--- title: "Set up the Document Extractor" description: "Set up the Document Extractor to automatically process receipts, invoices, and bank statements from email forwarding or direct uploads." section: "Document Extractor" canonical: "https://banksync.io/docs/document-extractor/extractor-setup" --- The Document Extractor automatically extracts data from documents (receipts, invoices, and bank statements) and syncs it to your connected sheet, database, or table. Process documents by forwarding emails or uploading files directly. > **Before you start:** Connect at least one integration (Google Sheets, Notion, or Airtable) under the Integrations tab. You will pick it on the Destination tab below. Also note your plan's monthly document quota: Free 10, Starter 30, Standard 100, Professional 1,000, Business 2,000 documents per month, resetting on the 1st. A Document Extractor feed is a kind of feed, so you create it from the Feeds tab and configure it across four tabs: **Document**, **Sources**, **Destination**, and **Mapping**. Progress autosaves as you go. **Create the feed** 1. **Open Feeds and add a feed** — In the Feeds tab, choose Add feed to start a Document Extractor. 2. **Pick what you are extracting** — When prompted "What kind of documents are you extracting?", choose a preset under General Documents (Receipts, Invoices, Document) or Financial Statements (Transactions, Balances, Trades, Holdings, Loans). For a bank statement, choose Transactions. 3. **Name it** — Give the feed a name like "Receipts" or "Vendor invoices" so it is easy to find later. [Video: Receipt-to-Spreadsheet Automation](https://www.youtube.com/watch?v=fSFszvGGYKA) [![Setting up the document extractor: picking Extract Data from Statements as the feed source, choosing the Receipts preset, uploading receipt-starbucks.pdf, and watching it move from processing to processed with the extracted $8.45 total.](https://cdn.banksync.io/videos/document-extractor-setup.poster.c9df1291f054188c.png)](https://cdn.banksync.io/videos/document-extractor-setup.b1b8e68de5f36c74.mp4) [Watch: Setting up the document extractor: picking Extract Data from Statements as the feed source, choosing the Receipts preset, uploading receipt-starbucks.pdf, and watching it move from processing to processed with the extracted $8.45 total.](https://cdn.banksync.io/videos/document-extractor-setup.b1b8e68de5f36c74.mp4) ## 1. Document: choose what to extract The **Document** tab is where you confirm the document type and which fields to pull out. 1. **Confirm the document type** — The Document tab shows the preset you picked (for example Receipts or Invoices). Switch presets here if you chose the wrong one. 2. **Review the extraction fields** — Each field (for example merchant, date, total, tax, line items) shows what the AI will pull out. Keep the ones you want and remove any you do not need. 3. **Reset to preset (optional)** — If you edited the fields and want to start over, use Reset to preset to restore the defaults for that document type. > **Mixed documents:** 'Create one feed per document type. A receipts feed and an invoices feed can both point at the same destination but extract different fields.' ![The extractor's document-type preset picker showing preset cards for general documents and financial statements, each with a short description of the fields it will extract.](https://cdn.banksync.io/screenshots/extractor/data-type-picker.662ebfac0b05ba45.png "The Document step: pick a preset that seeds the fields BankSync will extract.") ## 2. Sources: how documents arrive The **Sources** tab is where documents come in. You can use email forwarding, manual upload, or both. **Email inbox** 1. **Copy your inbox address** — Each feed has its own email address. Copy it from the Sources tab. 2. **Forward documents to it** — Forward any email with a document attached (or set your email client to auto-forward from specific senders). Attachments are picked up automatically. 3. **Restrict senders (optional)** — Add addresses to the whitelist so only forwards from people you trust are processed. **Manual upload** 1. **Drop files in** — Drag PDF, JPG, or PNG files onto the upload area (up to 10 MB each), or click to browse. 2. **Watch them process** — Each file becomes a job and is extracted within seconds. Recent documents and their status appear in the list. ## 3. Destination Choose where extracted data lands. This works the same as a sync feed: pick a connected integration, then the specific sheet, database, or table. ## 4. Mapping The **Mapping** tab connects each extracted field to a column in your destination. It has subtabs: **Fields**, **Enrichments**, and (once a destination table is chosen) **Advanced**. 1. **Fields: map to columns** — On the Fields subtab, select the column in your sheet, database, or table that each extracted field should write to. Unmapped fields are skipped. 2. **Enrichments: clean values (optional)** — On the Enrichments subtab, apply rules to clean or categorize values before they are written. 3. **Advanced: choose a row layout** — On the Advanced subtab, set Itemization: Row Per Document for one row per receipt/invoice, or Row Per Item for one row per line item. [Configure field mapping](/docs/document-extractor/field-mapping) ## Confirm it worked Forward an email or upload a file, then open the feed's history. The job should move from Processing to Completed (or Failed if something went wrong), and a new row should appear in your connected sheet, database, or table with the extracted values. > **If a document does not appear:** Check that the file is a PDF/JPG/PNG under 10 MB, that the sender is on the whitelist (if you enabled one), and that the Mapping tab has at least one field mapped to a column. If you have used your plan's monthly document quota, further documents are not processed until the counter resets on the 1st (or you upgrade). Faded or low-resolution scans extract less reliably, so review the values before relying on them. ## Related guides - [Sending documents by email](/docs/document-extractor/sending-documents-by-email): the inbox address, sender whitelist, and manual upload in depth. - [Map extracted fields to your destination](/docs/document-extractor/field-mapping): per-app mapping tips and row layouts. - [Extract and sync document data](/docs/document-extractor/sync-receipts): the overview of what the Document Extractor captures. - [Managing your plan](/docs/account-billing/managing-your-plan): check document usage against your monthly quota.