Most business owners treat a wrong AI answer as a minor annoyance — the kind of thing you notice, shrug at, and move on from. Here is a simple back-of-napkin calculation that shows why that shrug is expensive.
The math has three moving parts: how many people are asking AI about your business each month, how many of those queries produce a wrong or missing answer, and what each lost opportunity is actually worth to you.
Step one: estimate your monthly AI query volume
People use ChatGPT, Perplexity, and Google AI Overviews to ask questions like "best family dentist near me open Saturdays" or "24-hour plumber in Riverside" hundreds of thousands of times a day. For any individual business, the slice of that traffic is smaller — but not negligible.
A rough estimate for a mid-size local business: between 50 and 300 people per month are having a conversation with an AI assistant where your business name or category and location come up. A busy dental practice in a city of 150,000 people is probably closer to 100–150 per month. A specialty contractor in a smaller market might be at 30–50. If you have no idea where you stand, a free scan will surface the answers AI is actually giving about your business right now.
Step two: figure out how many wrong answers become lost customers
Not every wrong AI answer loses you a customer. Some people double-check; some call anyway. But a meaningful share don't — they move to the next result, or they assume the AI is right and show up at the wrong time, or they simply pick a competitor whose information looked cleaner.
A conservative estimate: when AI gives wrong or missing information, about 20–25% of the people who saw that answer will not reach you.
Take a dental practice called Riverside Family Dental. Suppose 120 people a month ask an AI assistant about dentists in the area, and the AI mentions them — but also says their Saturday hours are "unavailable" or lists an old phone number. At 25%, that's 30 people who never called.
Step three: multiply by what a new customer is worth
This is where it starts to sting.
A new patient at a dental practice pays roughly $200–300 for the first visit. But that's not the number that matters. The number that matters is lifetime value: a patient who returns for cleanings, fillings, and the occasional crown over five or more years is worth $2,000–4,000 to that practice. Most dentists keep a patient for seven to ten years.
So those 30 people who never called: at a conservative $2,000 lifetime value each, that's $60,000 in forgone revenue sitting inside one wrong AI answer — and that's per month.
Cut the estimate in half to account for people who would have left for other reasons. You're still looking at $30,000 a month.
For a plumbing company the math is different but no kinder. A single call-out averages $400–600. A household that's satisfied with their plumber calls back every 18 months for a decade, plus refers two or three neighbors. If AI says "no information available on after-hours service" and ten homeowners with failing water heaters go elsewhere, that's 10 jobs × $500 = $5,000 in immediate work, plus the referral chain those customers will never send your way.
The trust penalty nobody accounts for
There's a secondary cost that doesn't show up in any calculation: the friction that wrong AI information creates even when someone still reaches you.
A customer who arrives during hours the AI told them you were closed, or who spent ten minutes calling a disconnected number the AI listed, arrives slightly less confident before you've said a word. That's not fatal, but it's a headstart in the wrong direction at exactly the moment you need them to say yes.
The reasons AI gets business information wrong are usually fixable — stale citations, missing structured data, a Google Business Profile that hasn't been touched in two years. But fixing them requires knowing which details are wrong in the first place.
What drives the AI's answers
AI assistants pull from a mix of sources: your Google Business Profile, structured data on your website, review platforms, and cached copies of pages that may be months or years old. When those sources disagree with each other — or are simply sparse — the AI either guesses or hedges with phrases like "hours may vary" or "I'm not certain about their current offerings."
Adding LocalBusiness schema to your website is one of the most direct signals you can send. It tells AI crawlers exactly what your hours, services, phone number, and address are, in a format they're designed to read and trust over looser sources.
Run the numbers for your business
Take five minutes and do this on paper. Estimate your monthly AI query volume — honestly, anywhere from 30 to 200 is a reasonable range for most local businesses. Multiply by 20%. Multiply by your average customer lifetime value. Multiply by 12. That number is your annual exposure from a single category of wrong or missing information.
Then run a free scan to see exactly what AI is saying about your business today — phone number, hours, services, location. Whatever's wrong is the place to start.