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

# Overlap

> Overlap is a technique in fixed-size chunking where adjacent chunks share a set number of tokens at their boundaries.

*Technical implementation* · *AI Search Infrastructure*

## Definition

Overlap is a technique in fixed-size chunking where adjacent chunks share a set number of tokens at their boundaries. A chunk beginning at position 500 and ending at 1000 might share tokens 950–1000 with the next chunk beginning at 950.

## Why It Matters for AI Search

Reduces the chance of a relevant sentence falling at a boundary and being lost across two incomplete chunks. The tradeoff: larger indexes and the possibility of retrieving the same content twice from adjacent overlapping chunks. Standard overlap windows run 10–20% of chunk size. For content strategy, overlap is a mitigation for poor structure — self-contained sections reduce its necessity.

## Related Terms

<CardGroup cols={2}>
  <Card title="Fixed-size chunking" href="/ai-search-glossary/fixed-size-chunking" />

  <Card title="Chunking" href="/ai-search-glossary/chunking" />

  <Card title="Embedding" href="/ai-search-glossary/embedding" />

  <Card title="RAG" href="/ai-search-glossary/rag" />

  <Card title="Vector database" href="/ai-search-glossary/vector-database" />
</CardGroup>

## Relevant Plate Lunch Collective Services

[AI SEO](https://www.platelunchcollective.com/services/ai-seo)