Branded vs Unbranded AI Queries

In AI search, as in traditional search, queries fall into two buckets:

Branded queries name a brand. "Is [brand] good for marathon running?" or "[brand A] vs [brand B]" are branded. The shopper already has the brand in mind and wants validation, comparison or a specific detail.

Unbranded queries describe a need, use case or category without a brand. "Best running shoes for flat feet under 150 dollars" is unbranded. The shopper has intent and no preference, and the assistant decides which brands to surface.

The gap between them is large. On branded queries the bar is low: the assistant needs to know the brand and say something accurate about it. On unbranded queries the assistant has to pick your products out of the whole category, and the brands that win there capture every shopper who does not already have a preference.

In most categories, unbranded questions far outnumber branded ones. That is why unbranded AI visibility is usually the highest-value surface a retailer has, and the one most retailers have not looked at.

Branded queries reward entity clarity. If ChatGPT knows who you are and what you do, it can answer branded queries about you. The signals that win are consistent entity optimization, complete Organization schema, and third-party mentions that reinforce the basic "who you are" story.

Unbranded queries reward category authority. To be selected as the answer to "best [category] for [use case]", you have to be among the 2-3 brands the AI considers the top of that category. That means:

• Being mentioned as a top choice in multiple authoritative third-party roundups and review guides
• Having deep content on the use case itself as well as your product
• Ranking well in traditional search on the category, since assistants still draw on underlying search signals
• Having complete product schema so product cards can surface with rich detail
• Having reviews with specific praise for the use case
• Having product information that actually answers the use-case question: the attributes, descriptions and Q&A an assistant compares on

Running both question sets for the same brand usually shows one pattern: strong on its own name, thin on the category. See our write-up on branded vs unbranded AI visibility. Closing that gap is usually the largest single AI visibility opportunity for an established brand.

Because the work is so different, measurement and optimization should split too:

Branded query panel. 10-20 queries that name your brand. Include direct "what is X?" questions, "is X good for Y?" questions, and comparisons with known competitors. Measure citation accuracy, sentiment, and share of voice against named competitors. Fix gaps via entity optimization: schema, third-party profiles, consistent naming.

Unbranded question set. 30 to 60 questions that describe your category's buying scenarios without your name. "Best [category] for [use case]", "[category] alternatives to [competitor]", "[category] under [price]". Measure whether your products appear at all, in what position and as what. Fix gaps through category content, third-party roundups and, most directly, product information that answers the use-case question.

Run both sets per assistant on a schedule and treat the unbranded gap as the growth surface. This is how the Paz platform tracks visibility, with the question sets written for the shoppers you sell to, and a Paz expert turns each unbranded gap into a product information fix for you to approve. See the AI visibility service.