---
title: "Export and save results from Strata | Oplema"
description: "Send a finished table to a download, a local folder, Cloud or a connector such as Postgres with Output. Only Cloud and connector outputs can be scheduled."
url: "https://oplema.com/docs/strata/io/output"
updated: "2026-11-08"
---

# Output

Output ends a flow by delivering a table: as a browser download, to a folder on your computer, to Cloud or through a connector. Output is what makes Run deliver something. Without one, Run has nothing to do.

**Learn this in context:** [Export results to CSV and make a first chart](https://oplema.com/learn/export-to-report/export-and-chart) (Build a report from a CSV export)

## Download a table as CSV

Three orders flow into an Output set to Download. The Run button appears because this flow has an Output. The other examples in these docs have no Output, so they have no Run button: their previews show results, but nothing is delivered.

Result
| order | customer | amount |
| --- | --- | --- |
| 1001 | Ada | 120 |
| 1002 | Lin | 75 |
| 1003 | Sam | 240 |

1. Look at the preview. It shows the three rows Output will deliver, orders 1001, 1002 and 1003.
2. Select Run local at the top. Your browser downloads `orders.csv` with those three rows.
3. Change Filename to `march` and Format to `JSON array (.json)`, then select Run local again. This time you get `march.json`.

## When to use

Use Output when the table is finished and has to go somewhere: a CSV or Excel file you can open, a shared folder or a database table another team reads. If the table only feeds a dashboard, use Vista Source instead; it doesn't write anything.

## Choose a destination

| Destination | What happens on Run | Use it when |
| --- | --- | --- |
| Download | Your browser downloads the file, like any other download. | You want one copy now. Nothing to set up. |
| Local folder | The file is written to a folder you pick on your computer. Your browser asks for access to the folder. Some browsers don't offer folder access; you then see "Folder access is unavailable in this browser." | You run the flow repeatedly and want the file in the same place each time. |
| Cloud | The file is written to a Cloud folder. Running again overwrites it in place. | Other people or a schedule need the file. |
| Connectors | The table is written through a saved connection. | The result belongs in a database, a spreadsheet or shared storage. |

Download and Local folder can't be scheduled. To schedule a flow, choose Cloud or a connector. See [Scheduling](https://oplema.com/docs/cloud/scheduling).

File destinations can write CSV, compressed CSV, JSON, JSON Lines, Parquet, Arrow and Excel.

## Connectors

Output can write to these connectors. Read each [connector guide](https://oplema.com/docs/strata/io/connectors) for setup, permissions and limits.

- **Sheets and files:** [Google Sheets](https://oplema.com/docs/strata/io/connectors/google-sheets), [OneDrive](https://oplema.com/docs/strata/io/connectors/onedrive), [Dropbox](https://oplema.com/docs/strata/io/connectors/dropbox).
- **Databases:** [Supabase](https://oplema.com/docs/strata/io/connectors/supabase), [SQL Server](https://oplema.com/docs/strata/io/connectors/sql-server), [Postgres](https://oplema.com/docs/strata/io/connectors/postgres), [MySQL](https://oplema.com/docs/strata/io/connectors/mysql), [BigQuery](https://oplema.com/docs/strata/io/connectors/bigquery), [Snowflake](https://oplema.com/docs/strata/io/connectors/snowflake), [Databricks](https://oplema.com/docs/strata/io/connectors/databricks).

Database destinations can create a new table, append rows or replace the existing rows. Connector writes cost $0.01 per GB after the first 50 MB of each write, and Cloud writes are charged the same way. Connector traffic passes through Oplema servers.

To feed a dashboard without writing a file, use [Vista Source](https://oplema.com/docs/strata/io/vista-source).

To read data into a flow, use [Input](https://oplema.com/docs/strata/io/input).

## Inputs and outputs

### Inputs

1 port

**In Required**

Accepts: Tables, Geospatial data.

Accepts one connection on this port.

### Outputs

0 ports

No output ports

This block has no outgoing connections.

## Configure the block

All 73 settings

**Destination**

Where to deliver the output.

Default: `File`

Options: File, Cloud, Connectors.

**File delivery**

Where the file goes: a folder on your computer, or a download in your browser.

Default: `Download`

Options: Local folder, Download.

**Cloud Folder**

Cloud folder to write the file into. Re-runs overwrite the file in place.

**Connector**

Connector to write to.

Default: `Google Sheets`

Options: Google Sheets, OneDrive, Dropbox, Supabase, SQL Server, Postgres, MySQL, BigQuery, Snowflake, Databricks.

**Connection**

The help for this setting changes with the connector you choose.

**Write as**

Columns with values Sheets can't preserve accurately are exported as text.

Default: `Google Sheet`

Options: Google Sheet, File.

