# FAQs

## Docspire Knowledge Hub

*Everything you need to know about AI-powered document processing, extraction, validation, and workflow automation.*

***

### Platform Overview

#### \[What] What is Docspire?

Docspire is an AI-powered document intelligence platform purpose-built for industries that handle complex, document-heavy operations. It automates the extraction, validation, and integration of data from documents like invoices, mortgage applications, pay slips, bank statements, purchase orders, and more.

Backed by Astera, Docspire combines adaptive OCR, context-aware AI extraction, and configurable business rules to process documents with up to 99.5% accuracy across 40+ languages. Unlike generic OCR tools that simply scan text, Docspire understands document context the way industry professionals do — recognizing that a 1003 loan application needs different validation than a tax return, or that invoice discrepancies matter more than typos.

The platform connects every stage of document processing, from ingestion and classification through validation and delivery, with human review only for exceptions.

#### \[What] What industries does Docspire serve?

Docspire serves mortgage lending, accounts payable, logistics, retail, manufacturing, and any document-intensive operation. The platform ships with pre-trained workflows for:

* **Mortgage processing** — applications, pay stubs, bank statements, tax returns, ID documents, property appraisals, insurance policies, title documents
* **Accounts payable** — invoices, purchase orders, receipts, credit notes, vendor statements
* **Logistics & supply chain** — shipping documents, bills of lading, customs forms
* **Retail** — receipts, inventory documents, sales data
* **Manufacturing** — supplier invoices, quality reports, shipping docs

It also handles general PDF data extraction, form processing, and custom document types for any industry.

#### \[What] What file formats does Docspire support?

Docspire accepts a wide range of file formats: PDF, images (JPG, PNG), Excel spreadsheets (XLS, XLSX), CSV, Word documents (DOCX), plain text (TXT, RTF), and EDI files. The platform handles both digital-native documents and scanned images, including low-quality scans, sideways photos, and handwritten content.

#### \[Why] Why choose Docspire over generic OCR or document processing tools?

Generic OCR tools extract text without understanding what the text means. They require rigid templates, break when document layouts change, and need separate configuration for every document variation. Docspire is different in several key ways:

* **Context-aware extraction** — The AI understands document meaning, not just character shapes. It knows that "Total Due" on an invoice is different from "Total Deposits" on a bank statement.
* **No template maintenance** — Handle layout variations automatically without creating and maintaining templates for each vendor or document source.
* **Built-in validation** — Configurable business rules catch errors automatically. Three-way matching, calculation verification, and cross-field validation happen during processing.
* **Feedback-based learning** — When extraction needs improvement, provide natural language feedback and reprocess. The system learns from corrections.
* **Pre-trained industry workflows** — Go live in minutes with workflows already tuned for mortgage, AP, and other document types. No model training required.
* **End-to-end platform** — Ingestion, classification, extraction, validation, human review, and output delivery in one system. No need to stitch together multiple tools.

#### \[What] What languages does Docspire support?

Docspire supports document processing across 40+ languages. The adaptive OCR and AI extraction models are trained to handle multilingual documents, including those with mixed-language content. This makes the platform suitable for global organizations processing documents from vendors, applicants, and partners in different regions.

#### \[What] What accuracy rate does Docspire achieve?

Docspire achieves up to 99.5% extraction accuracy across handwritten forms, scanned PDFs, and unstructured documents. Each processed document shows its AI Extraction Accuracy as a percentage confidence score, so you can see exactly how confident the system is for every document. Documents with lower accuracy scores are automatically flagged for human review.

### Getting Started

#### \[How] How do I get started with Docspire?

Getting started takes under 5 minutes. Sign up at docspire.ai, and you will be taken to the Home dashboard. From there, click "Upload Documents" to upload your first file, optionally tag it with a Business ID and document type, then click "Process" in the Document Queue. Sample documents are also available in the platform so you can test processing immediately. No model training, no implementation delays.

