Local businesses, generative answers, and the next decade of customer acquisition

If you run a service business that sells inside a thirty-mile radius — a contractor, a clinic, a law firm, an HVAC company, a restaurant — the next ten years of customer acquisition will be decided by whether your name comes up when a buyer asks an AI assistant who should I call.

This is already happening. The shift is just easier to spot in some categories than in others.

I want to walk through how local search is changing specifically, because the dynamics are different from B2B and the work is different.

The new local journey

A buyer with a problem in 2026 — say, a homeowner whose water heater starts leaking on a Saturday morning — used to do this: open Google, type plumber near me, scan the local pack, click two or three of the top results, and call one.

Now they do this: open ChatGPT or the AI Overview at the top of Google’s results, ask “who should I call for a leaking water heater in [their city]”, read the response, and call the first business named. They never opened the local pack. They didn’t compare three options. They asked an assistant and acted on the answer.

This compresses the entire local-search funnel into a single decision moment. The business named first in the AI response wins. Second place is invisible. Third place doesn’t exist.

What AI Overviews actually evaluate for local

AI Overviews on Google — the most measurable of the AI surfaces for local — pull from a specific combination of signals. Understanding the combination is the work.

First: Google Business Profile completeness. AI Overviews lean heavily on the entities Google already understands. A complete, claimed, verified GBP with hours, services, address, photos, reviews, and Q&A is the foundational asset. A GBP that’s half-filled or claimed by someone who left two years ago is an automatic disqualification.

Second: review velocity and recency. Not total review count — though that matters — but how recent the most recent reviews are, how varied they are, and whether the responses to them come from the business. A business with three hundred reviews from 2019 and nothing since loses to a business with thirty reviews from the past ninety days. The model reads this as currency.

Third: schema markup on the business’s own website. LocalBusiness or ProfessionalService schema with the address, hours, services, and aggregate rating. Most local businesses’ websites have none of this. The ones who do appear in AI Overviews disproportionately to their size.

Fourth: editorial mentions. A local newspaper article that names the business. A trade-press piece. A blog post by a nearby influencer or community organization. Each editorial mention is a citation the model can use to triangulate the business’s reputation in the local context.

Fifth: structural answer pages on the business’s own website. If a homeowner is asking “what does a leaking water heater repair cost in [city],” and the local plumber has a page titled exactly that, with a clear answer and a service-area schema, the model retrieves and cites that page. Most local businesses have no such pages. The ones who publish them dominate.

The compounding effect at the local level

Local AI search compounds faster than national or B2B search, because the competitive set is smaller. There are thirty plumbers in a typical metro. The top five capture the bulk of demand. To be in the top five named by AI assistants, a business doesn’t need to outrank a global category — it needs to outrank twenty-five competitors, most of whom are doing none of the work above.

This is why local is, paradoxically, the easiest category to win in AI search right now. The bar is lower. The competitive set is smaller. The compounding kicks in faster. A local business that starts the program this quarter can be the default answer in their metro by the end of the year.

The flip side: once a business is the default, displacing them is an eighteen-month project for any competitor. The local AI-search winners of 2026 will be category-defining brands in their metros through at least 2030.

What this means for the marketing director of a local business

The shift in tactical priorities is concrete.

Stop spending on Google Ads bid-ups in your local pack. The clicks are getting more expensive every quarter and they go to a smaller share of buyers every quarter.

Start publishing answer pages on your own website. One page per question your buyers actually ask, schema-tagged, indexed within hours of publish, on a regular cadence. Twenty pages in the first quarter is a reasonable start. Thirty a month is the program.

Audit your Google Business Profile this week. Fill every field. Respond to every review of the past two years. Update photos. Add the service-area Q&A. The GBP is doing more work in AI Overviews than it’s done in the entire history of local SEO, and most businesses haven’t noticed.

Get cited locally. Sponsor a community event. Get profiled by the local paper. Show up in industry forums. Each editorial mention is a citation surface the model can use.

Track which prompts mention you each quarter. “Who is the best [your service] in [your city]” is the foundational prompt. There are about a hundred variants. Run them, monthly, against five AI platforms. The number that comes back is the only metric in local-AI search that matters.

The decade ahead

The businesses that figure this out before their competitors will spend the back half of the 2020s as the default recommendation in their categories. The businesses that wait will spend that period trying to dislodge incumbents who started a year or two earlier.

There is no shortcut. The work is the work. But the leverage of doing the work right now, when most local businesses haven’t even noticed the shift, is the highest it will ever be.

---

Luke LaFave is the founder of LaFave Consulting. The studio works with local service businesses on the exact program described above — GBP, schema, answer-page publishing, and monthly AI-citation auditing.