# Upload and Process Documents ### Overview This guide walks you through uploading documents to Docspire, processing them with AI extraction, and viewing the results. You can upload single or multiple documents and process them individually or in batches. ### Supported File Formats Docspire accepts the following file formats: • PDF • Text (TXT, RTF) • Excel (XLS, XLSX) • CSV • Images (JPG, PNG) • Word (Docx) • EDI ### Step 1: Upload Documents
1\. From the Home dashboard, click Upload Documents. The Upload Documents dialog opens. 2\. (Optional) Enter a Business ID to tag documents with your internal reference (e.g., PROJECT-123). 3\. (Optional) Select a Document Type from the dropdown. If left empty, Docspire will auto-detect the document type. ![](https://3990462849-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2Fgjpo1gyiI3QKDSNO9maQ%2Fuploads%2FNYDjT2m7KJzAjpOymqKH%2F11%20-%20Upload%20Options.png?alt=media\&token=bb41bd9b-2fda-480f-8cc2-d934d583427c) 4\. Click or drag files into the upload area, or click to browse your computer. Selected files appear below the upload area with a count (e.g., "2 file(s) selected").
5\. Click Upload Documents to upload the files. You will be redirected to the Document Queue screen. ### Step 2: View in Document Queue After uploading, your documents appear in the Document Queue with a Pending status. 6\. Locate your uploaded documents in the list. Latest ones appear on top. Manually uploaded files show "Manual Upload" as the Source.
### Step 3: Process the Document 7\. In the Document Queue, locate the document you want to process. 8\. Click the Process button in the Actions column.
The status changes to Running while Docspire extracts data from your document, and you are redirected to Document Tracking screen.
### Step 4: Track Processing You can monitor document processing in real-time.\ An info panel in the top-right corner shows "Processing Document" while extraction is in progress.
The tracking view displays: • Fields Extracted – Number of fields successfully extracted • Validations Passed – Number of validation rules passed • Validations Failed – Number of validation rules that failed • AI Extraction Accuracy – Confidence percentage for extracted data ### Step 5: View Results Once processing completes, the document status changes to Success. 9\. In the Document Queue, click View Results for the processed document.
The Results view opens, showing the original document alongside the extracted data. #### Refined View The default Refined View displays extracted data in a structured format: • Header fields appear at the top (e.g., Total Volume, Total Gross Weight, Loading Port) • Table data appears in a structured grid with columns matching the source document • The original document is displayed on the left for side-by-side comparison
#### JSON View 10. Click JSON View to see the extracted data in JSON format. This is useful for: • Reviewing the exact data structure • Copying data for use in other applications • Verifying field names for API integration
### Step 6: Download Results Export extracted data in your preferred format: 11\. Click the Download as JSON dropdown in the top-right corner.
12\. Select your preferred format:
FormatDescription
JSONStructured data format for API integration and development
CSVComma-separated values for spreadsheet applications
ExcelMicrosoft Excel format with formatted columns
The file downloads to your computer. ### Document Statuses
StatusDescription
PendingDocument is uploaded and waiting to be processed
RunningAI extraction is in progress
SuccessProcessing completed successfully
In ReviewDocument requires human review (validation failed or low confidence)
ErrorProcessing failed due to an issue with the document
### Available Actions Depending on document status, the following actions are available in the Document Queue:
ActionWhen AvailableDescription
ProcessPendingStart AI extraction on the document
TrackRunning, SuccessView detailed processing status and metrics
View ResultsSuccess, In ReviewSee extracted data alongside the original document
ReprocessSuccess, In ReviewRun extraction again with optional feedback
### Tips for Best Results • Use clear, high-quality scans – Better image quality improves extraction accuracy. • Specify document type when known – Selecting the correct type helps the AI extract the right fields. • Use Business IDs for tracking – Tag documents with your internal reference numbers for easier searching. • Process sample documents first – Test with a few documents before bulk processing. • Review accuracy scores – Documents with lower accuracy may need manual review. ### Related Features **Document Configuration** – Customize extraction fields for your document types. **Validation Rules** – Set up automatic data quality checks. **Integrations** – Connect email or webhooks for automated document ingestion. **Human Review** – Review and correct documents flagged for attention.