#### \[How] How do I upload and process a document in Docspire?

The process follows six steps:

* **Upload** — Click "Upload Documents" on the Home dashboard. Optionally enter a Business ID and select a document type (or let Docspire auto-detect). Drag and drop files or browse to select them.
* **View in Document Queue** — Uploaded documents appear with "Pending" status. Latest uploads show at the top.
* **Process** — Click the "Process" button in the Actions column. The status changes to "Running."
* **Track** — Watch processing in real time on the Document Tracking screen. It shows fields extracted, validations passed/failed, and AI accuracy.
* **View Results** — When processing completes (status: "Success"), click "View Results" to see extracted data alongside the original document in Refined View or JSON View.
* **Download** — Export extracted data as JSON, CSV, or Excel.

#### \[What] What are the main sections of the Docspire interface?

The Docspire web client is organized into four primary sections:

* **Main** — Core document processing functions including the Home dashboard (upload, metrics) and Document Queue (view, process, review documents).
* **Automation** — Workflow configuration, integrations setup (email, webhooks, cloud storage, databases, ERPs, CRMs), and validation rules management.
* **Monitoring** — Document Tracking (real-time processing status), Analytics Dashboard (performance insights), and Process Notifications (email alerts).
* **Account** — User Management and organization settings.

#### \[What] What metrics are shown on the Docspire Home dashboard?

The Home dashboard displays four operational metrics: Documents Processed (cumulative count of successfully processed documents), AI Accuracy Rate (average AI extraction accuracy percentage, updated per batch), Active Workflows (number of currently running workflows), and Integrations Connected (number of active integrations). These give you an at-a-glance view of your processing operations.

#### \[How] How can I request a custom demo of Docspire?

From the Home dashboard, click the "Request for Custom Demo" button. You will need to provide your contact details (name, company, email). You can also optionally specify your primary use case, expected monthly document volume, current systems you want to integrate with, specific requirements, and even upload sample documents. The Docspire team will then deliver a personalized demonstration tailored to your actual documents and business needs.

### AI Extraction & Processing

#### \[What] What is intelligent document processing (IDP)?

Intelligent document processing (IDP) is the modern approach to converting unstructured and semi-structured documents into structured, actionable data. It uses artificial intelligence to understand documents contextually (regardless of format), handle layout variations automatically, process multiple document types with minimal configuration, learn from user feedback, and integrate seamlessly with existing business systems. This replaces the older approaches of manual data entry and rigid template-based OCR.

#### \[How] How does Docspire's AI extraction work?

Docspire's processing pipeline runs through several stages:

* **Ingestion** — The document is received (via upload, email, webhook, or cloud storage).
* **OCR Extraction** — Optical Character Recognition extracts text and layout data from the document, whether it is a digital PDF or a scanned image.
* **AI Processing** — Machine learning models classify the document type and extract structured data. The AI understands context, identifying fields like vendor name, invoice total, line items, applicant information, and more based on the document type.
* **Validation** — Extracted data is checked against configured business rules, including format checks, range validations, cross-field calculations, and required field enforcement.
* **Human Review** — Documents with low confidence scores or failed validations are routed for manual verification.
* **Output Generation** — Validated data is formatted and exported to the target system in JSON, CSV, Excel, or sent directly via webhook or database integration.

#### \[What] What is auto-classification and how does it work?

Docspire's AI automatically determines the document type (invoice, purchase order, bank statement, ID, pay slip, mortgage application, receipt, credit report, etc.) when you upload a document without specifying a type. This auto-classification uses AI-based analysis with confidence scoring. You can also manually select a document type from the dropdown during upload if you prefer. Supported classification types include Bank Statement, Credit Report, ID, Invoice, Mortgage Application, Pay Slip, Purchase Order, Receipt, and a configurable "Other" category for custom documents.

