If you run a service business in 2026 and you have not asked ChatGPT to recommend a contractor in your trade and your city in the last 90 days, do it now. Then ask Perplexity. Then ask Google's AI Overview. The answers you get back are the answers your future customers are getting back. If you want a structured way to do exactly that, we wrote a ten-minute walkthrough for how to check if ChatGPT can actually see your business before you start on the fixes below. If your business is not in those answers, you are invisible to a third of the buyers who are quietly moving their initial search out of Google and into an LLM, and the share is growing every quarter.
Updated July 2026: before you do any of this, make sure the AI crawlers can even read your site. A single robots.txt line or a Cloudflare setting can block ChatGPT entirely. See is your website blocking ChatGPT, and how to check in 5 minutes.
Updated August 2026: if the AI already names you but describes the wrong company, that is a separate problem with a separate fix. See what to do when ChatGPT confuses your business with a competitor.
The fix is not exotic. It is not a hack. It is a small, specific set of structural changes to where your business shows up online and how it is described — most of which a competent service-business owner can put in motion in a week. Here is what AI search actually pulls from, why Bing Places is the lever almost nobody on your block is using, and the five-step plan to get your business cited the next time a customer asks an AI for the best roofer, cleaner, electrician, or tile installer in your service area.
Why does AI search matter for a service business in 2026?
Because the buying journey now starts in an LLM for a meaningful and growing share of customers, and the LLM picks the shortlist. Pew Research's 2025 data put ChatGPT's weekly active user base over 400 million globally, with a steep adoption curve among adults under 40 who are also the largest homeowner cohort in their first or second home — exactly the buyer for a roof replacement, a recurring cleaning contract, a kitchen tile job, or an electrical upgrade. Google's AI Overview now appears on close to 60% of search queries and is the only thing many users read before deciding which one or two businesses to call. The five-link Google ranking page that used to send buyers to your website is, for an increasing share of searches, a single AI-generated paragraph that names a few businesses and links to a few sources.
If your business is in the paragraph, the phone rings. If your business is not, the customer never sees you. The fight has moved from "rank in the top three on Google" to "be the business the AI cites when it generates the answer." Both still matter. The second is the one almost nobody is working on yet, which is exactly why the window to win is right now.
And once the AI does send a homeowner to your site, the site itself has to close them — most contractor sites still fail that step for a very predictable set of reasons. If your traffic is climbing without leads, that gap between visit and phone call is the problem to fix first; we broke it down in why your contractor website isn't getting leads.
What does AI search actually pull from?
LLM-powered search pulls from four sources at once, and the order matters. Knowing the order tells you where to spend your time.
- Live web search via the model's search partner. ChatGPT uses Bing as its primary web index. Perplexity uses a mix that leans on Bing and a few other crawlers. Google's AI Overview uses Google. Claude can search via its own connectors. The first thing the model does on a "best plumber in Lakewood" query is search the index — and the index ranks businesses that are claimed, structured, reviewed, and consistent.
- Structured business listings. Google Business Profile, Bing Places for Business, Apple Business Connect, and the major aggregators (Yelp, BBB, Angi, Houzz). The LLM trusts a business that exists in multiple structured listings with the same name, address, and phone number more than a business that exists in one.
- Schema markup on the business's own website. LocalBusiness, Service, FAQ, and Review schema on the contractor's site is read by the crawler and used to populate the AI's understanding of what the business does, where it operates, and what other people have said about it. A site with proper schema is a site the AI can quote with confidence.
- Third-party citations. Local press, directory listings, contractor association pages, and trade body memberships that mention the business by name. The AI pattern-matches: if three independent sources call you a "GAF Certified roofer in Toms River," the AI is more likely to surface you as a GAF Certified roofer in Toms River than a business that says it about itself in one place.
The model assembles its answer from all four at once. Win in one and you are in the running. Win in three and you start showing up consistently.
Step 1 — Claim Bing Places (the lever almost nobody is using)
If you do nothing else after reading this, claim Bing Places for Business this week. ChatGPT routes its web search through Bing, and the Bing local pack is the dataset it leans on for "near me" queries. A Bing Places listing takes about 20 minutes to claim, requires the same data you already have in your Google Business Profile, and is the single highest-leverage AI-search action a contractor can take in 2026.
The contractor share who have claimed Bing Places is somewhere in the low double digits — most of your competitors are not on it. That gap is the opportunity. The listing needs the same care a GBP listing needs: exact business name, exact NAP, accurate categories, real hours, real service area, and at least five photos. Bing imports a chunk of your GBP data on first claim, but the categories and the descriptions should be edited to read naturally in Bing too. Once it is live and verified, you are in the index ChatGPT reads when someone asks it to recommend a roofer in your town.
