Reading Your Results
Understand visibility scores, mentions, sentiment, and share of voice.
Visibility Score
The visibility score ranges from 0 to 100 and measures how often and how prominently your brand appears in AI responses. It factors in mention frequency, position within the response, presence of citations, and sentiment.
Score Benchmarks
- 0-20 (Low): Rarely mentioned in AI responses. Infrastructure review needed.
- 20-50 (Moderate): Appearing in some prompts but significant room to improve.
- 50-80 (Good): Consistently mentioned across key prompts.
- 80-100 (Excellent): Frequently appearing across most relevant prompts.
Mention Counts
A mention is any instance where an AI response includes your brand or product name. The dashboard shows total mentions across all tracked prompts, along with a breakdown by AI engine (OpenAI, Google, Perplexity — plus Claude on Enterprise).
Trends over time are just as important as the raw count. Rising mentions suggest your content strategy is working. Declining mentions may indicate competitor activity or AI model updates affecting your visibility.
Using Mention Data
- Compare mentions across engines to identify where you're weakest
- Look for prompts where you have zero mentions to find content gaps
- Track weekly and monthly trends to measure strategy effectiveness
Sentiment Analysis
Being mentioned isn't always positive. AEKO analyzes the context of each AI response to classify mentions as positive, neutral, or negative.
Positive
The AI recommends your product, highlights strengths, or uses phrases like "best," "popular," or "highly rated" alongside your brand.
Neutral
The AI mentions your brand as part of a list or provides factual information without a positive or negative stance.
Negative
The AI cites drawbacks, references negative reviews, or presents other products as better alternatives in the same context.
Tip
When you spot negative mentions, review the original AI response text. Understanding the cause lets you respond with review management, product page improvements, or FAQ content.
Share of Voice
Share of Voice (SOV) compares your brand's mention percentage against competitors for the same prompts. It shows which competitors dominate and how your position shifts over time.
| Brand | Share of Voice |
|---|---|
| Brand A | 35% |
| Brand B | 25% |
| Your Brand | 20% |
| Others | 20% |
Source Analysis
AI engines reference and cite various web sources when generating responses. AEKO tracks which URLs are cited so you can understand what sources AI engines trust.
The Domain Type Distribution chart shows the composition of cited sources (official, blog, marketplace, news media, beauty retail, community, etc.). Click a type to filter by that category.
The Citations by Domain view shows each domain's citation share, type classification, and AI platform distribution. Click a domain to expand its individual URL list.
Source Quality and Visibility
AI engines cite authoritative, well-structured sites more often. Pages with JSON-LD structured data, clear product information, and thorough review content are more likely to be cited.
Drift Detection
Drift detection tracks when any brand (yours and competitors) appears in or disappears from AI responses over time. The summary cards lead with your own brand — My Brand Appeared, My Brand Dropped, and My Brand Avg Rank (across prompts where AI mentions your brand) — with total all-brand movement shown as a secondary line.
Brand Appeared
A brand started appearing in a new prompt. When your own brand is among them, it's highlighted separately.
Brand Disappeared
A brand dropped out of a prompt where it was previously mentioned. Own brand disappearances require prompt attention.
Warning
Disappeared events require prompt attention. Check which prompts lost your mention and strengthen your content in those areas. Ignoring drift gives competitors a chance to consolidate their position.
Comparing results by context
Prompts tracked with context are reported separately for each customer situation. This shows where your brand comes to mind and where competitors still own the answer.
Context
The customer situation sent with a prompt run. Examples: sensitive skin, gift purchase, ingredient checking, or US availability.
Response differences
Compare mentions, citation sources, sentiment, and competitor exposure by context. The same prompt can surface different brands when the shopper situation changes.
How to Use
- Compare exposure by customer situation: See where your brand and competitors are mentioned for each context
- Find content gaps: Identify contexts where competitors appear but your brand is missing
- Check sources: Review which pages get cited by context and prioritize PDP or content fixes
Interpreting Trends
When reading time-series data, focus on medium-term trends rather than short-term fluctuations. AI engine responses can shift significantly around model updates, so 2-4 week trends are more meaningful than day-to-day changes.
Correlate visibility changes with the optimization actions you've taken. Infrastructure improvements (robots.txt, JSON-LD) tend to reflect quickly, while content strategy changes may take several weeks to show impact.
Tips
- If your visibility score drops sharply, check infrastructure issues first (robots.txt changes, site downtime)
- If a competitor's SOV spikes, investigate their recent content activity
- If mentions are rising but sentiment is declining, a negative review or issue may have surfaced