#### \[What] What data can Docspire extract from documents?

Docspire extracts header fields (like totals, dates, vendor names, reference numbers), line-item tables (with quantities, descriptions, amounts, unit prices), and complex nested data structures. The specific fields depend on the document type. For example:

* **Invoices** — vendor name, address, tax ID, invoice number, date, due date, line items, tax details, totals, payment terms
* **Purchase Orders** — vendor name, PO number, quantities, unit prices, total prices
* **Bank Statements** — transaction debits, credits, balances, total deposits, total withdrawals
* **Pay Slips** — gross pay, net pay, deductions, pay period
* **Mortgage Applications** — applicant info, employment history, income verification, property details, debt-to-income ratios, loan-to-value ratios
* **IDs** — date of birth, expiry date, document number

Extracted data is available in both a structured Refined View and raw JSON View.

#### \[How] How do I export extracted data from Docspire?

After processing, click "View Results" for a document, then use the download dropdown in the top-right corner. You can export as JSON (for API integration and development), CSV (for spreadsheet applications), or Excel (formatted columns). Alternatively, you can configure webhook output integrations to automatically send processed data to external systems like ERPs, CRMs, or databases.

#### \[What] What is feedback-based reprocessing?

Feedback-based reprocessing lets you improve extraction results by providing natural language instructions. When a document's initial extraction is not accurate enough, click "Reprocess" in the Document Queue, optionally change the document type, and enter feedback describing what needs to be fixed (up to 500 characters). The AI uses your feedback to re-extract data more accurately. This is useful when specific fields are missed, data formatting is wrong, table columns are misaligned, or additional fields need to be captured.

#### \[How] How do I write effective feedback for reprocessing?

Effective feedback is simple and specific. Use these patterns:

* **Tell it where to look:** "Invoice number is in the top-right corner."
* **Clarify the right column:** "Unit price is under the 'Rate' column."
* **Specify the format:** "Format dates as YYYY-MM-DD."
* **Fix quality issues:** "Remove extra spaces from all field values."
* **Combine instructions:** "Quantity is in whole numbers. Price is under 'Unit Cost' column. Remove currency symbols from amounts."

You do not need technical jargon. Plain language instructions work best.

#### \[What] What is the difference between Refined View and JSON View?

**Refined View** shows extracted data in a structured, business-friendly format with header fields at the top and table data in a grid. The original document is displayed side by side for comparison. This view is best for reviewing and verifying extraction results.

**JSON View** shows the raw extracted data in JSON format. This is useful for reviewing exact data structures, copying data for use in other applications, and verifying field names for API integration.

#### \[Why] Why does Docspire not require template creation or model training?

Traditional document processing tools require you to create rigid templates for every document layout or train ML models with hundreds of labeled samples. Docspire uses pre-trained, context-aware AI models that already understand common document structures across industries. The platform's adaptive OCR and AI extraction handle layout variations automatically. When adjustments are needed, you provide natural language feedback rather than building new templates. This is what makes it possible to go live in minutes instead of weeks.

### Document Management

#### \[What] What is the Document Queue?

The Document Queue is the central repository displaying all documents in your Docspire account. This includes sample files, manually uploaded documents, and files received through integrations. It provides a unified interface to view document metadata, check processing status, initiate actions (process, review, view results, reprocess), and manage documents through filtering, searching, and bulk operations. Each document shows its file name, type, status, source, AI extraction accuracy, validation score, and available actions.

#### \[What] What are the document statuses in Docspire?

Documents progress through these statuses:

* **Pending** — Uploaded and waiting to be processed
* **Running** — Currently being processed by the AI
* **Success** — Processing completed successfully, extracted data is available
* **In Review** — Requires human verification due to validation failures or low confidence scores
* **Error** — Processing failed due to technical issues or unsupported format
* **Rejected** — Rejected with notes through the human review process

#### \[What] What is Document Tracking in Docspire?

Document Tracking lets you monitor the complete lifecycle of any document from ingestion to final output delivery. It displays real-time status updates for each processing stage (Ingestion, OCR Extraction, AI Processing, Validation, Human Review, Output Generation), with timestamps and duration for each stage. You can search documents by ID, name, business ID, status, or date. It also provides a complete audit trail of all system and user actions for compliance and troubleshooting.

