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

# Incremental

> Incremental sync only transfer new or updated data, minimizing data transfer

### Overview

Incremental syncing involves transferring only new or updated data, thus avoiding duplication of previously replicated data. This is achieved through deduplication using a unique primary key specified in the model. For initial syncs, it functions like a full refresh since all data is considered new.

### Example

### Initial State

Suppose the following records already exist in our source:

| Name             | Plan     | Updated At |
| ---------------- | -------- | ---------- |
| Charles Beaumont | freemium | 6789       |
| Eleanor Villiers | freemium | 6790       |

### First sync

In this sync, the delta contains an updated record for Charles:

| Name             | Plan     | Updated At |
| ---------------- | -------- | ---------- |
| Charles Beaumont | freemium | 6791       |

After this incremental sync, the data in the warehouse would now be:

| Name             | Plan     | Updated At |
| ---------------- | -------- | ---------- |
| Charles Beaumont | freemium | 6791       |
| Eleanor Villiers | freemium | 6790       |

### Second sync

Let's assume in the next delta both customers have upgraded to a paid plan:

| Name             | Plan | Updated At |
| ---------------- | ---- | ---------- |
| Charles Beaumont | paid | 6795       |
| Eleanor Villiers | paid | 6795       |

The final data at the destination after this update will be:

| Name             | Plan | Updated At |
| ---------------- | ---- | ---------- |
| Charles Beaumont | paid | 6795       |
| Eleanor Villiers | paid | 6795       |