A regional insurance agency outperformed State Farm, Allstate, and GEICO on ChatGPT for local queries. Here's the David vs. Goliath playbook.
Introduction
State Farm has 19,000 agents. Allstate has 10,000. GEICO spends over $2 billion a year on advertising. Progressive and Liberty Mutual aren't far behind.
So how does a 12-person independent insurance agency in Columbus, Ohio end up being the name ChatGPT recommends when someone asks "Who's a good insurance agent in Columbus?"
Not by outspending the nationals. Not by out-advertising them. By out-specifying them.
This is the story of how a regional agency used one strategic insight to beat billion-dollar carriers on the fastest-growing discovery platform in the world. And why the same approach works for any local business competing against national brands in AI search optimization.
(Note: agency name and identifying details have been modified for confidentiality. The market, competitive dynamics, and results are based on real engagement data.)
The insight that changed everything
When someone asks ChatGPT "Who's a good insurance company?", the answer is predictable: State Farm, Allstate, GEICO, Progressive. These are entities with massive web presence, billions of mentions, Wikipedia pages, and decades of brand recognition. A regional agency can't compete at that level. Ever.
But here's what the agency's owner noticed: nobody was asking ChatGPT that question. His potential customers weren't asking "who's a good insurance company?" They were asking things like:
"Who's a good independent insurance agent in Columbus, Ohio?"
"Can you recommend an insurance agent near me who handles both home and auto?"
"I need an insurance agent for my small business in Columbus. Who should I call?"
These queries have three attributes that change the competitive dynamics entirely.
They're local. National carriers have strong generic entity recognition, but their local agent presence is fragmented across thousands of individual agents with inconsistent web presence.
They specify "agent," not "company." The user wants a person or office to talk to, not a 1-800 number. National carrier websites are designed for online quoting, not for generating "talk to a local agent" recommendations.
They include specific needs. "Handles both home and auto." "Small business insurance." "Independent agent." These specificity markers push AI toward matching detailed entity profiles, not generic brand names.
The strategic insight: compete for the specific query, not the generic one. Let State Farm own "best insurance company." Own "best independent insurance agent in Columbus who handles commercial and personal lines."
Building the entity for the specific query
The agency followed a focused 5-month strategy built around this insight.
Month 1: Entity definition and structured data.
They defined their entity precisely: "[Agency Name], an independent insurance agency in Columbus, Ohio, offering personal insurance (auto, home, umbrella), commercial insurance (general liability, commercial property, workers' compensation, business auto), and life insurance. Serving Columbus, Dublin, Westerville, Worthington, and surrounding Franklin County communities since 2011."
That description was engineered to match the specific query patterns their customers use. It includes: the "independent" differentiator, the city name, specific insurance types (matching how people ask AI about insurance), specific service-area communities, and founding year (trust signal).
They implemented Insurance Agency schema (a specific subtype of Local Business), Agent schema for each licensed agent, Service schema for each insurance line, and FAQ schema on their website.
Months 2 to 3: Hyper-local citation building.
Instead of chasing national insurance directories (where they'd be a tiny listing among thousands), they focused on local authority:
Columbus Chamber of Commerce, Dublin Chamber of Commerce, Westerville Area Chamber, Worthington Partnership, Franklin County business directories, Ohio Department of Insurance agent lookup (verifiable credential), Independent Insurance Agents & Brokers of Ohio (IIABO) member directory, Trusted Choice agency locator, BBB, and 15 additional local business and community directories.
They also secured mentions in two local publications: a Columbus Dispatch small business resource roundup and a ThisWeek Community News article about independent agents vs. captive agents.
Total citations by Month 3: 38, with a heavy concentration in Columbus-specific and Ohio-specific sources. This geographic signal density was critical: it told AI tools that this was the definitive independent insurance entity for Columbus.
Months 3 to 4: Content targeting the queries nationals can't answer.
They published 10 pieces of content that specifically addressed questions where national carriers' websites are useless:
"Independent vs. Captive Insurance Agents in Columbus: Which Is Right for You?"
"Small Business Insurance in Columbus, Ohio: What You Need and What It Costs"
"How to Bundle Home and Auto Insurance in Ohio (And Whether It Actually Saves Money)"
"Columbus Homeowner's Insurance: What Influences Your Rate and How to Lower It"
"Workers' Comp in Ohio: A Small Business Owner's Guide"
Each article positioned the agency as the local expert. National carrier websites publish generic insurance education content. This agency published Columbus-specific, Ohio-specific content that AI tools could reference when someone asked a local question.
Months 3 to 5: Review diversification.
The agency already had 85 Google reviews. They added active review collection on Yelp (grew to 18 reviews), BBB (grew to 12 reviews), and Facebook (grew to 24 recommendations). Several reviews specifically mentioned "independent agent" and "Columbus" in the review text, reinforcing the entity signals.
The result: local beats national
By Month 5, here's what happened when someone asked AI about insurance in Columbus:
Query: "who's a good insurance agent in columbus, ohio?"
ChatGPT's response: Named the agency first, describing them as "an independent insurance agency serving Columbus and surrounding communities, offering personal and commercial insurance lines." Then mentioned that "national carriers like State Farm and Allstate also have agents in the Columbus area."
The regional agency was named before the national carriers. Not because it was bigger. Because it was more specifically relevant to the query.
Query: "can you recommend a small business insurance agent in columbus?"
All three platforms named the agency. None named a national carrier. The national carriers' websites aren't structured to answer "who's a good agent for small business insurance in Columbus" because their entity data is organized around the corporate brand, not individual agents.
Query: "what's the best insurance company?"
State Farm, Allstate, GEICO, Progressive. As expected. The agency didn't appear for generic national queries. That was never the goal.
The agency didn't beat the nationals at everything. It beat them at the queries that actually produce local customers.
Why specificity beats scale in AI recommendations
This case study illustrates a principle that applies far beyond insurance: AI recommends the most relevant entity for the specific query, not the biggest entity in the category.
National brands have enormous generic entity authority. But their entity data is broad, not deep. State Farm is known for "insurance." The regional agency is known for "independent insurance in Columbus for small businesses and homeowners." When the query is specific, the specific entity wins.
This works because AI tools are trying to provide the most helpful answer. Recommending State Farm to someone who asked for an independent agent in Columbus is a bad answer. AI tools actively try to match specificity when the query provides enough context.
Every local and regional business competing against national brands has this same opportunity. The nationals own the generic queries. You can own the specific ones. And the specific ones are where the customers are.
Competing against national brands in your market? Run your free AI visibility audit at yazeo.com and find out what queries AI recommends you for, what queries go to the nationals, and where the specific query opportunities are that nobody has claimed yet.
Key findings
A 12-person regional agency outperformed billion-dollar national carriers on ChatGPT for local, specific insurance queries.
The strategic insight was competing for specific queries ("independent insurance agent in Columbus") rather than generic ones ("best insurance company").
38 geographically concentrated citations on Columbus and Ohio-specific sources created stronger local entity authority than the nationals' broad but thin local presence.
Content targeting local-specific questions gave AI tools answers that national carrier websites couldn't provide.
By Month 5, the agency was named before national carriers in AI responses to local queries, generating new client inquiries at zero per-acquisition cost.