#### \[How] How does the human review process work?

When a document is flagged for review (due to low AI confidence or failed validation), it enters "In Review" status. Reviewers can access the Review interface, which shows the extracted data alongside the original document. Error fields are highlighted with validation messages. Reviewers can edit data inline in both Refined View and JSON View. After reviewing, they can either Approve (accept the data with or without edits, completing processing with Success status) or Reject (reject with notes, setting the document to Rejected status). Validation rules can also assign specific team members as reviewers for specific types of issues.

#### \[How] How do I search and filter documents in the Document Queue?

The Document Queue provides three filtering options: a Document Type filter (select specific types from the "All Types" dropdown), a Priority filter (filter by processing priority levels), and a Search Bar (find documents by name, ID, or partial text). You can also customize which columns are visible by clicking the "Columns" button and toggling fields on or off. For bulk operations, select multiple documents using checkboxes to delete them in one go.

#### \[What] What is a Business ID and why should I use it?

A Business ID is an optional internal reference tag (e.g., an applicant ID, vendor ID, or project number) you can assign to documents during upload. It allows you to associate documents with your internal tracking systems. Business IDs are alphanumeric, can include underscores and hyphens, and have a maximum of 50 characters. Using Business IDs makes it easier to search for and group related documents later. When sending documents via webhook, you can include the BusinessId field in the form data.

### Validation Rules

#### \[What] What are validation rules in Docspire?

Validation rules are configurable quality control checks that run automatically on extracted data after processing. They let you enforce business policies, ensure data accuracy, and route documents requiring attention to the right people. For example, you can flag invoices over $10,000 for manager approval, verify that line totals match quantity times unit price, ensure required fields are not empty, check that ID documents have not expired, or validate that calculated totals match extracted values.

#### \[How] How do I create a validation rule?

Navigate to Automation > Validation Rules. Then follow these steps:

* Enter a descriptive **Rule Name** (e.g., "ExpenseApproval" or "NullVendor")
* Select **Type** — Document (for individual documents) or Workflow (for workflow-processed documents)
* Select the **Document Type** (Invoice, Purchase Order, Bank Statement, etc.)
* Choose the **Field to Validate** from the dropdown (fields are type-specific)
* Set the **Action On Failure** — "Flag for Review" (assigns to a reviewer) or "Error" (stops processing)
* If flagging for review, assign a **Reviewer** from your team
* Write the **Validation Expression** — a condition that must be true for the document to pass (e.g., `total < 10000` or `IsNotNull(vendorName) = true`)
* Enter a **Show Message** that explains the failure
* Click **Add Rule**

#### \[What] What validation expressions does Docspire support?

Validation expressions are conditions that evaluate to true when data is valid. Supported patterns include:

* **Null checks:** `IsNotNull(vendorName) = true`
* **Threshold comparisons:** `total < 10000`
* **Calculation verification:** `Abs(quantity * unitPrice - totalPrice) < 0.1`
* **Positive value checks:** `amount > 0`
* **Date comparisons:** `expiry_date > current_date`
* **Subtotal verification:** `Abs(cal_subtotal - subtotal) < 0.2`

The `Abs()` function is recommended for numeric comparisons to handle rounding differences.

#### \[What] What happens when a validation rule fails?

There are two possible actions when a validation rule fails:

**Flag for Review** — The document is marked as "In Review" and assigned to the designated team member. Processing continues, but the document requires human attention before data is finalized. The assigned reviewer gets a notification.

**Error** — Document processing stops entirely. The error message you configured is displayed, and the document cannot proceed until the issue is resolved (typically by reprocessing with feedback).