**Save file**

Create a new file on every run, or replace the contents of a selected file.

Options: Choose how to save, Create new file, Replace selected file.

**File**

Replace this file's contents, keeping its name and location.

**Sheet**

Spreadsheet to write to, or Create new spreadsheet.

**Name**

Name of the new spreadsheet.

Default: `Output`

**Folder**

Drive folder for the new spreadsheet or file. The default is My Drive.

Default: `root`

**Tab**

Tab to write to, or Create new tab.

**Tab name**

Name of the new tab.

Default: `Sheet1`

**Connection**

• The first 50 MB of each write is free. Additional data is charged at the transfer price and needs available usage. For Excel, the whole workbook counts. • Replacing an Excel sheet keeps the other sheets. • For other formats, Replace replaces the entire file.

**Write**

Create a new file, or replace the contents of a selected file.

Options: Choose how to save, Create new file, Replace selected file.

**Folder**

OneDrive folder for the new file, or the folder that holds the file to replace.

**File**

Replace this file, keeping its name and location.

**If file exists**

Fail if the name is taken, or save a new file with an available name on each run.

Default: `Fail`

Options: Fail, Keep both.

**Excel tab**

Other tabs are kept.

**Tab name**

Name of the new Excel tab.

Default: `Sheet1`

**Connection**

• The first 50 MB of each write is free. Additional data is charged at the transfer price and needs available usage. For Excel, the whole workbook counts. • Replacing an Excel sheet keeps the other sheets. • For other formats, Replace replaces the entire file.

**Write**

Create a new file, or replace the contents of a selected file.

Options: Choose how to save, Create new file, Replace selected file.

**Folder**

Dropbox folder for the new file, or the folder that holds the file to replace.

**File**

Replace this file, keeping its name and location.

**If file exists**

Fail if the name is taken, or save a new file with an available name on each run.

Default: `Fail`

Options: Fail, Keep both.

**Excel tab**

Other tabs are kept.

**Tab name**

Name of the new Excel tab.

Default: `Sheet1`

**Connection**

• The first 50 MB of each write is free. Additional data is charged at the transfer price and needs available usage. • Use your administrator’s host, port, database and database-user credentials, with the required TLS setting. • Your administrator must allow Oplema’s backend network access and database permissions. Local/private databases need a route from that backend. • Replace keeps the table’s columns and configuration. Append and Replace require compatible columns. Empty replacement is blocked. • Replace requires TRUNCATE permission and is blocked when row-level security applies to this connection, or by dependent foreign keys.

**Schema**

Schema that holds the destination table.

Default: `public`

**Table**

Table to write to, or Create new table.

**Table name**

Name of the new table.

Default: `output`

**Connection**

• The first 50 MB of each write is free. Additional data is charged at the transfer price and needs available usage. • Use your administrator’s host, port, database and database-user credentials, with the required TLS setting. • Allow Oplema’s backend network access and the MySQL user’s connecting host. Local/private databases need a route from that backend. • Append and Replace require compatible columns in an InnoDB table. Replace keeps the table; triggers and foreign-key cascades still apply. Empty replacement is blocked. • A failed initial load can leave a newly created empty table.

**Table**

Existing tables must use InnoDB.

**Table name**

Created in this connection’s database.

Default: `output`

**Connection**

• The first 50 MB of each write is free. Additional data is charged at the transfer price and needs available usage. • Use SQL username/password authentication and a host with port, or host\\instance without a port. Verify server certificates in production. • Your administrator must allow Oplema’s backend network access and database permissions. • Replace keeps the table’s columns and configuration; constraints and triggers stay active. Append and Replace require compatible columns. Empty replacement is blocked. • Omit identity, computed and rowversion columns.

**Schema**

Schema that holds the destination table.

Default: `dbo`

**Table**

Table to write to, or Create new table.

**Table name**

Name of the new table.

Default: `output`

**Connection**

• The first 50 MB of each write is free. Additional data is charged at the transfer price and needs available usage. • Copy the Project URL from Supabase → Connect and a secret key from Settings → API Keys. Secret keys can bypass row-level security. • Input and Output can share a connection. For Output, copy a PostgreSQL connection string from Connect → Session pooler; replace \[YOUR-PASSWORD\] with your database password and URL-encode special characters. • Replace keeps the table’s columns, policies and indexes. Append and Replace require compatible columns. Empty replacement is blocked. • Replace requires TRUNCATE permission and is blocked by dependent foreign keys.

**Schema**

Supabase-managed schemas are hidden.

Default: `public`

**Table**

Table to write to, or Create new table.

**Table name**

New tables enable row-level security and remove public access.

Default: `output`

**Project selection**

Pick from your projects, or enter a project ID directly. Use the ID, not the display name.

Default: `Choose project`

Options: Choose project, Enter project ID.

