Superlines tracks several core metrics to help you understand and optimize your brand’s AI search visibility.
Brand Visibility
What it measures: The percentage of AI responses that mention your brand.
Brand Visibility = (Responses mentioning your brand / Total responses) × 100
| Benchmark | Rating |
|---|
| > 30% | Excellent |
| 20–30% | Good |
| 10–20% | Fair |
| < 10% | Needs improvement |
Brand Visibility is the single most important metric — it tells you how visible your brand is when people ask AI about topics relevant to your business.
Share of Voice (SOV)
What it measures: Your brand mentions as a percentage of all brand mentions (yours + competitors).
Share of Voice = (Your brand mentions / All brand mentions) × 100
| Benchmark | Position |
|---|
| > 25% | Market leader |
| 15–25% | Strong presence |
| 5–15% | Moderate |
| < 5% | Needs attention |
Key difference from Brand Visibility
- Brand Visibility = how often you’re mentioned out of all responses
- Share of Voice = how often you’re mentioned compared to competitors
SOV tells you about your competitive position. If your SOV is low but Brand Visibility is decent, it means competitors are also being mentioned frequently.
Citation Rate
What it measures: The percentage of AI responses that cite (link to) your website.
Citation Rate = (Responses citing your website / Total responses) × 100
| Benchmark | Rating |
|---|
| > 20% | Excellent |
| 10–20% | Good |
| 5–10% | Fair |
| < 5% | Needs improvement |
Citations are a strong signal of content authority. AI models cite sources when your content is seen as authoritative and relevant. Focus on creating comprehensive, well-structured content to improve citations.
Share of Citations
What it measures: Your website citations as a percentage of all citations in branded responses.
Share of Citations = (Your citations / Total citations in branded responses) × 100
Mentions
The raw count of times your brand is mentioned across all AI responses. While Brand Visibility is a percentage, Mentions gives you the absolute number — useful for understanding volume and trends.
Average Position
What it measures: The average rank at which your brand is mentioned in AI-generated answers. Position 1 means your brand is mentioned first, Position 2 means second, and so on.
Average Position = Volume-weighted average of your brand's mention order across all responses
| Benchmark | Rating |
|---|
| 1.0–2.0 | Excellent — consistently mentioned first |
| 2.0–3.5 | Good — among the top recommendations |
| 3.5–5.0 | Fair — mentioned but not prominently |
| > 5.0 | Needs improvement |
Only responses where your brand is actually mentioned are included in the calculation. Responses with no mention of your brand do not affect your Average Position. See Brand Visibility and Mentions for how often your brand appears. Lower is better.
Average Position Ranking
The leaderboard ranks all brands by their volume-weighted average position across your tracked prompts. This lets you see exactly where your brand stands relative to competitors — who gets mentioned first, and who gets mentioned last.
- Rank change shows whether your relative position improved or declined vs. the previous period
- Brands with fewer than 3 appearances are excluded to filter noise
- The ranking updates daily and supports competitor comparison in the trend chart
Average Position is particularly useful for understanding quality of visibility. A brand can have high Brand Visibility (mentioned often) but poor Average Position (mentioned last). Improving your position means AI models see your brand as a more primary recommendation.
Sentiment Score
The average sentiment of AI responses that mention your brand:
- Positive — the AI presents your brand favorably
- Neutral — factual mention without strong sentiment
- Negative — the AI raises concerns about your brand
Metrics by engine
All metrics can be filtered and broken down by AI engine. This helps you identify which engines your brand performs best on, spot engines where competitors are winning, and prioritize content strategies per platform. 
