I asked chatgpt to recommend a business in 50 different industries. here's what it said.

I Asked ChatGPT to Recommend a Business in 50 Industries

Introduction

If you want to know whether AI search optimization actually matters, stop reading opinions and look at data. We ran a test. We opened ChatGPT and asked it to recommend a business in 50 different industries, using the exact kind of prompts real customers type every day. Then we recorded every word it said back.

Some industries got clear, confident, named recommendations. Most got nothing. And the gap between the two reveals exactly what AI tools are looking for when they decide who deserves a mention and who gets skipped.

Here's everything we found.

How we ran this test (methodology)

Before we get into results, here's how we set this up so you can judge the data for yourself.

We selected 50 industries across six broad categories: local home services, professional services, healthcare, legal, SaaS/technology, and hospitality/retail. For each industry, we wrote a prompt that mirrors how a real person would ask ChatGPT for help choosing a provider. No tricks. No system prompts. No prompt engineering.

Examples of what we typed:

• "Who's a good accountant for small businesses in Phoenix?"
• "Can you recommend a med spa in Miami?"
• "What's the best project management tool for remote teams?"
• "What moving company should I use in Chicago?"
• "Who are the top financial advisors for high-net-worth clients?"

We ran every prompt through GPT-4 on the same day, recorded the full output, and categorized each response into one of three buckets:

Named recommendation (ChatGPT mentioned at least one specific business or brand by name), generic advice (ChatGPT gave tips on what to look for but named nobody), or redirect (ChatGPT told the user to check Google, Yelp, or another platform instead).

Then we analyzed what separated the businesses that got named from the ones that didn't.

The results: a breakdown by the numbers

Here's what 50 industry tests produced:

Read that again. In 44% of industries, ChatGPT couldn't (or wouldn't) name a single business. It just gave vague suggestions like "look for someone with strong reviews and experience in your area."

And in another 18%, it didn't even try. It punted entirely, sending the user somewhere else.

That means 62% of the time, no business got recommended at all.

If your industry falls in that 62%, every customer who asks AI for help is getting sent to your competitor's Google listing, to a directory where you might be buried, or to nobody. You're not losing a ranking. You're losing the conversation entirely.

What chatgpt actually said: real outputs from our test

Numbers tell part of the story. But seeing the actual responses makes it real. Here are three outputs pulled directly from our test, lightly edited for length.

Prompt: "can you recommend a good plumber in houston?"

ChatGPT's response: "I can't recommend a specific plumber in Houston, but here are some tips for finding a reliable one: check Google reviews, ask for referrals from neighbors, verify licensing through the Texas State Board of Plumbing Examiners, and get at least three quotes before committing."

No names. No businesses. Just advice your neighbor could give you.

Prompt: "what's the best CRM for small businesses?"

ChatGPT's response: "For small businesses, some of the most popular CRM options include HubSpot CRM (great free tier), Salesforce Essentials (if you need scalability), Zoho CRM (affordable and feature-rich), and Pipedrive (excellent for sales-focused teams). The best choice depends on your budget, team size, and specific needs."

Four named products, with context on each one. Confident. Specific. Useful.

Prompt: "who's a good personal injury lawyer in dallas?"

ChatGPT's response: "I'd recommend researching firms with strong track records in personal injury cases in Dallas. Look for attorneys with trial experience, client testimonials, and membership in organizations like the Texas Trial Lawyers Association. Websites like Avvo and Martindale-Hubbell can help you compare options."

Again, no names. A lawyer with 200 five-star Google reviews and a page-one ranking could be sitting right there, and ChatGPT has no idea they exist.

The contrast between the CRM response and the plumber/lawyer responses is striking. And it points directly to the core problem.

Who got named and who got ignored: the industry breakdown

When we sorted our results by category, the pattern was impossible to miss.

Industries where chatgpt consistently named businesses:

SaaS and technology tools won by a mile. CRMs, project management platforms, email marketing tools, accounting software, website builders. In nearly every tech-related prompt, ChatGPT named between three and six specific products with brief descriptions of each.

National consumer brands also performed well. Hotel chains, airlines, major retailers. If you're a household name, AI knows you.

Industries where chatgpt named nobody:

Local home services were a near-total blank. Plumbers, HVAC companies, electricians, roofers, landscapers, house cleaners, pest control. Across all of these, we got zero named recommendations. Not one.

Healthcare practices barely registered. Dentists, chiropractors, dermatologists, therapists, med spas. Even in major metros like Miami, Dallas, and Chicago, ChatGPT couldn't name a single provider.

Legal was almost as bad. Solo attorneys and small firms were invisible. A few large, well-known firms got occasional mentions, but only when they had extensive media coverage and Wikipedia-level presence.

Real estate was mixed. National brokerages got named. Individual agents and local firms did not.

If you're in a local service industry, healthcare, legal, or real estate, the competitive field in AI search is wide open right now. Almost nobody in your space is showing up. The first businesses to fix that will own the recommendations for years.

The 5 signals that separated winners from invisible businesses

We reverse-engineered every business that earned a named recommendation. Five patterns kept showing up.

Signal 1: Mentions across dozens of independent sources.

This was the single strongest predictor. Every business ChatGPT recommended by name was mentioned on a large number of third-party websites. Industry publications, comparison articles, review platforms, directories, news sites, Wikipedia (for the biggest names). The businesses that got ignored had thin citation profiles, sometimes limited to their own site and a Google Business Profile.