Step 2 — Make every business listing say the exact same thing
LLMs use NAP consistency — the exact match of name, address, and phone number across every listing — as a trust signal. The more places your business name, address, and phone match, the more confident the AI is that you are a real, single business. The more your listings disagree (one says "ABC Roofing LLC," another says "ABC Roofing," another says "ABC Roofing Co."), the more the AI hedges and the less likely it is to name you.
Standardize on one version of the business name, one address, and one phone number. Then audit, in this order:
- Google Business Profile
- Bing Places
- Apple Business Connect
- Yelp
- BBB
- Houzz (for remodelers/installers)
- Angi
- Facebook Business Page
- LinkedIn Company Page
- Your own website footer and contact page
Every one of those should match exactly. The free tools that scan citation inconsistency (Moz Local, Whitespark, BrightLocal) are worth a one-time pass. The paid services that "auto-fix" them are usually not worth the recurring cost — fix once, then keep the list current as you add locations.
Step 3 — Get the reviews that make AI describe you as "highly rated"
LLMs do not just count stars. They read the review text and use the descriptive language as signal. A business with 80 four-and-five-star reviews that contain phrases like "showed up on time," "explained the work," "fair price," "cleaned up after themselves" will get described to the AI's user with those phrases. A business with 10 reviews that all say "great" will get described as "rated well" and nothing else, and is less likely to make the shortlist on a query that asks for a recommendation with a reason.
Two practical moves: ask for reviews at the moment the customer is happiest (the day of completion, in person, with a QR code or a text link to your GBP review URL), and never coach the language. Authentic detail beats generic praise every time, and the AI can tell the difference between 50 reviews that say "good job" and 50 that say "Dave caught a flashing detail at the chimney that two other roofers missed and the price was the same as the lowest quote." The second business is the one the AI cites.
Step 4 — Add the schema markup the AI can actually read
A LocalBusiness schema block on every page of your website tells the AI's crawler the structured facts about your business in a single, parseable JSON object. Without it, the AI is guessing from prose. With it, the AI is reading exact facts. The schema does not show up to a human visitor — it is in the page's source — but it is one of the most impactful 30-minute changes a contractor site can make for AEO (Answer Engine Optimization, the new acronym for what we used to call SEO when the answers were ten blue links).
The schema types that matter most for a service business:
- LocalBusiness (or the specific subtype: Plumber, RoofingContractor, HousePainter, etc.) on the homepage and footer
- Service for each service page (roof replacement, gutter installation, etc.)
- FAQPage on any page with a Q&A section
- Review and AggregateRating if you publish reviews on your site
- areaServed as a list of City entities for every town you cover
If your site is on WordPress, Schema Pro or Yoast SEO Premium can install most of this. If your site is custom, a developer can add it in under a day. If you do not know whether your site has any schema markup, paste your homepage URL into Google's Rich Results Test and the output will tell you what is there and what is missing.
Step 5 — Write site copy in question-and-answer form
The fastest way to get the AI to quote your site is to write content in the shape the AI is generating. That shape is question + answer, with the answer in the first sentence. The H2 on the page is a question. The first sentence after the H2 is the direct answer. The rest of the section adds the why and the context. When the AI generates its response to a user's question, it pulls from sources that already answered the question cleanly — your page becomes one of those sources.
This is the same pattern Google's own People Also Ask is built on, and the same pattern the question-driven blog posts on our portfolio sites use. It feels mechanical at first, then it becomes how you write because it is also how people read.
How long does this take to show results?
Most contractors who execute the five steps see the first AI-search citations inside 60 to 90 days. The Bing Places claim is the fastest signal — Bing reindexes new claimed listings inside a few weeks. The citation cleanup compounds over the next quarter as the AI's training and live search both catch up to the consistent data. The schema and content changes show up as soon as the next crawl of your site, but the AI's confidence in citing you builds with every reinforcing source. The contractors who started this in early 2026 are already getting cited regularly by ChatGPT and Perplexity. The ones who start now are mid-pack. The ones who start in 2027 are behind.
How does Current Digital handle AEO for clients?
We build every client site with the schema, the LocalBusiness subtype, the areaServed city list, the FAQ schema, and the question-and-answer content structure built in from day one. The Bing Places claim is part of the launch checklist alongside the Google Business Profile. The citation cleanup is part of the first 90 days. The blog and service pages are written to be the source the AI cites when it answers a question about a roofer in Ocean County, a cleaner in Monmouth, a tile installer at the Jersey Shore, or a property manager in Rutland County. The whole site is built to be quotable, not just rankable.
If your business is invisible in ChatGPT today and you want to fix it, run your site through our free site grader for a baseline score on AEO, schema, and technical health, then read our local SEO checklist for contractors and our schema markup guide for the implementation detail. Or book a call and we will tell you, honestly, whether your current site is salvageable or whether AEO is the right reason to rebuild.