#### \[How] How do I enable or disable a validation rule without deleting it?

In the Current Validation Rules table, each rule has a Status toggle in the Status column. Click the toggle to switch between active (green) and inactive (gray). An inactive rule is preserved with its configuration but will not be applied during document processing. This lets you temporarily pause rules without losing their setup.

### Integrations

#### \[What] What integrations does Docspire support?

Docspire supports a wide range of integration types:

* **Email Integrations** — Connect your email inbox to automatically receive and process document attachments via IMAP
* **Cloud Storage** — Monitor folders in SharePoint, Google Drive, OneDrive, S3, or Azure Blob for new documents
* **API Upload** — Receive documents via REST API
* **Database** — Connect to MySQL, PostgreSQL, SQL Server, ODBC, Redshift, and more
* **CRM Systems** — Sync data with Salesforce, Zoho, Dynamics CRM, HubSpot, IBM suite
* **ERP Systems** — Sync data with SAP suite, Oracle, QuickBooks, NetSuite, JD Edwards, IBM suite
* **Webhooks** — Both input (receive documents from external apps) and output (send processed data to custom endpoints)

#### \[How] How do I set up email integration for automatic document ingestion?

Navigate to Automation > Integrations, find the Email Integrations card, and click Configure. Choose between Basic Authentication (username/password) or Modern Authentication (OAuth 2.0 for Microsoft 365 and Google Workspace). Enter your IMAP server URL, port (typically 993), credentials, and select the email folder to monitor (e.g., INBOX). Optionally set up email filters to only process messages from specific senders or with specific subject keywords. Enable "Mark Email As Read" to prevent duplicate processing. Common server settings: Gmail uses imap.gmail.com:993, Outlook uses outlook.office365.com:993.

#### \[What] What are webhook integrations and how do they work?

Webhooks provide a flexible API-based approach for integrating with external systems. There are two types:

**Webhook Input** — External systems send documents to Docspire via HTTP POST requests. When you configure webhook input, Docspire generates a unique endpoint URL and secret key. External systems include the secret key in the `X-Docspire-Secret` header and send documents as `multipart/form-data`.

**Webhook Output** — Docspire sends processed data to your external systems automatically after document processing completes. You configure the target URL, HTTP method, authentication (supports No Auth, NTLM, API Key, Basic Auth, OAuth 2.0, Bearer Token, and AWS Signature), and optional custom headers. You can set up multiple output webhooks to deliver data to different destinations simultaneously.

#### \[How] How do I send documents to Docspire via the webhook API?

After configuring Webhook Input (Automation > Integrations > Webhooks > Webhook Input tab > Save Configuration), you will receive a Webhook URL and Secret Key. Send documents via HTTP POST with `Content-Type: multipart/form-data`, the `X-Docspire-Secret` header containing your secret key, and the following form fields:

* **files** (required) — The document file(s) to process
* **DocumentType** (optional) — The document type
* **AllowedDocumentTypes** (optional) — Comma-separated list of allowed types for this request
* **BusinessId** (optional) — Your internal reference identifier

A ready-to-use cURL example is provided in the configuration dialog for quick testing. You can also use Postman to test the integration.

#### \[What] What authentication methods does Docspire's webhook output support?

Webhook output supports seven authentication methods: No Authentication, NTLM (Windows-based systems), API Key, Basic Authentication (Base64-encoded username/password), OAuth 2.0 (token-based for modern APIs), Bearer Token (JWT or opaque tokens), and AWS Signature (Version 4 for Amazon Web Services endpoints). You can also add custom HTTP headers to every outgoing request.

#### \[How] How do I connect Docspire to QuickBooks?

You can connect Docspire to QuickBooks using the Webhook Output integration. Configure a webhook output with your QuickBooks API endpoint URL, select the appropriate HTTP method (typically POST), set up OAuth 2.0 authentication with your QuickBooks credentials, and add any required custom headers. Once configured, processed invoice data from Docspire is automatically sent to QuickBooks. Docspire provides a dedicated guide for the QuickBooks webhook output integration in its documentation.

