> ## Documentation Index
> Fetch the complete documentation index at: https://docs.getcited.in/llms.txt
> Use this file to discover all available pages before exploring further.

# Position Bias

> The tendency for users and LLMs to give disproportionate attention and credibility to items listed first.

**Position bias** is the tendency for users (and LLMs themselves) to give disproportionate attention and credibility to items listed first in a sequence. In AI-generated answers, the first brand mentioned receives more reader attention than the fifth brand mentioned — even when the AI does not explicitly rank its suggestions. This effect is well-documented in traditional search (users click the first result far more than the tenth) and carries over into AI answer consumption.

## Why it matters

Position bias means that [average position](/glossary/average-position) is not just a vanity metric — it has real downstream impact on brand perception and consideration. A brand consistently mentioned in positions 1-3 receives materially more reader engagement than a brand consistently mentioned in positions 7-10. Moving from position 6 to position 3 can produce larger business impact than increasing [mention rate](/glossary/mention-rate) by several percentage points.

## How it applies in practice

Cited tracks average position as a core metric precisely because of position bias. The [Cited Index](https://www.getcited.in/cited-index) shows a median average position of 5.6 across all brands — roughly middle-of-pack. Top-decile brands achieve average position 3.3, placing them consistently in the high-attention zone where position bias works in their favor.

## Related concepts

* [Average position](/concepts/metrics/average-position) — the metric that captures position bias effects
* [How Perplexity ranks sources](/concepts/platforms/how-perplexity-ranks-sources) — the platform with the most explicit position signal
* [Share of voice](/glossary/share-of-voice) — a complementary competitive metric