How to Show Up in ChatGPT When People Ask for Recommendations

Your customers stopped Googling. They're asking ChatGPT instead.

Five years ago, a homeowner with a leaking water heater opened Google and typed "plumber Austin." Today, that same homeowner opens ChatGPT and asks: "Who's the best plumber in Austin for emergency water heater repair?"

ChatGPT responds with three to five specific recommendations. The homeowner picks one. They call. They book. They never see a single traditional search result.

This guide shows you exactly how AI search works, how ChatGPT decides which businesses to recommend, and what you need to do this week to start showing up. Whether you run a dental practice, a plumbing company, a law firm, or an e-commerce brand, the playbook is the same. It's called AI Search Optimization (also called AIO, or AEO — Answer Engine Optimization), and it's the most important marketing shift since Google launched in 1998.

Real queries people are asking ChatGPT right now

Let's ground this in reality. Here are actual queries we've seen customers ask ChatGPT, Perplexity, and Claude in the last 30 days:

For every one of these queries, ChatGPT returns a short list — usually three to five specific businesses, with names, addresses, key differentiators, and sometimes phone numbers. No blue links. No ad carousels. Just a recommendation, the way a knowledgeable friend would give it.

That's it. That's the whole game. Not rank #1 in Google. Not outbid your competitors for paid ads. Just make sure that when ChatGPT generates its answer, your name shows up in the list.

How ChatGPT (and Perplexity, and Claude) decide which businesses to recommend

AI search doesn't work like Google. There are no keyword density scores, no backlink counts plugged into a PageRank formula. Instead, language models ingest billions of pages, extract entities and facts, and when asked a recommendation question, generate an answer based on what they "know" about businesses in that category and location.

Here are the seven factors that determine whether ChatGPT will recommend you:

Factor 1: Structured data (JSON-LD schema)

The critical schemas for businesses seeking AI recommendations:

Without schema, AI has to guess what you do from reading prose. With schema, it knows. Guess which one ends up in the recommendation list.

Factor 2: Content that answers questions directly

Language models love answer-first content. That means the first sentence of a paragraph should directly answer the question implied by the heading. No lead-ins, no fluff.

If a customer might ask ChatGPT "how much does a root canal cost in Denver?" and your dental site has a page that answers exactly that in the first sentence, you become source material. The AI reads it, remembers it, and cites you. For the deep dive on this technique, see our guide on answer-first content structure.

Factor 3: Entity consistency (NAP everywhere)

AI models build an internal "entity" for your business by cross-referencing mentions across the web. If your business name, address, and phone number are consistent across your website, Google Business Profile, Yelp, Facebook, Apple Maps, BBB, and industry directories, the AI builds a single strong entity. If they're inconsistent, the AI gets confused and downgrades confidence — meaning it won't recommend you.

"Smith & Sons Plumbing LLC" on your site and "Smith and Sons Plumbing" on Yelp might look the same to you. To an AI disambiguation system, they might be two different businesses, neither of which it's confident enough to mention.

Factor 4: Reviews and ratings

ChatGPT heavily weights review signals when making recommendations. In our testing, businesses with 50+ reviews averaging 4.5 stars show up in AI answers roughly 4x more often than businesses with fewer than 20 reviews — even when the content quality is identical.

Two pieces of this matter:

Without both, you're leaving signal on the table.

Factor 5: Freshness

ChatGPT and Perplexity both strongly prefer recent content. Pages with a dateModified from the last 90 days get meaningfully more citation weight than pages that haven't been touched in two years. Stale content reads as abandoned content, and AI models avoid recommending businesses that look abandoned.

Update your hours seasonally. Refresh your service pages quarterly. Publish new content monthly. This isn't optional — freshness is a ranking factor.

Factor 6: Citations from authoritative sources

When a local news outlet, an industry blog, or a respected directory mentions your business, ChatGPT's training data picks it up. Mentions in Austin American-Statesman, Dentistry Today, or the Chamber of Commerce carry more weight than mentions in random blog comments.

Cultivate earned media. Get quoted in local news. Get listed in industry-specific directories. Every high-authority mention strengthens your entity profile.

Factor 7: llms.txt file

The newest and most underused tool: a plain text file at yoursite.com/llms.txt that tells AI crawlers exactly what your business does, which pages matter most, and how to describe you. Think of it as a briefing document you hand directly to ChatGPT.

We wrote the full guide on this: llms.txt: The AEO File That Tells AI About Your Business. If you haven't created one, you're skipping the easiest win in AI search.