### Notifications & Monitoring

#### \[What] What notification events can I configure in Docspire?

Docspire supports automated email notifications for seven event types: Document Processing Complete, Document Processing Failed, Validation Failure, Human Review Required, Workflow Succeeded, Workflow Failed, and Integration Established. Each recipient can be subscribed to specific events and configured with an appropriate notification frequency. The Recent Notifications section also logs all triggered notifications with details, timestamps, and severity levels.

#### \[How] How do I set up process notifications?

Navigate to Monitoring > Process Notification. Click "+ Add Recipient" to add a new notification recipient. Enter the email address, select the notification frequency (e.g., Immediate for real-time alerts), and check the boxes for which events should trigger notifications (Document Processing Complete, Failed, Validation Failure, Human Review Required, Workflow Succeeded/Failed, Integration Established). Click the green checkmark to save. You can add multiple recipients with different event subscriptions. Each recipient's configuration can be edited or deleted later.

#### \[What] What analytics does Docspire provide?

Docspire's Analytics Dashboard provides use-case-specific insights for invoice processing (AP Automation) and mortgage application review. Available metrics include processing volume trends, accuracy rates over time, validation pass/fail ratios, and vendor-specific or application-specific performance breakdowns. The analytics help identify bottlenecks, track improvements, and optimize processing efficiency.

### Mortgage Processing

#### \[How] How does Docspire automate mortgage document processing?

Docspire accelerates mortgage origination by processing the full range of borrower documents: applications (including 1003 forms), pay stubs, bank statements, tax returns, ID documents, property appraisals, insurance policies, and title documents. The platform automatically classifies each document, extracts applicant information, employment history, income and asset details, property details, existing debts, and insurance coverage. It then cross-references data across documents, calculates debt-to-income ratios, flags missing or expired documents, flags potentially fraudulent documents, and generates compliance reports. This reduces application processing from days to hours while improving accuracy in risk assessment.

#### \[What] What mortgage-specific fields does Docspire extract?

For mortgage applications, Docspire extracts and validates fields including loan-to-value ratio, disposable income, property value, applicant information, employment history, credit scores, account statuses, payment history, and more. Validation rules can be configured to check these fields automatically. For example, you could validate that the loan-to-value ratio is within acceptable limits or that ID documents have not expired using the expression `expiry_date > current_date`.

#### \[Why] Why is automated mortgage document processing better than manual review?

Manual mortgage document review typically takes 40+ hours per application and involves multiple people touching each file. Automation with Docspire accelerates loan approvals by 60%, reduces compliance risk with automated KYC verification, cuts operational costs by up to $500K annually per 100 loans processed, ensures regulatory compliance through consistent rule application, enhances borrower experience with faster decisions, and reduces operational costs by up to 60%. The platform handles the routine processing while humans focus only on exceptions and final approvals.

### Accounts Payable Automation

#### \[How] How does Docspire automate accounts payable workflows?

Docspire automates the full invoice-to-payment cycle. Invoices arrive via email, portal, or scan. The AI automatically classifies them, extracts vendor information (name, address, tax ID), invoice details (number, date, due date), all line items (descriptions, quantities, amounts), payment terms, bank details, and tax breakdowns. The system can perform three-way matching (comparing invoice against purchase order and receipt/delivery note), validate amounts and calculations, route for approval based on amount or department, and export directly to your accounting system or ERP. Early adopters report 80% reduction in processing time and elimination of manual data entry errors.

#### \[What] What is three-way matching and does Docspire support it?

Three-way matching is a standard AP control that compares three documents: the vendor's invoice, the original purchase order, and the goods receipt or delivery note. The goal is to verify that what was ordered, what was received, and what is being billed all match. Docspire supports this through its validation pipeline, cross-source validation capabilities, and configurable business rules. You can set up validation rules that flag discrepancies between invoiced amounts and purchase order values, enabling automated exception handling for mismatches.

