# What "Visible" Means — AI Visibility and LLM Brand Discovery

Source: https://www.geekswithgeeks.com/en/ai-visibility/vis-what-visibility-means

> Define mention, citation, recommendation, accuracy and sentiment as separate measurable outcomes.

## Five things to track

**Mention**: the assistant names your brand. **Citation**: it links to one of your pages as a source. **Recommendation**: it suggests you for the user's need ("best for small teams"). **Accuracy**: what it says about you (features, price, location) is correct. **Sentiment**: how it describes you (positive, neutral, negative). Being mentioned with wrong details can be worse than not being mentioned, so measure all five, not only mentions.

## A scorecard row

One record per question-and-answer. Over many questions these become rates you can chart.

```json
{
  "prompt": "best lightweight project tracker for a 5-person team",
  "assistant": "assistant-x (web search on)",
  "date": "2026-10-01",
  "mentioned": true,
  "position": 3,
  "cited_our_url": false,
  "recommended": true,
  "facts_correct": {"price": false, "free_plan": true},
  "sentiment": "positive"
}
```

## Accuracy can be a quick win

Fixing a wrong price or outdated feature on your own pages and key third-party listings is often easier than winning new mentions, and it protects trust right away.

**Quiz:** An assistant names your brand but gives the wrong price. What does this show?

- [ ] That the price is irrelevant
- [ ] Perfect visibility
- [ ] That no one saw it
- [x] A mention exists but accuracy is poor and needs fixing

*Answer:* A mention exists but accuracy is poor and needs fixing. Mentions and accuracy are separate; both must be tracked.