**Project ID**

Google Cloud project ID, not its display name.

**Project**

Destination project; also runs the load job.

**Dataset**

Output doesn't create datasets.

**Table**

Table to write to, or Create new table.

**Table name**

Column names must use letters, numbers or underscores and start with a letter or underscore.

Default: `output`

**Connection**

• The first 50 MB of each write is free. Additional data is charged at the transfer price and needs available usage. • Use your account identifier, programmatic access token and optional warehouse; blank uses your Snowflake default. • The token needs warehouse/data access, and network policies must allow Oplema’s backend. Replace an expired token by creating a connection with a new one. • Append and Replace require compatible columns. Replace keeps the table; all batches commit together. • Writes support ordinary, untagged tables without access or masking policies. Empty replacement is blocked. • A failed initial load can leave a newly created empty table.

**Database**

Database that holds the destination table.

**Schema**

Schema that holds the destination table.

**Table**

Table to write to, or Create new table.

**Table name**

Names preserve their exact spelling.

Default: `OUTPUT`

**Connection**

• The first 50 MB of each write is free. Additional data is charged at the transfer price and needs available usage. Databricks limits each write to about one million encoded parameter characters. • Find Server hostname and HTTP path in SQL Warehouses → your warehouse → Connection details. • Use a personal access token with warehouse/data access; create a connection with a new token when it expires. Allow Oplema’s backend to access the workspace and download query results. • Append and Replace require compatible Unity Catalog Delta tables without row filters or masks. Replace keeps the table. Empty replacement is blocked. • A failed initial load can leave a newly created empty table.

**Catalog**

Catalog that holds the destination schema.

**Schema**

Schema that holds the destination table.

**Table**

Table to write to, or Create new table.

**Table name**

Names preserve their exact spelling.

Default: `output`

**Write**

Replace overwrites existing rows.

Default: `Replace`

Options: Replace, Append rows.

**Folder**

The local folder the file is written to.

**Filename**

Name of the saved file. The format supplies the extension.

Default: `output`

**Filename**

The format supplies the extension.

Default: `output`

**Format**

File format to write. It also sets the file extension.

Default: `CSV (.csv)`

Options: CSV (.csv), Compressed CSV (.csv.gz), JSON array (.json), JSON Lines (.jsonl), JSON Lines (.ndjson), Parquet (.parquet), Arrow IPC (.arrow), Arrow IPC (.ipc), Excel workbook (.xlsx).

**Sheet**

Name of the sheet to write in the Excel workbook.

Default: `Sheet1`

**If file exists**

What happens when the folder already has a file with this name: replace it, keep both or stop with an error.

Default: `Replace file`

Options: Replace file, Keep both, Stop with error.

**If sheet exists**

Create a missing sheet; replace or stop if it already exists. Other sheets are preserved.

Default: `Replace sheet`

Options: Replace sheet, Stop with error.

**Separator**

The character placed between values in the CSV file.

Default: `Comma`

Options: Comma, Semicolon, Tab, Pipe, Custom.

**Custom separator**

A separator of one to four UTF-8 bytes.

**Include column names**

Write the column names as the first line of the CSV file.

Default: `true`

**Quoting**

Automatic quotes a value only when it needs quotes. Always quotes every value.

Default: `Automatic`

Options: Automatic, Always.

**Encoding**

UTF-8 with BOM adds a marker at the start of the file. Some spreadsheet apps use it to recognize the encoding.

Default: `UTF-8`

Options: UTF-8, UTF-8 with BOM.

**Line endings**

The characters that end each line: LF, or CRLF, which is common on Windows.

Default: `LF`

Options: LF, CRLF.

**Null marker**

The text written for a missing value. Leave empty to write nothing.

## Output FAQs

**How do I save as an Excel file?**

Set Format to Excel workbook (.xlsx), so the Output block writes an xlsx file.

**Why doesn't my flow produce anything when I run it?**

Run delivers to Output blocks. A flow with no Output has nothing to deliver, although its previews still show results.

**Which destinations can I schedule?**

Cloud and connectors. Schedules reject Download and Local folder because no browser is open during a scheduled run.

**What is the difference between Output and Vista Source?**

Output writes a table to a destination when you run the flow. Vista Source writes nothing; it only makes the table available to Vista.

**What does writing through a connector cost?**

The first 50 MB of each connector or Cloud write is free. Additional data is charged at the transfer price and needs available usage.

**How large can a connector write be?**

There's no fixed size limit. A write stops when it passes the free 50 MB and your available usage. Databricks limits each write to about one million encoded parameter characters.

### Prepare your data. Build it in Strata.

Strata opens in your browser, with no account.

[Open Strata](https://oplema.com/editor?path=%2F%3Fopen%3Dstrata) [About Strata](https://oplema.com/strata)
