How to get AI to recommend your product

People now ask ChatGPT and Perplexity "what's the best tool for X" instead of Googling. If AI engines don't mention your brand, you're invisible in that channel. Here's the honest playbook — what works, what doesn't, and how to measure it.

Why AI recommendations matter

When someone asks an AI assistant for a product recommendation, the answer comes from what the model learned in training plus what it can verify on the live web. Unlike Google's ten blue links, there's no page two — you're either in the answer or you're not. Brands that AI engines cite get the click; brands they can't verify don't exist in that conversation.

The playbook (4 steps)

1. Make your site AI-readable

AI crawlers need the same clarity humans do, in machine form: semantic HTML, a descriptive title and meta description, JSON-LD structured data, a sitemap, and an llms.txt file at your root summarizing what your product does. If your robots.txt blocks AI crawlers (GPTBot, ClaudeBot, PerplexityBot and friends), nothing downstream works — check this first. Run the free AI-readiness audit to see where you stand in two minutes, no login.

2. Earn citations from sources AI trusts

Engines don't just read your site — they read what others say about you. Reviews, comparison articles, directory listings, and open datasets are what get cited in AI answers. One honest benchmark with public methodology beats ten press releases: it's the kind of source Perplexity actually links to. (We publish ours weekly at the Quotalis benchmark.)

3. Keep your brand facts consistent

AI engines cross-check. If your pricing, feature list, or company description differs between your homepage, your llms.txt, and third-party directories, models hedge or skip you. One canonical source of truth, repeated everywhere.

4. Measure what AI actually says

This is the step everyone skips: ask the engines about your brand the way customers do, repeatedly, and record what comes back. AI answers vary run to run — a single check proves nothing; a weekly time series shows whether you're gaining ground. Quotalis does exactly this across ChatGPT, Claude, Gemini and Perplexity: mentions, citations, and share of voice against your competitors.

What doesn't work

• Keyword stuffing for AI. There's no keyword box to stuff — models summarize meaning, not match strings.
• One viral post. Training data and citation graphs reward sustained, consistent presence, not spikes.
• Asking once. A single "does ChatGPT know me" check is a anecdote. Track it weekly or you're flying blind.

FAQ

How do I get ChatGPT to recommend my product?

Make your product verifiable: clear pages with structured data, an llms.txt file, citations from trusted sources, and consistent brand facts everywhere. Then measure with a tracker like Quotalis to see whether ChatGPT actually started mentioning you.

What is llms.txt and does it help?

It's a markdown summary of your site for AI crawlers — robots.txt for language models. It helps engines describe you accurately. Check yours with the free audit.

How do I know if AI search engines talk about my brand?

Ask them directly, repeatedly — or use Quotalis, which tracks mentions and citations across all four major engines and benchmarks you against competitors.

Related: AI visibility tools compared · weekly benchmark · our methodology.