Help & User Guide

Quick start

Converting a file takes three steps:

1. Choose your input — click Choose file and pick a CSV, JSON, XLSX, or Parquet file, or switch to URL mode and paste a link to a publicly accessible file.
2. Choose your output format — select CSV, JSON, XLSX, Parquet, SQL, or SQLite from the dropdown.
3. Click Convert — the result appears immediately below with Download and Copy buttons.

No sign-up required for files up to 5 MB, 10 conversions per day. Create a free account to track your usage.

Input methods

File upload

Click the file button (or drag and drop) to select a local file. Supported input formats:

• CSV / TSV — comma or tab-separated values. Delimiter is auto-detected; override it in options if needed.
• JSON — array of objects [{...}] or newline-delimited {...}\n{...}.
• XLSX — Excel workbooks. The first sheet is read by default.
• Parquet — columnar data format common in data engineering.

URL input

Switch to the URL tab and paste any public link — a REST API that returns JSON, a raw file on GitHub, an S3 pre-signed URL, or a Google Sheets CSV export. The service fetches the file and converts it without you having to download it first.

To export a Google Sheet as CSV: File → Share → Publish to web → select the sheet → CSV → Copy link.

Output formats

• CSV — plain-text table, compatible with Excel, Google Sheets, pandas, R, and almost every database. Best for flat, tabular data.
• JSON — array of objects. Perfect for passing data to APIs, web apps, or testing tools. Copy directly to clipboard without downloading.
• XLSX — Excel workbook. Supports multi-sheet output for nested JSON data (see Nested JSON).
• Parquet — binary columnar format with zstd compression. Ideal for large datasets going into data lakes, dbt, BigQuery, or Spark.
• SQL — a CREATE TABLE statement followed by INSERT INTO rows inside a transaction. Ready to paste into any SQL client to seed a database.
• SQLite — a self-contained .sqlite database file. Open it in DB Browser for SQLite, DBeaver, or query it with Python's sqlite3 module. No server needed.

Nested JSON

Most online converters break when your JSON has nested objects or arrays. DataFlow Convert handles these automatically.

Nested objects → dot-notation columns

A field like {"user": {"name": "Alice"}} becomes a column called user.name in CSV or XLSX output. This happens automatically — no configuration needed.

Arrays of objects → strategy picker

When your JSON has an array-of-objects field (e.g. a list of orders inside each customer), converting to XLSX shows a strategy picker:

• Single sheet — arrays are exploded into rows. One parent record with 3 orders becomes 3 rows. Good for small datasets.
• Multi-sheet (recommended) — each array field gets its own sheet. The parent sheet has one row per record; child sheets have one row per array element, linked by a _row column. Keeps your data relational and readable.

Multi-sheet is only available for XLSX output. For CSV, arrays are always exploded into rows.

Column configuration

After selecting a file or URL, click Configure columns → to choose which columns to include in the output and what to name them.

What you can do

• Exclude columns — uncheck any column to leave it out of the output entirely.
• Rename columns — type a new name in the text field to the right of the original name. The original name stays as the source; only the output name changes.
• Reorder columns — use the ▲ ▼ buttons to move columns up or down. The output will follow this order.
• Select all / None — shortcuts at the top of the panel.

Column configuration works for all input and output formats. The schema is fetched from the server, so it works for large files too.

JSON Tools

Available at /tools. These work on raw JSON text — no file upload needed.

JSON Repair

Paste broken or partial JSON and get back a valid, formatted version. Common problems it fixes automatically:

• Truncated fragments — missing closing brackets from copy-pasting a portion of a large JSON
• Trailing commas — {"a": 1,}
• Single quotes — {'key': 'value'}
• Missing quotes around keys — {key: "value"}

JSON Path Extract

Paste a full JSON document and type a dot-notation path to extract a nested value — without manually cutting and risking broken brackets.

• data.bookies.pl → extracts the Polish bookie entry
• orders.0.price → extracts the price from the first order
• Leave the path empty to pretty-print the whole document

Data preview

After converting to CSV or JSON, the first 50 rows appear automatically as a scrollable table below the Download and Copy buttons. You can verify the output looks correct before downloading.

• Long cell values are truncated in the table; hover over a cell to see the full value.
• The row count shows "First 50 of ~12,000 rows" when the file has more rows than shown.
• Binary formats (XLSX, Parquet, SQLite, SQL) do not show a preview.
• Files larger than 2 MB do not show a preview.

Batch conversion (Pro)

Convert multiple files at once at /batch. All files are converted to the same output format and downloaded as a single ZIP.

• Up to 50 files per batch, up to 500 MB total.
• Files that fail to convert are listed in a _errors.json file inside the ZIP — the rest still convert successfully.
• Each file in the ZIP is named after the original file with the new extension.

Batch conversion requires a Pro subscription.

Plans & limits

Manage your subscription from the account page. Cancel at any time — you keep Pro access until the end of the billing period.

Privacy

• No storage. Files are processed in memory and deleted immediately after conversion. Nothing is written to disk beyond the temporary processing window.
• No tracking. We do not sell data or use third-party analytics. Conversion events are logged internally (format, size, timestamp) for quota enforcement — file contents are never logged.
• Encrypted in transit. All connections use HTTPS (TLS 1.2+).

See the full Privacy Policy.

FAQ