How AI assistants choose which brands to recommend

When someone asks an AI assistant “What’s the best invoicing tool for freelancers?”, the answer usually contains a short list of named products. Being on that list — and near the top — is quickly becoming as important as ranking on the first page of Google.

This post explains, in practical terms, what drives those recommendations and which levers you control.

Two sources of knowledge

Modern assistants combine two things when they answer:

1. What the model learned during training. Brands that were widely and consistently discussed on the web before the model’s training cutoff are “known” to it. This changes slowly — only when a new model version ships.
2. What it finds with live web search. ChatGPT, Perplexity, Claude and Gemini can all search the web while answering. The pages they retrieve — and cite — strongly shape the final list.

The second source is the one you can move this month.

What the retrieved pages tend to be

Look at the citations behind answers in your category and you will typically find a mix of:

• Listicles and “best X for Y” roundups on publisher and affiliate sites.
• Comparison and “alternatives to” pages, including vendors’ own.
• Review platforms and directories relevant to your industry.
• Community discussions (forums, Q&A sites).
• The vendors’ own product, pricing and documentation pages.

If your competitors appear on the pages that get cited and you don’t, the assistant has little reason to name you.

Signals that make a page easy to use

Assistants summarise. Pages that state clearly who the product is for, what it does, what it costs and how it compares are easier to quote than vague marketing copy. Useful patterns:

• A one-sentence description of what you sell and to whom, near the top of your homepage.
• Dedicated pages for the use cases buyers actually ask about (“for freelancers”, “for agencies”).
• Honest comparison pages (“X vs Y”) and “alternatives to Y” pages.
• FAQ sections with direct answers, marked up with FAQPage schema.
• Up-to-date pricing on a crawlable page.

Why you have to measure it

AI answers vary by engine, by wording and over time. A single manual check tells you very little. To know where you stand you need to:

• Ask many realistic buyer questions, not just your brand name.
• Ask several engines, since each searches and ranks differently.
• Repeat on a schedule and look at trends.
• Record which sources were cited, so you know where to act.

That is exactly the loop Mentioned automates. You can see your starting point in under a minute with the free audit.