How to Get Cited by Claude and Gemini

Most advice about AI visibility treats every platform the same. Optimize your website, add schema, get listed in directories. That's not wrong — but it glosses over something important: Claude and Gemini pull from meaningfully different sources than Perplexity or ChatGPT, and understanding that difference is what separates a decent AI presence from a strong one.

If you've already done the foundational work — clean schema, good service pages, consistent NAP — this is the layer on top.

How Claude sources its answers

Claude is built by Anthropic. Until recently, it operated almost entirely from training data: a snapshot of the web captured before a cutoff date, weighted toward authoritative, high-quality sources.

That's changing. Claude now has web search access for many queries on Claude.ai, which means Anthropic's crawler (ClaudeBot, user agent `Claude-User`) is actively indexing pages to retrieve current information. But unlike Perplexity, which does a live search for nearly every query, Claude's web search is selective — it tends to search when the query is clearly time-sensitive or when its training data confidence is low.

The practical implication: getting Claude to cite your business requires showing up in *both* layers. Your site needs to be crawlable by ClaudeBot for real-time queries. And you need an established presence on sources that carried enough authority to appear in Claude's training data.

What "training data authority" looks like for local businesses

Most local businesses aren't cited by Claude from their own websites alone. Claude's training data skewed heavily toward editorial content: news articles, industry publications, authoritative directories, and Wikipedia.

If a regional magazine ran a "best of" list featuring your restaurant, that article is more likely in Claude's training data than your restaurant's homepage. If your business is mentioned in an industry trade publication, a chamber of commerce feature, or a local news story, those citations carry more weight than anything on your own site.

This is why PR and earned media contribute to AI visibility in a way that's underappreciated. It's not just about backlinks — it's about being mentioned, in plain text, on sources that AI models were trained to trust.

Practical steps to improve your Claude presence

First, check whether ClaudeBot is blocked on your site. Review your `robots.txt` file and look for any block on `ClaudeBot` or `Anthropic-AI`. Some SEO plugins add blanket bot blocks that catch AI crawlers unintentionally. If Claude can't crawl you, it can't retrieve you for real-time queries.

Second, audit your directory presence on sources that carry editorial weight. Yelp, TripAdvisor, Houzz, Avvo, Healthgrades, G2 — the specific directories matter by industry. These were crawled extensively before Claude's training cutoff and are almost certainly in the corpus. Being well-represented there, with accurate information and specific descriptions, feeds directly into Claude's understanding of your business.

Third, pursue one or two editorial mentions per quarter. A profile in a local business journal, a contributed article in an industry publication, a quote in a regional news story. These are long-game plays, but they're the clearest path to being present in AI training data for future model versions.

How Gemini sources its answers

Gemini is Google's AI. That single fact shapes everything about how it retrieves local business information.

Gemini draws heavily from Google's own data: the Knowledge Graph, Google Business Profile, Google Maps, Google's index of structured data. When someone asks Gemini "best HVAC company in Hamilton," it's not doing a neutral web search — it's leaning into data Google already has high confidence in.

This makes Gemini fundamentally different from Claude. For Gemini, your Google presence is your AI presence.

What a strong Gemini presence requires

Your Google Business Profile is the single most important asset. Gemini reads it directly. This means your GBP needs to be fully built out — not just claimed, but optimized.

Service categories matter. Use the most specific primary category available for your business type. Add all relevant secondary categories. The GBP category structure maps closely to how Gemini understands what kind of business you are and which queries you're relevant to.

Your service list inside GBP matters more than most businesses realize. Each service you add, with a description, is another explicit signal about what you offer. Gemini uses this to match queries like "who does roof inspections in Barrie" — and if that service isn't listed in your GBP, you're a weaker match than a competitor who listed it.

GBP Q&A is underused and valuable. You can add questions and answers directly to your own profile. Think about the questions Gemini might try to answer from your profile — "do they offer emergency service?", "what areas do they serve?", "are they licensed and insured?" — and answer them explicitly in the Q&A section.

Schema markup through Google's lens

Gemini also reads structured data, but through Google's indexing rather than direct crawl in the way Perplexity works. This means the same schema best practices apply — correct business type, full NAP, `areaServed` with city-level specificity — but what Gemini weighs most heavily is *consistency between your schema and your GBP*.

If your schema says you serve Burlington and Oakville, but your GBP only lists Burlington, Gemini has conflicting data from two sources it trusts. The confident citation goes to businesses where every data source agrees.

Google reviews feed Gemini's understanding of your business in the same way they feed traditional local search. Specific review text that names services and locations is signal that carries through to Gemini's recommendations.

The combined strategy

For Claude: focus on editorial presence and directory authority. Make sure ClaudeBot can crawl you. Pursue earned media on industry-relevant and local publications.

For Gemini: treat your Google ecosystem as infrastructure. Fully built GBP, complete service list, consistent schema, active review acquisition. Every Google optimization you do for local SEO is also an AI visibility optimization for Gemini.

The good news is that these strategies compound. Strong directory listings help both. Consistent NAP across all sources helps both. The foundational work isn't wasted — you're just adding platform-specific layers on top.

If you haven't mapped where your current AI visibility stands across Claude, Gemini, ChatGPT, and Perplexity, a Signal Check gives you a clear starting point. It takes about 30 seconds and shows you which platforms are citing you — and which ones have never heard of you.