---
schema_version: "1"
id: 2c24b95bf0076ade2152b3d4d3089201d4238c089ac789a0381c5fdaba7d033c
title: Delta Extract
collection: admin
version: current
locale: en-us
portal_url: https://docs.salesforce.com/admin/current/en-us/data/264-0-0/c360_a_incremental_replace.md
canonical_url: https://help.salesforce.com/s/articleView?id=data.c360_a_incremental_replace.htm&release=264.0.0&type=5
source_site: help.salesforce.com
content_type: documentation
content_hash: aaaa3074409d3cf774b3a9459bed87fe9f58c2227bfef75e3859f0b87941f611
description: "Marketing Cloud Engagement supports two types of incremental extracts: delta extract by date and by number."
products:
  - Data 360
release: "264"
---
# Delta Extract

Marketing Cloud Engagement supports two types of incremental extracts: delta extract by date and by number.

## Delta Extract by Number

When Delta Extract by Number is selected, there is a cascading list of fields in the numeric Data Extension data type. Choose a numeric field as the basis for the high-water mark between consecutive extracts.

With this extract method, the Data Extension data is sorted in ascending order according to the Numeric field selected. For the first extract, a high-water mark is tagged next to the ordered highest number (maximum) and data below that mark is sent. By the time of the second extract, more data can accumulate, resulting in a new maximum value. The second extract reflects all rows of data between the initial max value and the updated max value. With each new extract, all new rows of data between the penultimate max value (second-to-last-identified max value) and the last-identified max value get sent.

## Delta Extract by Date

When Delta Extract by Date is selected, there is a cascading list of fields in the Data Extension whose data type is date time. You are prompted to choose a date time field as the basis for the high-water mark between consecutive extracts.

With this extract method, the Data Extension data is sorted in ascending order according to the Date Time field selected. For the first extract, a high-water mark is tagged next to the ordered most recent date time, and data below that mark is sent. By the time of the second extract, more data can accumulate, resulting in a new most recent date time or maximum value. The second extract reflects all rows of data between the initial max value and the updated max value. With each new extract, all new rows of data between the penultimate max value, or second-to-last-identified max value, and the last-identified max value get sent.