TL;DR
- Each AI assistant reads a different local data pipe: ChatGPT leans on Foursquare and Bing Places, Google's AI tools use Business Profile and Maps data, Siri uses Apple Maps.
- Foursquare's Places API is the dominant source behind ChatGPT's local answers — independent analysis puts its share above 70% of ChatGPT's local business results.
- Ranking well in Google's map pack does not carry over to AI recommendations. Overlap between the two is well under half.
- Consumer use of AI for local recommendations grew fast in 2026, so this isn't a niche channel anymore — it's becoming a normal path to discovery.
- LocalBusiness schema is still worth deploying for consistency and Google's AI features, but its direct effect on ChatGPT and Perplexity citations is unconfirmed.
Local business AI search refers to the process by which assistants like ChatGPT, Perplexity, Gemini, and Siri surface specific businesses — plumbers, dentists, coffee shops — in response to conversational, location-based queries. It behaves differently from Google local search because the underlying data sources, ranking logic, and citation mechanics are not the same.
The plumber question: who actually gets recommended
Ask ChatGPT "who's a good plumber near me" and it won't query Google Maps. It has no direct pipe into Google Business Profile. Instead, it typically draws from a combination of Bing Places for Business listings and, as of a licensing deal that took effect in July 2026, Yelp review data. Layered on top of that is Foursquare, which became an official OpenAI data partner and now supplies the bulk of ChatGPT's point-of-interest results — its Places API covers more than 100 million locations across 200+ countries, and outside analysis has found it accounts for over 70% of what ChatGPT surfaces for local queries.
This matters because a business that has meticulously optimized its Google Business Profile — photos, categories, posts, review responses — may be functionally invisible to ChatGPT if its Foursquare, Bing, or Yelp presence is thin or inconsistent. The optimization target is not the same platform you've spent a decade tuning for.
Perplexity behaves more like a web-search hybrid: it crawls and cites live pages, directory listings, and review aggregators, so a strong web presence (consistent citations across Yelp, TripAdvisor, local news, association directories) carries more weight than in ChatGPT's more closed local-data pipeline. Gemini, being Google's own model, has the tightest integration with Business Profile and Maps signals, which puts it closest to traditional local SEO behavior. Siri, unsurprisingly, defaults to Apple Maps data — meaning Apple Business Connect listings matter there in a way none of the others touch.
Why your Google map pack ranking doesn't transfer
A business owner who ranks #1 in the local map pack for "emergency plumber [city]" might assume that authority carries over. It largely doesn't. Data on this shows fewer than half of Google map pack winners (45%) also appear in AI recommendation results, and citation overlap with ChatGPT's sources is even thinner — only about 12% of what ChatGPT cites matches Google's first page of results.
The practical read: local GEO is a separate discipline from local SEO, not an extension of it. You can win the map pack and still be absent from ChatGPT's answer, and vice versa. Each system has independent ranking logic, independent data ingestion, and — in ChatGPT and Perplexity's case — no transparency into how weighting works.
What profile fields actually seem to matter
Since none of the major AI vendors publish their local ranking criteria, the safest approach is to treat every data source as if it will eventually be checked and to keep them identical:
- NAP consistency — name, address, phone number must match exactly across Google Business Profile, Bing Places, Yelp, Apple Business Connect, and Foursquare's own claimed listing. Discrepancies (a suite number here, a missing one there) reduce confidence signals for AI systems the same way they do for traditional local SEO.
- Category and attribute completeness — Foursquare and Yelp both use structured category taxonomies. A plumber only listed as "Home Services" instead of "Plumber" with sub-attributes (emergency service, licensed, insured) may be filtered out of narrower queries.
- Review volume and recency across all platforms, not just Google — since ChatGPT now licenses Yelp data, a business with strong Google reviews but a stale or thin Yelp profile is under-optimized for that specific pipeline.
- Hours and geo-coordinates accuracy — assistants answering "is this open now" or "near me" queries depend on structured, current hours data. Errors here cause a business to be silently dropped from time- or distance-sensitive answers rather than actively penalized.
LocalBusiness schema: useful, but don't oversell it
LocalBusiness schema — Schema.org markup specifying address, geo coordinates, opening hours, and business type — is worth implementing on your website regardless. It's cheap, it's a known input for Google's AI features and rich results, and it creates one more consistent, machine-readable record of your NAP data that reduces ambiguity for any crawler that does read it.
What it doesn't have is confirmed value for ChatGPT or Perplexity specifically. Neither company has disclosed whether their crawlers preserve or weight schema during extraction. The technical capability to parse it exists — these are HTML-crawling systems, after all — but there's no published confirmation that Foursquare-sourced or Bing-sourced local answers in ChatGPT are influenced by your on-site schema at all. Treat it as good infrastructure hygiene, not a local-GEO growth lever on its own.
Why this now matters for the emergency plumber
The shift from Google-first to AI-first local discovery isn't hypothetical anymore. BrightLocal's research found 45% of U.S. consumers used an AI tool for a local recommendation in 2026, up from just 6% the year before. That's a fast enough curve that a business ignoring its Foursquare, Yelp, and Bing Places presence today is making a bet that this channel stays marginal — a bet the data doesn't support.
FAQ
Does optimizing my Google Business Profile help with ChatGPT?
Indirectly at best. ChatGPT has no direct GBP integration and instead pulls from Bing Places, Foursquare, and licensed Yelp data, so a strong GBP alone won't guarantee a ChatGPT citation.
Which single data source matters most for ChatGPT local results?
Foursquare, currently the dominant source behind ChatGPT's local business answers as an official OpenAI data partner covering over 100 million points of interest globally.
Is LocalBusiness schema worth implementing for a small business?
Yes for consistency and Google's AI features, but treat it as hygiene rather than a proven ChatGPT or Perplexity ranking factor, since neither platform has confirmed how or whether it uses schema during crawling.
