A growing share of your customers no longer type "plumber London Ontario" into Google and scan ten links. They ask ChatGPT, Microsoft Copilot, or Google's AI results a full question, "who's a reliable plumber in London that does emergency calls?"and get three names back.
Either you're one of those names or you don't exist for that customer. AI assistants don't pull recommendations from thin air, though. They pull from sources you can influence, and most of your competitors haven't started.
Where AI answers actually come from
When an assistant recommends local businesses, it's drawing on some mix of:
- Live web search. ChatGPT and Copilot run real searches (Bing powers much of this) and read the results before answering. Google's AI results are built directly on its index and Business Profile data.
- Business listings and reviews. Google Business Profile, Bing Places, Apple Maps, and review platforms are structured, trusted data, exactly what AI systems prefer to cite.
- Websites they can read and extract from. Fast, crawlable pages with clear, self-contained answers get quoted; slow, vague ones get skipped.
- Third-party mentions. Directories, local news, "best of" lists, and community discussions (Reddit and local Facebook groups included) act as corroboration that you're real and reputable.
There's no form to submit and no "AI optimization" service to buy. Being recommendable is what falls out of doing ordinary local visibility work unusually well.
The playbook
1. Make your listings bulletproof, on all three platforms
Google matters for Google AI; Bing matters for ChatGPT and Copilot, which is why we tell every client to stop ignoring it. Claim and complete Google Business Profile, Bing Places, and Apple Business Connect. Identical name, address, phone, and hours everywhere, assistants cross-check, and inconsistency reads as unreliability.
2. Earn reviews with substance
AI answers routinely justify recommendations with review language: "highly rated for fast emergency service." Volume matters, but descriptive reviews matter more, they give the model something concrete to repeat. Ask customers to mention the specific job. Then reply, because response patterns signal an attentive business.
3. Publish pages that answer questions in full sentences
Assistants extract; they don't interpret marketing. A page that says "Emergency plumbing in London, Ontario, average response under 90 minutes, available weekends" is quotable. "Your comfort is our passion" is not. Structure service pages with real questions as headings and direct, self-contained answers under each, the same FAQ discipline that wins Google's featured results wins AI citations.
4. Keep the technical floor high
If a crawler can't read your site quickly, you're not in the answer. That means fast load times, working mobile layout, clean headings, and structured data (schema markup) describing your business, services, and FAQs. Decide deliberately to allow AI crawlers in robots.txt, blocking them removes you from the sources those assistants read. This is standard technical SEO; AI search just raised the price of skipping it.
5. Get mentioned where AI reads
A profile on relevant directories, a story in the local paper, a listing in "best pizza in London" roundups, a genuine presence in community discussions, each is an independent witness saying you're legitimate. You can't fake this layer, but you can earn it: sponsor something local, pitch the story, ask a happy customer who's active in the community group.
6. Measure it
AI referrals show up in analytics as traffic from chatgpt.com, copilot.microsoft.com, perplexity.ai, and Google's AI surfaces. Watch that segment monthly, it tells you whether this work is compounding. A privacy-friendly tracker like NX DAT shows these referrers without a GA4 archaeology dig. And periodically just ask the assistants your customers' question, "best [your service] in [your town]"and see who they name.
How long does this take?
Listings and website fixes influence AI answers within weeks, because assistants search live. The mention layer, reviews, press, community reputation, compounds over months. Start with the mechanical work (listings, site, schema), then feed the reputation layer continuously. If you'd rather see your starting point first, a free SEO audit covers the technical half in ten minutes, and our digital marketing team handles the rest as a program.
Frequently asked questions
Can I pay to be recommended by ChatGPT or Google AI?
No. There's currently no paid placement inside organic AI recommendations. Ads may appear around AI results, but the recommendations themselves are drawn from search results, listings, reviews, and mentions, which is why the organic work above is the whole game.
Does blocking AI crawlers in robots.txt protect my content?
It keeps your pages out of those systems, including out of their recommendations. For a local business trying to be found, blocking AI crawlers is usually self-defeating. Make the choice deliberately rather than by default.
Why does Bing matter if nobody uses it?
People don't browse Bing, but ChatGPT and Microsoft Copilot do. Bing's index and Bing Places data feed the answers those assistants give. Ten minutes claiming your Bing listing buys presence in every tool built on it.
How is this different from regular SEO?
It's the same foundation, fast site, complete listings, reviews, mentions, with heavier weight on extractable, plainly-written answers and cross-platform consistency. If your SEO is genuinely good, you're most of the way there; AI search mostly punishes the businesses that were coasting on thin profiles and slow sites.
How do I know if AI assistants are already sending me customers?
Check your analytics for referrers like chatgpt.com and perplexity.ai, and start asking new customers "how did you find us?""I asked ChatGPT" answers are already common in 2026. If you're tracking with NX DAT or GA4, segment those sources and watch the trend.