The 28-point AIO checklist

Here's the complete technical checklist our team runs on every client site. Aim to tick every box.

1. Answer-first paragraph structure on all key pages
2. H2 headings that match real user questions
3. FAQ section on every major landing page

1. Clear entity disambiguation (city, state, specialty in titles)

1. Google Business Profile fully completed and verified
2. Consistent NAP across top 20 directories
3. 50+ reviews on primary review platforms
4. Active blog with monthly publishing cadence
5. Author bios with credentials and Person schema

1. Site speed under 2.5 seconds LCP
2. Mobile-first responsive design
3. HTTPS with valid certificate
4. Semantic HTML (proper heading hierarchy)
5. Alt text on all images with descriptive context

How to test whether AI already recommends you

Don't guess. Test. Here are three practical checks you can run today:

Open ChatGPT and ask the exact query your customer would ask. "Best [your business type] in [your city]." Then ask three variations. Note whether you appear. If you don't, note who does — those are your real AI search competitors.

Repeat the test in Perplexity and Claude. The results will differ slightly because each AI is trained on different data, but consistent absence across all three is a red flag.

Use WeLead Lab's free website scanner to catch the technical issues that block AI crawlers — missing schema, slow load times, broken robots.txt directives. It takes two minutes and flags the critical blockers.

If you've added schema markup, confirm it actually parses. Our free schema markup validator tells you exactly which types you have, which are missing, and which have errors that make AI crawlers ignore them. Broken schema is worse than no schema — it signals carelessness.

Common AIO mistakes small businesses make

After auditing hundreds of local business sites, the same five mistakes keep appearing.

Stuffing keywords into title tags does nothing for ChatGPT. AI doesn't count keyword density. It extracts facts and entities. Stop writing for algorithms that retired a decade ago.

We regularly find sites that block GPTBot or ClaudeBot because someone copied a robots.txt from Stack Overflow in 2023. Unblock them. If AI can't read you, AI can't recommend you.

About 70% of local business sites have zero JSON-LD schema. They're invisible to machines. See our breakdown of the most common AEO mistakes for the full list and fixes.

Your website says "Suite 200," Google says "Ste. 200," Yelp says "#200," and Facebook has an old suite number from before you moved. AI entity resolution collapses under this kind of noise.

The three highest-leverage AIO assets, missing from 90% of small business sites. Add them and you leapfrog most of your competitors in one week.

Quick wins you can implement today

If you only do five things this week, do these:

Total time: about two and a half hours. Expected result: meaningful AI visibility lift within 30 days as ChatGPT's crawlers revisit your site.

Want ChatGPT to recommend your business? We build the whole system for you.

If reading through the 28-point checklist made your eyes glaze over, you're not alone. Most small business owners don't have time to become schema experts, entity consistency auditors, and AI crawler whisperers on top of running their business.

That's what we do. WeLead Lab builds the entire AI search optimization system — new website, full schema, llms.txt, crawler config, content strategy, the works — for $500/month, with the website built free. No setup fees, no six-month contracts, no "managed SEO" vagueness. We make ChatGPT, Perplexity, and Claude recommend your business, and we show you the query tests to prove it.

FAQ

Typically 30 to 90 days. AI crawlers revisit sites on different schedules than traditional search engines. GPTBot and ClaudeBot tend to refresh popular sites weekly and smaller sites monthly. Once they recrawl and the new schema, content, and llms.txt are ingested, your entity profile updates and you start appearing in relevant recommendations.

No. The subscription is for users, not businesses. There is no paid inclusion program in ChatGPT, Perplexity, or Claude (yet). AI search is currently a pure merit system — you show up because the AI has good data about you, not because you paid.

Mostly no. The same foundation — schema, content structure, llms.txt, entity consistency, fresh content — works across all major AI search tools. Each model has quirks, but 90% of the work benefits all of them simultaneously. Build once, rank everywhere.

Probably not entirely, but the share is shifting fast. Current estimates put AI search at roughly 27% of all query volume and growing. A realistic 3-year outlook has AI search handling 50%+ of informational and recommendation queries, with Google holding navigational and transactional searches longer. Either way, ignoring AI search today is like ignoring mobile in 2010.

They're the same thing with different labels. AEO (Answer Engine Optimization) is the most common term. AIO (AI Optimization) is a newer umbrella term. AI search optimization is the plain-English version. All three refer to the practice of making your business discoverable by AI tools like ChatGPT, Perplexity, and Claude. Pick whichever acronym you prefer — the tactics are identical.