Branded
Own-brand
“What does Brand Armor AI do?”
AccuracyPositioningSentimentHallucinations
Caution
The brand is guaranteed to be in the question, so a mention is not a discovery win.
Representation and discovery are different jobs
Branded vs non-branded prompts for AI visibility
Branded prompts reveal how AI describes a company it was explicitly asked about. Non-branded prompts reveal whether the company is discovered when buyers ask about a category, problem, comparison, or purchase. Blend them into one score and you can turn assisted recognition into a fake market win.
Branded question
“Is Brand Armor AI good for marketing teams?”
Use it to evaluate accuracy, sentiment, positioning, features, pricing statements, and risks.
Non-branded question
“Which platforms track visibility in AI answers?”
Use it to evaluate discovery, category inclusion, recommendation share, competitors, and sources.
Prompt intent matrix
Six prompt families, six measurement jobs
Non-branded
Category discovery
“What are the best AI visibility platforms for marketing teams?”
Unassisted discoveryRecommendation shareCategory authority
Caution
Broad wording may represent several buyer segments; cluster by fit and intent.
Non-branded
Problem or use case
“How can I find why competitors appear in ChatGPT instead of my company?”
Problem relevanceBuyer-job coverageSolution framing
Caution
Avoid vague educational prompts that no commercial page should reasonably win.
Competitor-branded
Competitor and alternative
“What are alternatives to [competitor] for crawler monitoring?”
SubstitutionComparative positioningShortlist entry
Caution
The named competitor receives an assisted advantage; report it separately.
Usually non-branded
Purchase decision
“Which AI visibility tools support multiple markets and scheduled prompts?”
Feature fitPurchase intentPackage relevance
Caution
Confirm that the compared facts are current before treating the answer as reliable.
Variant
Market and language
“Best AI visibility software for a Swedish ecommerce company”
Regional discoveryLanguage coverageLocal competitors and sources
Caution
Do not compare markets as if model, language, prompt, and location context were identical.
Separate scorecards
Do not let easy branded mentions inflate discovery
Keep each prompt family visible in reporting. An executive can then see that the brand is described accurately while still being absent from unassisted category recommendations.
| Scorecard | Set | Observed result | Reported metric |
|---|---|---|---|
| Own-brand accuracy | 10 branded prompts | 9 accurate answers | 90% accuracy |
| Category discovery | 20 non-branded prompts | 5 recommendations | 25% recommendation share |
| Problem relevance | 15 non-branded prompts | 6 relevant mentions | 40% problem coverage |
| Competitor substitution | 10 competitor prompts | 2 alternative mentions | 20% substitution coverage |
ANTI-PATTERN 01
Vanity-heavy
70% own-brand · 20% competitor · 10% discovery
Looks strong because the brand is named in most questions, but says little about being discovered.
ANTI-PATTERN 02
Discovery-only
80% category · 20% problem · 0% branded
Measures market discovery but misses inaccurate pricing, stale positioning, and hallucination risk.
ANTI-PATTERN 03
Paraphrase-heavy
50 versions of “best tool”
Creates execution volume without covering distinct buyer decisions or content opportunities.
A balanced operating mix
Weight category, problem, comparison, and purchase prompts around unassisted buyer discovery. Retain a smaller branded layer for accuracy and reputation. Add competitor and market variants only when they represent decisions your team can act on. The correct mix is the one that matches your business questions, not a universal percentage copied from another company.
From prompt to action
Different losses require different fixes
A wrong branded answer is an accuracy problem. Missing from a category answer is a discovery problem. Losing an alternative query is a positioning problem. The prompt type tells you where to investigate.
Explore content gap analysisBranded inaccuracy
Review current owned facts, cited sources, and conflicting third-party descriptions.
Category absence
Inspect competitor recommendations, category framing, and missing buyer-fit pages.
Comparison loss
Clarify tradeoffs, alternatives, use-case fit, proof, and independent validation.
Market inconsistency
Compare language, local sources, regional competitors, and market-specific content gaps.
Track the right questions
Keep brand accuracy, category discovery, comparison wins, and local-market visibility distinct
Brand Armor AI lets teams choose or generate representative prompts, accept them into recurring monitoring, and connect each result to citations, competitors, content gaps, blogs, and UGC suggestions.
What you can measure
Suggested and custom prompts for buyer-relevant monitoring
Recurring results without manually executing every question
Prompt-level competitors, citations, sentiment, and recommendation visibility
Country and language-aware monitoring for supported workflows
Frequently asked questions
Branded and non-branded prompt FAQ
What is a branded AI visibility prompt?
A branded prompt explicitly names your company, product, or competitor. It is useful for testing accuracy, positioning, sentiment, pricing, features, comparisons, and hallucination risk.
What is a non-branded prompt?
A non-branded prompt asks about a category, problem, use case, or desired outcome without naming your brand. It is the stronger test of whether AI systems discover and recommend you without being prompted.
Should branded prompts be excluded from visibility scores?
They should usually be separated rather than discarded. Branded prompts measure representation and accuracy, while non-branded prompts measure discovery and recommendation opportunity. Blending them can inflate a score.
Are competitor prompts branded or non-branded?
They are branded because they name a competitor, but they answer a different question: whether your brand is recognized as an alternative, substitute, or comparison candidate.
What prompt mix should a company use?
The mix should follow business goals. Discovery-focused teams should weight category, problem, and comparison questions heavily while retaining a smaller branded set for accuracy and reputation monitoring.
Further reading
Crawler identities and platform behavior change. Use provider documentation as the source of truth and review it before changing robots, firewall, or CDN rules.
Representative AI Search Prompt Library
Aleyda Solis outlines why branded, non-branded, competitor, buyer-stage, geography, and persona questions belong in a representative library.
Brand Armor AI prompt monitoring
Product details for monitored prompts, recurring results, citations, competitors, and content gaps.
Related guides
Continue the measurement journey
Measurement
Can you see ChatGPT visitor prompts?
Learn what ChatGPT referral data and crawler logs can reveal, what remains private, and how to build a reliable demand picture without inventing attribution.
Read guideCrawler intelligence
Does crawler traffic mean a recommendation?
Separate crawler access from citations, mentions, recommendations, human visits, and conversions using an evidence-based measurement ladder.
Read guideMeasurement
ChatGPT referrals vs crawler traffic
Understand the difference between a person arriving from ChatGPT and an AI system requesting a page from your server.
Read guideCrawler intelligence
AI crawler purpose guide
A practical guide to OpenAI, Anthropic, Perplexity, and Google crawler identities, controls, and the evidence each request provides.
Read guidePrompt strategy
How many prompts should you track?
Choose a defensible prompt-set size based on buyer journeys, coverage, model variance, markets, and the decisions your team needs to make.
Read guideMeasurement
How quickly does AI search update?
Understand the separate timelines for crawler access, retrieval, indexing, citations, recommendations, and future model training.
Read guide