# Meet your customers' agents where they work

Your customers' AI agents do their analysis in the data warehouse. Prequel syncs your product's live data there, white-labeled end to end, so customers can run the analysis they need.

[Book a demo](https://www.prequel.co/request-product-demo)

## Agents work in the data warehouse

When an agent needs data, the shortest path is a query. Your customers' warehouses are built for exactly that.

- **A query is a single step** — In the data warehouse, an agent reads your schema, writes a SELECT, and has structured results seconds later. That loop is what agents are built to do well.
- **Your data, next to all of theirs** — The analysis customers care about spans systems: your product's events joined against their billing, CRM, and support data. The data warehouse is the one place those joins happen.
- **Warehouse-native agents are on the rise** — Snowflake Cortex, Databricks Genie, and their peers run inside the customer's data platform itself. The data they can work with is the data that's synced in. Prequel syncs the data there.

## Two halves of agent-readiness

If the question is about one row, it belongs on MCP. If it's about the whole table, it belongs on exports.

### MCP — Point lookups and actions

"What's the status of invoice #4521?" Lookups and actions with small payloads are what MCP does well, and your customers' agents will expect the surface. Ship one.

[How agents should access your data](https://www.prequel.co/guides/mcp-vs-data-exports)

### Data Exports — Aggregates, trends, and joins

Whole-table questions run best where the data lives. In the customer's data warehouse, one SQL query does the job in seconds and costs fractions of a cent. Prequel keeps your product's data synced there continuously.

[Explore data export](https://www.prequel.co/data-export)

## Be part of the analysis customers run

A growing share of your customers' analysis happens outside your product: in their own data warehouse, run by their analysts and their agents, joined against the rest of their data.

Warehouse syncs put your product's data in the middle of that work. It lands as regular tables, current within minutes, ready for whatever the customer wants to build on it.

| | |
| --- | --- |
| **Before** | To use your data, a customer's agent logs into your product, drives a UI built for humans, and scrapes or exports what it finds. |
| **With Prequel** | The agent finds your data already sitting in the data warehouse it works in, and answers with a single query. |

[Book a demo](https://www.prequel.co/request-product-demo)

## Built for enterprise scale

- **Trillions of rows, monthly** — Prequel syncs trillions of rows to customer data warehouses every month, across 20+ databases, data warehouses, and object stores.
- **Multi-tenant by design** — Every query Prequel runs against your source is filtered by recipient ID, auditable in your own query logs.
- **White-labeled end to end** — Embedded in your product, styled to your brand. Your customers never know Prequel exists.
- **Nothing stored, SOC 2 Type II** — Data moves through ephemeral workers. No data is ever stored or retained on Prequel infrastructure.

## Go deeper

- [Data Export](https://www.prequel.co/data-export) — Let customers sync your data to their warehouse. Embedded in your product and fully white-labeled.
- [Customer Experience](https://www.prequel.co/customer-experience) — Prequel sends your customers clean, up-to-date data. There's nothing to clean, code, or parse.
- [Guide: MCP vs. Data Exports — How AI Agents Should Access Your Product's Data](https://www.prequel.co/guides/mcp-vs-data-exports) — MCP gives agents a way to look up records and take actions. Data exports give them a way to answer analytical questions.