#### \[Why] Why should AP teams automate invoice processing with Docspire?

Automating AP with Docspire delivers measurable business outcomes: 80% reduction in processing time, elimination of manual data entry errors, improved vendor relationships through faster payments, better cash flow management through early payment discount capture, enhanced compliance and audit readiness through complete processing trails, and the ability to track spending patterns for optimal procurement planning. Your team handles only exceptions, not routine data entry.

#### \[How] How does Docspire handle invoices from different vendors with different layouts?

Docspire's context-aware AI understands invoice structure regardless of layout. You do not need separate templates for each vendor. The AI recognizes common fields (vendor name, total, line items) across varying layouts automatically. When specific vendor formats need adjustments, you use the feedback-based reprocessing feature. For example, if a vendor labels their total as "Amount Due" rather than "Total," you can provide feedback like "Total is the final 'Amount Due' at the bottom, not the subtotal." The AI applies this learning to improve extraction accuracy.

### Security, Compliance & Fraud Detection

#### \[What] What fraud detection capabilities does Docspire offer?

Docspire employs multi-layered fraud detection that analyzes documents across multiple dimensions simultaneously. The platform examines metadata, font characteristics, and invisible micropatterns that fraudsters often overlook, including file creation timestamps, software signatures, and embedded properties that reveal a document's true origin and history. The system uses real-time anomaly detection powered by machine learning, multi-layer authentication, and cross-document validation to catch forged, altered, or inconsistent documents. This is particularly critical for mortgage processing (detecting fraudulent income statements, fabricated bank statements) and AP workflows (catching altered invoices, duplicate submissions).

#### \[How] How does Docspire ensure audit compliance?

Docspire maintains a complete audit trail for every document. The Document Tracking feature logs each processing stage with timestamps, durations, and status indicators. All system and user actions are recorded chronologically, including who reviewed a document, what changes were made, and when approvals or rejections occurred. This comprehensive traceability supports regulatory requirements in mortgage lending, financial services, and other compliance-sensitive industries. The validation rules engine ensures consistent rule application across all documents, eliminating human inconsistency.

### User Management & Team Collaboration

#### \[How] How do I manage users and roles in Docspire?

Navigate to Account > User Management. The page shows all users in your organization with their email, active status, and role (ADMIN or USER). Admins can invite new users, manage permissions, and deactivate accounts. The interface includes a Users tab (showing active members) and an Invited tab (showing pending invitations). You can search for users by email and export the user list to Excel. When setting up validation rules, you can assign specific team members as reviewers for documents that fail validation.

#### \[How] How do I invite new users to my Docspire organization?

From the User Management page (Account > User Management), you can invite new users by email. Invited users appear in the "Invited" tab with a pending status until they accept the invitation and create their account. Once active, you can assign them ADMIN or USER roles and designate them as reviewers for specific validation rules so they receive flagged documents for review.

### Pricing & Deployment

#### \[What] What pricing plans does Docspire offer?

Docspire offers tiered pricing that scales with your needs. There is a free tier for testing (50 pages/month), a plan at $99/month for power users, a $1,249/month plan for small businesses, and custom enterprise pricing for high-volume operations. The platform also offers a 14-day free trial so teams can connect their systems and begin processing documents immediately. There is $0 setup cost and no long implementation timelines.

#### \[How] How quickly can I deploy Docspire?

Docspire is designed for rapid deployment. You can go live in under 5 minutes. Sign up, upload documents, and start processing immediately with pre-trained industry workflows. No model training is required. No implementation delays. The platform uses adaptive OCR and pre-trained AI models, so there is no need for labeling training data or waiting for model convergence. For integrations (email, webhooks, cloud storage), setup is done through the UI with point-and-click configuration. Enterprise deployments with custom integrations can be set up with the Docspire team.