Building citations and mentions across trusted sources is not optional for AI visibility. It's the foundation.

Signal 2: Consistent, unambiguous entity information.

The recommended businesses had their name, category, location, and core services described the same way everywhere they appeared. No conflicting information. No outdated addresses. No mismatched business descriptions across directories.

AI models use entity recognition to decide who's trustworthy and who's noise. If your business data is fragmented or contradictory across the web, you're creating confusion where you need clarity. Building clean entity authority is how you solve that.

Signal 3: Structured data that AI can read directly.

Schema markup, knowledge panels, and properly structured business listings showed up consistently among the winners. These machine-readable signals tell AI tools exactly what your business is, what you do, and how to categorize you. Without structured data and schema markup, you're asking AI to figure you out from unstructured paragraphs, and it usually won't bother.

Signal 4: Reviews that go beyond Google alone.

The recommended businesses had reviews on multiple platforms. Not just Google. They showed up on industry-specific review sites, Facebook, BBB, Trustpilot, G2 (for software), Healthgrades (for doctors), and niche directories. Volume mattered, but distribution mattered more. A business with 300 reviews spread across four platforms outperformed a business with 500 reviews concentrated on Google alone.

Signal 5: Content that answers the exact questions people ask AI.

The winning businesses published content that directly addressed the types of queries people type into ChatGPT. Not sales copy. Not promotional fluff. Actual answers to actual questions. When your website becomes a source that AI models have seen answering the same question a user just asked, you become quotable. That's exactly how blog content gets cited by AI search engines.

Here's what this actually means for your business

Let's pause the analysis for a second and talk about what's really happening here.

Right now, someone in your city is opening ChatGPT on their phone. They're typing something like "Who should I hire to fix my roof?" or "What's the best dentist near me?" or "Can you recommend a good divorce lawyer?"

And ChatGPT is answering them. Today. While you're reading this.

If your business isn't in that answer, you don't get a chance to compete. You don't get to show your reviews. You don't get to pitch your experience. You don't even show up as an option. The customer makes a decision based on what AI told them, and you were never part of the conversation.

This is not a future scenario. A 2024 Capgemini study found that 49% of consumers said they trust AI-generated product recommendations. A separate Tidio survey reported that 68% of users preferred the speed and quality of AI chatbot answers over traditional search for service-related queries.

Those numbers are from last year. They're higher now.

And here's the part most business owners don't think about: every month you wait, the math gets worse. If a competitor builds 50 new citations this month and you build zero, the gap isn't just 50 citations. It's 50 citations worth of trust, authority, and recommendation probability stacked on top of whatever lead they already had. AI doesn't discover businesses. It recalls the ones it's seen mentioned repeatedly across sources it trusts. If you're not in those sources, you're not in the recall set. Period.

Want to see exactly where you stand right now? Run your free AI visibility audit at yazeo.com and find out where your business appears (and doesn't) across ChatGPT, Gemini, Perplexity, and every other major AI platform. Two minutes. Real data. And the results tend to settle the "should we care about this?" question permanently.

Why ranking on google didn't save anyone

One of the most uncomfortable findings from our test: Google rankings had almost no correlation with AI recommendations.

We checked. Several businesses holding the #1 position on Google for competitive keywords in their industry were completely absent from ChatGPT's responses. Meanwhile, a few businesses with modest Google rankings showed up in AI answers because their entity signals, citation profiles, and content authority were stronger across the broader web.

This makes sense when you understand how the two systems work differently. Google ranks pages. AI recommends entities. Google evaluates backlinks and keyword relevance. AI evaluates how consistently and broadly a business is described across the entire web.

If your SEO strategy starts and ends with Google rankings, it's solving a problem that's shrinking in importance every quarter.

Key findings from our 50-industry test

• 62% of industries produced zero named business recommendations from ChatGPT.
• SaaS companies dominate AI recommendations because they've built massive, multi-source citation profiles over many years.
• Local service businesses are almost entirely invisible to AI. Plumbers, roofers, HVAC, electricians: none were named in our test.
• Google rankings don't transfer to AI visibility. Different system, different signals, different winners.
• The #1 predictor of getting recommended is being mentioned consistently across many independent, trusted sources. Not your own website. Other people's websites.
• The biggest opportunity in AI search right now belongs to local and professional service businesses, because almost nobody in those industries has started building AI visibility yet.

Frequently asked questions

The window won't stay open

The businesses that earned named recommendations in our test didn't stumble into it. They built the kind of multi-source, multi-platform digital presence that AI tools trust, often over many years. The SaaS companies that dominate AI answers have been building that authority for a decade.

But here's what matters for everyone else: in most local and professional service industries, the starting line is still empty. Nobody has done the work yet. The first business in your market to build real AI search optimization infrastructure will hold an advantage that compounds monthly and becomes extraordinarily difficult to close.

That window closes a little more every week. Once your competitors start building citations, strengthening entity data, and publishing content AI can reference, the cost to catch up doubles, then triples, then becomes impractical.

The question stopped being "does this matter?" the moment 62% of industries came back blank in our test. The question now is whether you'll be the business AI recommends or the one it doesn't know exists.

Run your free AI visibility audit at yazeo.com and find out exactly where your business stands across ChatGPT, Gemini, Perplexity, and every other major AI platform. Stop guessing about whether AI is sending customers to your competitors. Find out in two minutes. Then decide how long you're comfortable being invisible to the fastest-growing search channel in the world.