A practical field guide from Automation Ace.
The short answer
A text-based file stores characters you can read directly, like CSV, JSON, XML, or TXT. A binary file stores encoded bytes, like PDF, DOCX, XLSX, images, or ZIP archives, that need special software to interpret. For automation, that difference decides whether a step can use a file's contents directly or must first convert it, for example with Files by Zapier Convert File to Text.
For the app itself, see Files by Zapier explained.
Binary vs text-based files compared
| Text-based files | Binary files | |
|---|---|---|
| What it contains | Human-readable characters | Structured bytes: layout, compression, images, metadata |
| Examples | TXT, CSV, JSON, XML, HTML, Markdown | PDF, DOCX, XLSX, PPTX, images, audio, video, ZIP |
| Open in a text editor? | Readable | Unreadable |
| Map into a text field? | Often directly, once read | Only after conversion or extraction |
| Send to AI | Directly as text | Convert to text first (or use a model that accepts the file type) |
| Parse with Formatter or Code | Yes | Not until converted |
| Common Zapier tool | Files by Zapier Extract Rows, Code steps | Files by Zapier Convert File to Text |
Text-based files
Text files are sequences of characters in an encoding such as UTF-8. Formats like CSV, JSON, and XML add structure through punctuation: commas and line breaks, braces and brackets, or tags. Because they are just text, automation tools can read, search, split, and build them with ordinary string operations.
- CSV: rows and columns; ideal for tabular imports and exports. See CSV automation in Zapier.
- JSON: nested objects and arrays; the language of APIs. See the JSON glossary entry.
- XML: tagged hierarchical data common in enterprise and legacy systems.
Binary files
Binary files store data in formats designed for a specific program. A DOCX is actually a compressed package of XML files and media; a PDF describes how to draw each page; an image stores pixel data. You cannot reliably pull meaningful text out of them with string functions, so automation needs a converter or a dedicated app action.
Some binary files contain no text at all, such as photos and scanned documents, and need OCR before any text can be extracted.
How Zapier handles files in steps
- File fields pass a file, often as a URL that Zapier downloads (hydrates) when a later step needs it. See hydrating files in Code steps and getting file contents.
- Text fields hold strings. To use a binary file's words in a text field, convert it first.
- Links vs files: sharing links are not files; some apps need a direct download link. See Dropbox direct download links and Google Drive link formats.
Choosing formats for your automations
- Moving data between systems? Prefer text formats: CSV for tables, JSON for nested data.
- Sending documents to people? Binary formats like PDF preserve layout and are harder to alter.
- Feeding AI? Convert to text, and send only what the model needs. See converting files to text for AI.
- Exporting from Google Workspace? Choose the format at export. See exporting PDFs and exporting CSV or Excel.
Common problems and fixes
- Garbled text in a field usually means a binary file was treated as text. Convert it first.
- Empty text from a PDF usually means a scanned image. Use OCR.
- Strange characters in a text file usually mean an encoding mismatch. Save as UTF-8.
- “File not found” or tiny files often mean a link to a web page was passed instead of the file. Use a direct download URL.
- Timeouts with very large files: reduce size or split the work. See timeout errors.
For ten ways to put this into practice, see 10 Files by Zapier use cases, then convert your first document with Files by Zapier.
Frequently asked questions
What is the difference between binary and text files?
Text files store human-readable characters, such as CSV, JSON, XML, and TXT. Binary files store encoded bytes that need specific software to interpret, such as PDF, DOCX, XLSX, images, and ZIP archives.
Is a CSV file binary or text?
CSV is a text-based format. Each line is a row and commas separate values, so automation tools can read and build it with string operations.
Is a PDF binary or text?
A PDF is a binary file. To use its words in an automation, convert it to text first, for example with Files by Zapier Convert File to Text. Scanned PDFs also need OCR.
Why does my automation show garbled text from a file?
A binary file such as a PDF or DOCX was probably read as text. Convert the file to text with an appropriate action before mapping its contents.
Which file format is best for moving data between apps?
Text formats: CSV for tabular data and JSON for nested data. Use binary formats like PDF when you need to preserve layout for people.
Disclaimer: Zapier features, plan availability, and settings can change. Confirm current details in Zapier's help documentation and the documentation for the apps that store your files before relying on a specific setting. This article may include links to apps, products, or services; some links may be affiliate links, which means Automation Ace may earn a commission at no extra cost to you.