invented example: "Maple Street Plumbing is a licensed plumbing company serving Columbus, Ohio, specializing in emergency repairs." Don't leave this to a slogan or an image.
The best-known research is "GEO: Generative Engine Optimization" (Aggarwal et al., KDD 2024). The authors tested nine content changes on 10,000 queries. Adding citations, quotations from credible sources and statistics raised visibility in generative-engine answers by up to about 40%. Keyword stuffing did not help. It is one benchmark on the engines of 2023-24, and effects vary by subject, so it points a direction; it does not promise a result.[3] Google's July 2026 guide points the same way: unique, useful content over tricks.[4]
So where you have real numbers, state them and say where they came from.
4. Keep your facts the same everywhere
Use one description of your business, and the same name, address, phone number and hours on your website, Google Business Profile, social profiles and directories. Conflicting details make it harder for a model to know which one is right.
Structured data (schema.org markup) can state these facts explicitly. Google says it isn't required for its generative AI search, so treat it as tidiness, not a lever.[4] If another business has a similar name, say who you are on your About page. We learned this the hard way: when we asked ChatGPT (search on) "What is Citefast?" on Sept 30, 2026, it described an unrelated citation tool at citefast.com (one run). More on that in Why doesn't ChatGPT mention my business?
5. Earn real mentions on other sites
AI answers draw on what the web says about you, not only on what you say. In a study of 75,000 brands, Ahrefs found that how often a brand was mentioned across the web correlated more strongly with appearing in Google's AI Overviews (0.664) than backlinks did (0.218). That is correlation, not proof, and it is about Google's AI Overviews, not ChatGPT.[5]
It still tells you where effort tends to go: customer reviews, local and industry directories, suppliers and partners who list you, local press, and honest answers in communities where your customers ask questions. Google says seeking inauthentic mentions isn't as helpful as it might seem[4], so no bought links, fake reviews or paid-for reviews.
6. What isn't worth your time
- Keyword stuffing. It did not help in the GEO study.[3]
- Anyone who guarantees ChatGPT placement.
- Fake or incentivised reviews.
- llms.txt as a ranking lever. Google says Google Search doesn't use it.[4] See What is llms.txt, and do you need it?
7. How long it takes, and how to tell if it is working
When ChatGPT searches the web to answer, changes to your pages and to the sites that mention you can show up within weeks. When it answers from what it learned in training, changes only appear as models are updated, which takes months. Nobody publishes timelines, so treat these as expectations, not promises.
Re-ask your questions on the same day each month, record how often you're named across several runs, and look for visits from chatgpt.com in your analytics.
Common questions.
No. Answers change between runs, between models and over time. In SparkToro's study, ChatGPT and Google's AI gave the same list of brands less than 1 time in 100. Be wary of anyone who promises placement.
There is no results list to climb. ChatGPT writes an answer and names a few businesses, and the names change from run to run. Instead of a position, track how often you are named across repeated questions.
Being cited means your page is linked as a source. It starts with ChatGPT's crawler being able to read your page, and with the page stating something clear and checkable that is worth citing: facts, numbers with named sources, and information that isn't available elsewhere.
Neither is a proven lever. Google says it doesn't use llms.txt and that structured data isn't required for its generative AI search. We keep both as tidiness.
AI tools generate each answer fresh, and the brands named vary a lot between runs. That is why we check each question several times.
Sources
- SparkToro, "NEW Research: AIs are highly inconsistent when recommending brands or products; marketers should take care when tracking AI visibility", January 28, 2026.
- OpenAI, "Overview of OpenAI Crawlers", read October 13, 2026.
- Aggarwal et al., "GEO: Generative Engine Optimization", KDD 2024 (arXiv preprint, 2023).
- Google Search Central, "Google's Guide to Optimizing for Generative AI Features on Google Search", last updated July 10, 2026.
- Ahrefs, "An Analysis of AI Overview Brand Visibility Factors (75K Brands Studied)", May 26, 2025.
Want to see where you stand?
If you'd rather have this done for you, see our generative engine optimization services.