# Pipeline FAQs

> Get answers to the most frequently asked questions about Rippling's data pipelines.

## Overview

Answers below cover pipeline questions that are not already explained in the other pipeline guides, including:

-   Maintenance when a third-party API changes
-   Embedding external BI dashboards
-   Initial import time and data warehouse connections
-   Encryption, retention, and third-party credentials
-   Multiple Rippling accounts
-   How Rippling AI uses pipeline data

## Core concepts

### Who is responsible for maintaining pipelines when a third-party API changes?

Custom pipelines built with Forward Deployed Engineering, Professional Services, or independently are the customer's responsibility to maintain when a third-party API changes.

### Can we embed external BI dashboards into Rippling templates?

No. You cannot embed third-party BI views from Tableau, Looker, or Sigma into Rippling templates, 1:1 pages, or review cycles. Instead, use pipeline-ingested data to recreate those data models and visualizations using native Rippling Reports, which can then be embedded in employee profiles and custom internal applications.

## Data architecture and ingestion

### How long does the initial pipeline import take?

Initial sync begins immediately but can take up to 24 hours to fully process depending on data volume.

For more information, see [Pipeline sync management guide](https://developer.rippling.com/documentation/rippling-platform/pipelines/sync-management).

### How does Rippling interact with data warehouses like Snowflake?

You can connect Snowflake directly to Rippling and scope access to specific schemas or views. Use Rippling Transformations alongside SQL and RQL to recreate complex data aggregations and reconcile metrics like sales quota attainment against external BI tools.

For more information, see [Get started with pipelines](https://developer.rippling.com/documentation/rippling-platform/pipelines/getting-started) and [Get Started with Transformations](https://developer.rippling.com/documentation/rippling-platform/transformations/getting-started).

### Are there pipeline considerations for multiple Rippling accounts?

Organizations with multiple Rippling accounts can connect external tools to each account individually. Rippling does not support cross-account data queries, and each account pulls data independently.

## Security, privacy, and compliance

### What are Rippling's data encryption and retention policies?

All ingested data is encrypted in transit using TLS and at rest. Data is stored and processed in AWS us-west-2 (Oregon, US). Data persists for the duration of your contract and is wiped according to standard offboarding protocols if you leave Rippling.

### How are third-party credentials handled during pipeline setup?

API keys and OAuth credentials belong to your external accounts. Rippling uses them solely to authenticate and pull authorized data.

## Rippling AI capabilities and data usage

### How does Rippling AI use pipeline data, and is it used to train models?

Rippling AI acts as an orchestration layer under strict contractual controls. Rippling never permits third-party model providers to train LLMs on your company data or inputs. Approved sub-processors like ChatGPT or Claude are used for real-time inference only. System improvements use aggregate, explicit product feedback.

## See also

-   [Get started with pipelines](https://developer.rippling.com/documentation/rippling-platform/pipelines/getting-started)
-   [Set up a custom pipeline](https://developer.rippling.com/documentation/rippling-platform/pipelines/custom-pipeline-setup)
-   [Permissions for pipelines](https://developer.rippling.com/documentation/rippling-platform/pipelines/permissions)
-   [Pipeline sync management guide](https://developer.rippling.com/documentation/rippling-platform/pipelines/sync-management)

Source: https://developer.rippling.com/documentation/rippling-platform/pipelines/faqs

> For the complete documentation index, see [llms.txt](https://developer.rippling.com/llms.txt). Public pages have a Markdown version at the same URL with .md added.
