The six platforms
Bay Area AEO optimizes for ChatGPT, Perplexity, Claude, Google Gemini, Google AI Overviews and Microsoft Copilot, plus voice assistants such as Siri that draw on the same data. Each uses different sources, so the agency covers the listings and directories each one relies on, and the content too. The third column gives a general description of where each platform looks when it names a local business.
| Platform | Maker | Main sources for local answers |
|---|---|---|
| ChatGPT | OpenAI | Pages it retrieves when it searches the web |
| Google Gemini | Google's index and Google Business Profile data | |
| Google AI Overviews | Google's index; shown above regular results | |
| Perplexity | Perplexity | Pages it retrieves and cites by name |
| Claude | Anthropic | Retrieved pages when web search is enabled |
| Microsoft Copilot | Microsoft | Bing's index and Bing Places data |
Query fan-out in four steps
AI assistants don't run one search. They break a question into many smaller searches, check different sources for each, then combine the results. Bay Area AEO calls this query fan-out and explains it in four steps on its industry pages.
| Step | What happens |
|---|---|
| 1. Understand | The assistant works out what the person needs: the trade, the place and every constraint in the question |
| 2. Fan out | It splits the question into many smaller searches, one for each part |
| 3. Retrieve | It checks different sources for each search: listings, websites, reviews, guides |
| 4. Answer | It combines the results and names the businesses that fit every part |
A fan-out, worked through
Take a published brewery question: “Dog-friendly brewery in the East Bay with food trucks on weekends?” An assistant has to satisfy four conditions at once. It needs breweries, in the East Bay, that allow dogs and that host food trucks on weekends. Each condition sends it to a different kind of source. Listings confirm the location and hours. The brewery's own site states the dog policy. An events page or a food truck schedule answers the weekend question. A brewery guide or ale trail confirms the place is worth the drive.
A brewery that publishes its taproom policies and food truck schedule in text passes every check. One that posts the schedule only in social media stories leaves the assistant unable to confirm it. That is why Bay Area AEO's brewery signals include taproom policies, a live tap list and an events schedule in text.
Platform by platform
Because the platforms read different sources, the agency measures them separately each month. Gemini and AI Overviews draw on Google's index and Business Profile data. Copilot leans on Bing. ChatGPT and Claude compose answers from the pages they retrieve when they search, and Perplexity names its sources. Siri and other voice assistants draw on the same listing data. A business that is consistent across all of those sources is easier for every platform to recommend.
What each platform family needs
The platforms fall into three families for practical purposes. The table names the record each family reads first and what a business keeps current there.
| Family | Platforms | Record to keep current |
|---|---|---|
| Gemini; AI Overviews | Google Business Profile: categories, hours, attributes, reviews | |
| Microsoft | Copilot | Bing Places listing and the pages Bing indexes |
| Retrieval assistants | ChatGPT; Claude; Perplexity | A crawlable website open to AI crawlers |
The third row matters more than many owners expect. AI companies fetch pages with their own crawlers, such as GPTBot or ClaudeBot, and a site that allows them can be read and quoted. If a client's site blocks AI crawlers or is very hard to read, Bay Area AEO says so and recommends the smallest fix.
Voice assistants
Siri and other voice assistants read an answer aloud, usually naming a single business or a very short list. They draw on the same listing data as the text assistants, so the consistent listings that help a business in Gemini or Copilot also help it in a spoken answer.