Branded vs. Unbranded Queries in AI Search and Tracking
Distinguishing between branded and unbranded AI prompts is essential for accurate performance measurement and strategy.
1Understanding AI Search Queries
AI search queries are categorized into branded and unbranded types. A branded query explicitly names your company or product, indicating existing brand awareness. An unbranded query describes a category, problem, or use case without naming a brand. Distinguishing these is vital because they represent different stages of the customer journey.
2The Blended Score Trap
Many teams mistakenly combine branded and unbranded data into a single visibility score. This creates a misleading metric that masks poor discovery performance. Because branded prompts almost always mention the brand, they inflate the total, hiding the fact that new, unbranded users may not be finding your company at all.
3Measuring Different Intentions
Branded prompts reveal how AI describes your brand and confirm user intent. Unbranded prompts measure your ability to reach new buyers who are still in the research phase. Treating these as separate metrics allows you to identify whether your challenge is poor brand messaging or a lack of overall market visibility.
4Implementing Weekly Tracking
AI responses are probabilistic, meaning the same prompt can yield different results across multiple attempts. Experts recommend running prompt tracking daily but reviewing the data on a weekly basis. This routine helps smooth out inconsistencies and allows you to identify meaningful trends rather than reacting to single, isolated AI search answers.
5Building a Prompt Library
Effective tracking begins with a clean, deduplicated library of prompts. Each prompt should be labeled by buyer stage, geography, and whether it mentions a competitor. By assigning an Ideal Customer Profile to every entry, you ensure that your data remains actionable and relevant to your specific target market and business goals.
6Calculating Visibility Rates
To calculate your visibility rate, divide the number of answers that mention your brand by the total number of tracked answers. Always use a clear denominator, such as 12 out of 60 prompts, rather than raw counts. This provides the necessary context to compare performance across different months and library sizes.
7The Branded-to-Unbranded Ratio
The branded-to-unbranded ratio is calculated by dividing your branded visibility rate by your unbranded visibility rate. A high ratio suggests you are primarily reaching people who already know you. A falling ratio indicates that your discovery efforts are improving, which is a critical sign of growth in the AI search era.
8Interpreting Search Results
When reviewing results, look beyond the numbers by sampling five to ten actual AI answers. Analyze how your brand is described and which sources the engine cites. Check if competitors appear where you do not. This qualitative review helps you understand the context behind your quantitative visibility rates and metrics.
9Scaling with AEO Tools
Manual spreadsheets work for small sets, but automated Answer Engine Optimization tools are necessary for scaling. Tools like HubSpot AEO track prompts across platforms like ChatGPT, Gemini, and Perplexity. They handle the daily data collection, allowing teams to focus on strategic decisions like closing visibility gaps and refining content.
10Actionable Reporting Standards
Reports should always state the date range, engines used, and the total number of answers analyzed. Clearly flag any changes made to the prompt library since the last report. By showing competitor share of voice and top cited domains, you provide a comprehensive view of your brand's standing in AI.
11Strategic Content Adjustments
If unbranded visibility is low, it often signals a need for better top-of-funnel content. For example, creating an FAQ glossary can increase citations for awareness-stage prompts. Use your tracking data to decide which specific topics to cover, ensuring your content directly addresses the questions potential customers are asking AI engines.
12Continuous Optimization Cycle
AI search is a dynamic environment that requires constant monitoring and adjustment. By separating branded and unbranded queries, you gain a clear view of your brand's health. Use these insights to refine your messaging, improve your source authority, and ultimately ensure your brand remains visible to new customers during their research.
Original article: HubSpot Blog — Branded vs. Unbranded Queries in AI Search and Tracking