Local search has been changing for years. Voice search was supposed to change everything. Featured snippets were supposed to change everything. AI-generated answers are the latest version of the same shift — and this one is real.
When someone asks ChatGPT to recommend an electrician in their city, or asks Perplexity for a good accountant nearby, they get an answer. Whether your business is in that answer depends on a handful of factors that most local business owners haven't had reason to think about yet.
The queries that matter for local businesses
Local AI search happens in two forms.
The first is explicit: "recommend a good HVAC company in [city]" or "who are the best family dentists in [area]." These are the easiest to target, because the intent is clear.
The second is category-level: "what should I look for in a plumber?" or "how do I choose a local accountant?" Here the model gives criteria first, then sometimes names providers. Getting into these responses is harder, but the brand association it builds is durable.
Both matter. Most local businesses only think about the explicit version.
What determines who shows up
For local queries, AI models draw heavily on:
Aggregator presence. Yelp, Google Business Profile, Angi, HomeAdvisor, Houzz, Thumbtack — depending on your trade. Models retrieve from these because they rank well for local service queries and aggregate real customer signal. If you're thin or absent on these platforms, you're thin or absent in AI responses too.
Review volume and recency. A business with 200 reviews on Google is a more established data point than a business with eight. The model's confidence in recommending you is higher because the signal is stronger.
Local editorial coverage. Mentions in local newspapers, regional business journals, and local blogs establish geographic context. A regional publication saying "the best roofers in Northern Virginia include..." is directly useful to a model trying to answer that query.
Chamber of commerce and local directory listings. These carry moderate weight, but consistent NAP (name, address, phone) data across directory sites helps models resolve your identity with confidence.
What doesn't work
Paid local directory listings that have no traffic and no editorial standing are low value. A presence on 200 random citation sites with identical, thin descriptions is not equivalent to 20 reviews on Google and a mention in the local business journal.
Your own website also has less pull than you'd expect for local AI search — not because it doesn't matter, but because models are more persuaded by third-party signal than self-description.
Where to start
If you haven't done this already:
- Claim and complete your Google Business Profile. Make sure the category, description, and service area are accurate. Respond to reviews.
- Get onto the aggregator relevant to your trade. An HVAC company should be on Angi and HomeAdvisor. A restaurant should be on Yelp. A law firm should be on Avvo and Justia.
- Ask satisfied customers for reviews. Volume matters. Ten new reviews over six months is more useful than five reviews from three years ago.
- Look for a local press mention. A single citation from a credible local publication establishes geographic context that no number of directory listings can replicate.
Know your baseline
Before doing any of this, it's worth knowing your current AI visibility across the major models. You might already be appearing for some queries and not others. Knowing the gap tells you where to focus effort.
Run your free audit and see how your local business appears across ChatGPT, Claude, Perplexity, and Gemini.