Whitespark spent this summer measuring when Google shows an AI Overview on local queries, and the numbers should reorder your week. AI Overviews now appear on 68% of local business searches. The classic map pack shows on just 39%. Slice the data by intent and it gets sharper: queries about price, cost, or buying trigger an AI answer more than 80% of the time. A hybrid query like “average cost of dental implants in Phoenix” hits 97%. Search Engine Journal published the full breakdown in late July.
Now put that next to a habit most local service businesses share: refusing to publish pricing. Those two facts collide head on. Your highest-intent searchers, the people literally asking what your work costs, now get their answer from an AI Overview. And because your site says nothing about price, the AI assembles its answer from whoever does. That is usually a national lead broker like Angi, a franchise competitor with a real pricing page, or a directory quoting stale numbers about your market. You appear nowhere in the answer that decides whether the phone rings.
The Query You Refuse to Answer Is the One AI Owns
Owners hide pricing for reasons that feel sound. Every job is different. Competitors will undercut you. A published number becomes a negotiation ceiling. Fair concerns, all three. But the search behavior does not care. When someone types “how much does a water heater replacement cost,” Google no longer sends them to your contact form to find out. It writes them an answer on the spot, with citations, and moves on. If your site gives it nothing to quote, you are not in the conversation. Silence is not neutral anymore. It is a forfeit.
The citation math makes the forfeit expensive. Omniscient Digital analyzed more than 23,000 AI citations and found that owned content accounts for only about 23% of citations even on branded queries. The other 77% comes from third parties: directories, review sites, Reddit threads, news mentions. You already control a minority of what AI says about your business. Cost is the one topic where you hold the primary source, because nobody knows your prices better than you do. Giving that away too means the machine describes your market, your trade, and your price range using everyone’s data but yours.
There is a window here, and it is open because your competitors share your fear. In most local markets, almost nobody publishes usable pricing. The first plumber, med spa, or landscaper in a metro area to put honest ranges on a well-built page becomes the default citation for every cost query in that category. That position compounds. AI systems keep returning to sources that answered well before.
Build One Cost Page Per Service. Here Is the Template
Pick your highest-ticket service and build a single page that answers the money question directly. Not a blog post buried three clicks deep. A standing page, linked from your site’s main menu, titled the way people actually search: “What Does Water Heater Replacement Cost in Boise?” Then answer it immediately. Kevin Indig’s analysis of 1.2 million ChatGPT responses found that 44.2% of citations come from the first 30% of a page. An answer that lives below four paragraphs of throat-clearing may as well not exist.
The page needs five parts, in this order:
- The range, first sentence. “Most water heater replacements in the Boise area run $1,400 to $3,200 installed.” One sentence, real numbers, no hedging preamble.
- A variables table. Tank versus tankless, capacity, venting changes, code upgrades, access difficulty. Each row names the variable and how it moves the price. Tables give AI discrete facts it can quote with confidence.
- Local factors. Permit costs in your county, climate quirks, regional code requirements. This is what separates your page from the national averages Angi publishes.
- A short FAQ in plain speech. “How much to fix a leaky faucet?” reads differently than “faucet repair cost analysis.” Write the questions the way customers say them on the phone.
- A visible date stamp. “Prices reviewed August 2026.” Then keep the promise quarterly.
Here is a copy-ready opening block. Swap in your trade and your numbers:
Most [service] jobs in [city] run $[low] to $[high]. The biggest price drivers are [variable 1], [variable 2], and [variable 3]. Below is what moves a job toward each end of that range, what [county] permits add, and the questions we answer most often on estimates. Prices reviewed [month, year]. Every written estimate is firm before work begins.
That last sentence does real work. It publishes a range while making plain that the binding number arrives with the written estimate. You have answered the search without pricing the job sight unseen.
The Tradeoffs, Straight
Will competitors see your prices? Yes, and they already know them. Anyone in your trade can price your work within 10% from a single job photo. The pricing page reveals nothing to professionals. It reveals everything to the customer choosing between you and a company that made them call to ask. You are not leaking secrets. You are converting an open secret into a citation asset.
Will customers hold you to the number? Some will try, which is why you publish ranges plus the variables that move them, never flat rates. A range with named drivers educates the caller before the first conversation. Operators who run this play report better-qualified calls, because the price shoppers who would never pay your rates screen themselves out before dialing.
One thing that will not save you: schema markup on its own. An Ahrefs test this year found that adding structured data alone did not move AI citations. Schema helps machines parse a page that already answers well. It cannot make an empty page citable. Spend the afternoon writing real numbers, not decorating thin copy with JSON-LD. And keep the page alive. Ahrefs’ study of 17 million AI citations found cited content averages 25.7% fresher than what merely ranks. A cost page last touched in 2024 loses to a competitor’s page dated last quarter.
Consistency matters as much as freshness. An August 10 roundup from Mean CEO’s startup blog framed local search in 2026 as a verification contest: engines cross-check your claims across your Google Business Profile, your site, and directories, and contradictions kill confidence. If your cost page says $1,400 to $3,200 and a directory quotes your old $999 special from 2023, fix the directory the same day the page goes live.
What It Costs and How to Prove It Worked
The do-it-yourself version costs an afternoon per service. You know your price history; the writing is transcription. If you want monitoring, AI-citation trackers like Otterly and Profound run roughly $50 to $350 per month for a single-location business. If you hand the whole thing to an agency, a Chicago-area local SEO shop, The Search Sherpa, published an unusually candid pricing guide last week: this work runs $1,000 to $3,000 per month folded into a normal local SEO retainer. The same guide flags the scam pattern to avoid: any vendor selling guaranteed AI placement as a $199 flat add-on is selling something nobody controls.
Measurement is simpler than the tool vendors make it sound. Write down ten money queries for your business, the exact phrases with “cost” and “price” and your city in them. Once a month, run each one through Google, ChatGPT, and Perplexity, and record whether your business is named or cited. That cited-or-not log is your scoreboard. Support it with two numbers you already have: branded searches in Search Console, which rise when AI answers put your name in front of people, and calls plus direction requests on your Business Profile.
Give it 90 days before judging. Citation patterns move slower than rankings, but they stick harder once they move. The operators who publish real numbers this quarter are buying the default position for every cost query in their market, and defaults are hard to unseat. We built AU to flag exactly these windows while they are still open. Subscribe and you will catch the next one on time.
