The 4 query types that actually matter for AEO

Most teams tracking their AEO visibility are monitoring the wrong queries. They check whether ChatGPT knows their brand name, whether Gemini mentions them when asked directly, whether Perplexity includes them in a list. These are branded queries. You're measuring your ego, not your funnel.

The queries that shape purchasing decisions — the ones where LLMs actually influence which product someone chooses — fall into four categories. If you're not tracking all four, your visibility score is misleading.

1. Category queries

"Best [product category] for [use case]"

This is the highest-intent query type. Someone asking "best project management tool for remote engineering teams" is actively evaluating options. If your brand doesn't appear in the LLM's response, you've been eliminated from consideration before the prospect even visits your site.

Category queries are where AEO matters most. You can have perfect brand sentiment, great reviews, a strong website — and still lose if the LLM doesn't include you in the consideration set.

Track 2-3 category queries that match your primary use cases. Make sure the phrasing matches how real users ask, not how your marketing team would frame it.

2. Comparison queries

"[Your brand] vs [competitor]"

Comparison queries are where LLMs write your competitive narrative. If you don't have comparison content on your own site, the LLM assembles one from whatever sources it finds — G2 reviews, Reddit threads, blog posts from affiliates with their own agenda.

The fix is straightforward: create honest, detailed comparison pages for every major competitor. Not "why we're better at everything" (which nobody believes) but genuine head-to-head analysis that acknowledges trade-offs. LLMs favor balanced, authoritative content. Giving them a well-structured comparison from your own domain increases the chance they reference your framing.

Track comparison queries for your top 3-5 competitors. Pay attention to which direction the LLM favors and what reasons it cites.

3. Problem queries

"How to [solve the problem your product addresses]"

Problem queries catch users before they've decided to buy anything. They're searching for solutions, not products. If the LLM mentions your product as part of the solution, you've entered the consideration set at the earliest possible moment.

Problem queries are the hardest to win because the LLM isn't looking for product recommendations — it's looking for informational content. The brands that show up here tend to have deep, practical content about the problem space. Not "5 ways our product solves X" but genuinely useful content that happens to mention the product as one approach.

Track 2-3 problem queries that match your product's core value proposition. These are the queries where content marketing has the highest leverage.

4. Reputation queries

"Is [your brand] good?" or "[your brand] reviews"

Reputation queries are your early warning system. When someone asks an LLM about your brand's quality, the response is a weighted summary of everything the model has seen about you. If the sentiment shifts negative, it's usually a lagging indicator of a real problem — bad reviews, PR issues, or competitive content that's reshaping the narrative.

Track 1-2 reputation queries monthly. Don't obsess over them, but don't ignore them. A gradual sentiment shift from "neutral" to "negative" is worth investigating. A sudden shift is worth investigating immediately.

Why branded queries are misleading

Branded queries ("What is [your brand]?" or "Tell me about [your brand]") are the easiest to score well on and the least commercially useful. Of course the LLM knows your brand exists — it's been trained on the internet. The question isn't whether it knows you, but whether it recommends you.

Teams that only track branded queries get inflated visibility scores that mask real gaps. Your score looks great, but you're invisible in the category and comparison queries where decisions actually happen.

Setting up your query portfolio

For a useful AEO audit, we recommend this mix:

• 2-3 category queries (highest priority)
• 2-3 comparison queries (one per major competitor)
• 1-2 problem queries
• 1 reputation query

Run these through onpage.app and you'll get a visibility picture that actually maps to your funnel, not just your ego.