#### \[What] What is Docspire's relationship with Astera?

Docspire is an AI-powered document intelligence platform backed by Astera. Astera is a data management and integration company with decades of experience serving financial institutions, insurance companies, and enterprises. Docspire was purpose-built by Astera's AI team to solve document processing challenges in mortgage lending, accounts payable, and other document-intensive industries, with advisory from Astera's CPTO Jay Mishra and CTO Mike O'Quinn who have led enterprise data projects for organizations like Wells Fargo, Raymond James, and Kaiser.

#### \[How] How does Docspire scale for high document volumes?

Docspire's architecture is flexible and scalable, designed to handle anywhere from a few hundred files to millions of pages. The platform supports bulk document uploads, automated ingestion through email and webhook integrations, and parallel processing. Whether you process 50 pages a month or millions, the infrastructure scales with your needs. Enterprise plans include custom configurations for high-volume processing environments.

### Technical & Troubleshooting

#### \[How] How do I improve extraction accuracy for a specific document?

Several approaches can improve accuracy:

* **Use clear, high-quality scans** — Better image quality directly improves extraction accuracy.
* **Specify the document type** — Selecting the correct type during upload helps the AI extract the right fields.
* **Use feedback-based reprocessing** — Provide natural language feedback about what went wrong (e.g., "Total amount is in the last row of the table") and reprocess.
* **Check the accuracy score** — Documents with lower AI Extraction Accuracy percentages may need manual review or more specific feedback.
* **Process sample documents first** — Test with a few documents before running bulk processing to identify any adjustments needed.

#### \[Why] Why is my webhook input returning a 401 Unauthorized error?

A 401 error means the `X-Docspire-Secret` header in your request does not match the Secret Key configured in your webhook input. Verify that you copied the secret key exactly (case-sensitive) and that it is included in the header. If you recently regenerated your credentials, all connected systems need to be updated with the new secret key. Also ensure the Content-Type is set to `multipart/form-data` and the files field contains valid document files.

#### \[Why] Why are my email-ingested documents not appearing in the Document Queue?

Check these items: verify the IMAP server URL and port are correct, ensure IMAP is enabled in your email account settings, confirm TLS/SSL is enabled, check that the Email Folder name matches your actual mailbox folder (e.g., "INBOX"), verify that email filters (From Filter, Subject Filter) are not too restrictive, and ensure the "Read Unread Only" setting is configured appropriately. If using Gmail, make sure you are using an app-specific password rather than your regular account password.

#### \[How] How do I prevent duplicate document processing from email?

Enable the "Mark Email As Read" option in your email integration configuration. This marks emails as read after processing their attachments, preventing them from being picked up again. You can also enable "Read Unread Only" to ensure the system only processes new, unread emails. Using both settings together provides robust duplicate prevention.

#### \[How] How do I rotate webhook credentials for security?

In the Webhook Input configuration dialog, click the "Regenerate Credentials" button. This generates a new Secret Key and Webhook URL. Important: regenerating credentials immediately invalidates the existing ones, so you must update all connected systems with the new credentials before they can send documents again. It is best practice to rotate credentials periodically and to update connected systems promptly.

#### \[What] What should I do if a document's processing status shows "Error"?

An "Error" status means processing failed, typically due to a technical issue with the document file or an unsupported format. Check the Document Tracking view for detailed error logs. Common fixes include ensuring the file is not corrupted, verifying it is a supported format (PDF, JPG, PNG, DOCX, XLS, XLSX, CSV, TXT, RTF, EDI), re-scanning at higher quality if it is a scanned document, or trying a different file format. If the error was caused by a validation rule set to "Error" action, review the validation message and either adjust the rule or correct the source document.

***

*Docspire is an AI-powered document intelligence platform by Astera. Visit* [*docspire.ai*](https://docspire.ai/) *for more.*