Amazon Textract AnalyzeExpense
aws.amazon.com
What it is
Amazon Textract AnalyzeExpense extracts financial text and data from invoices and receipts through an API or web service. Results can include grouped line items and summary fields such as purchased items, prices, receipt headers, and vendor names. The API can normalize field names and column headers to help downstream applications compare documents. It can also identify vendors from logos and extract items, quantities, and prices when those details have no labels or column headers. Supply documents as image bytes or an Amazon S3 object; AWS advises PDF, JPG, or PNG files, and invoice and receipt processing is English only. AnalyzeExpense processes synchronously. Asynchronous invoice and receipt analysis uses StartExpenseAnalysis and GetExpenseAnalysis. The listed rate is $0.01 per page for the first 1 million pages per month in US West (Oregon), then $0.008 per page; custom high-volume pricing is available. New AWS customers can use the Free Tier for three months, up to 100 pages per month. AWS says inputs may be used to improve Textract and related AI technologies, with an Organizations opt-out policy.
Spec sheet
- What it does
- Amazon Textract is a managed machine-learning service that extracts printed text, handwriting, forms, tables and structured data from scanned documents.Read 15 Oct 2026 · docs.aws.amazon.com
- Purpose
- AnalyzeExpense processes invoices and receipts to extract financially related text and data.Read 2 Oct 2026 · docs.aws.amazon.com
- Integrations
- AWS lists robotic-process-automation partners Blue Prism, Ripcord and UiPath, plus service and software partners including Accenture, Deloitte, Box and PwC.Read 15 Oct 2026 · aws.amazon.com
- Compliance
- AWS states that Textract is covered by HIPAA, SOC, ISO and PCI compliance programs, subject to applicable conditions.Read 15 Oct 2026 · docs.aws.amazon.com
- Access controls and monitoring
- Textract security documentation covers IAM, CloudTrail logging and monitoring, data protection, resilience, and VPC endpoints.Read 2 Oct 2026 · docs.aws.amazon.com
- API surface
- The public API includes AnalyzeExpense, StartExpenseAnalysis, GetExpenseAnalysis, AnalyzeDocument, AnalyzeID, AnalyzeLending, adapter-management and text-detection operations.Read 15 Oct 2026 · docs.aws.amazon.com
- Asynchronous expense processing
- StartExpenseAnalysis begins asynchronous invoice or receipt processing and GetExpenseAnalysis retrieves results after completion.Read 15 Oct 2026 · docs.aws.amazon.com
- CLI and console
- Textract can be used through the AWS Management Console, AWS CLI and API.Read 15 Oct 2026 · aws.amazon.com
- Custom Queries
- Custom Queries adapters can be trained from as few as ten annotated sample documents in the AWS Console.Read 15 Oct 2026 · aws.amazon.com
- Deployment model
- Textract is delivered as an AWS-hosted service accessed through regional HTTPS endpoints; AWS does not provide a self-hosted Textract server.Read 15 Oct 2026 · docs.aws.amazon.com
- Documentation depth
- AWS documentation includes a conceptual Developer Guide, detailed API Reference, getting-started procedures, quotas, IAM, encryption and code examples.Read 15 Oct 2026 · docs.aws.amazon.com
- Expense analysis
- AnalyzeExpense synchronously analyzes invoices and receipts and returns line-item groups and summary fields in JSON.Read 15 Oct 2026 · docs.aws.amazon.com
- Extracted data
- Results include line item groups and summary fields, such as purchased items, prices, receipt headers, and vendor names.Read 2 Oct 2026 · docs.aws.amazon.com
- File and size limits
- Supported formats are JPEG, PNG, PDF and TIFF; synchronous requests allow one-page PDF/TIFF files up to 10 MB, while asynchronous PDF/TIFF files can be up to 500 MB and 3,000 pages.Read 15 Oct 2026 · docs.aws.amazon.com
- Forms
- The service extracts related key-value pairs from document forms while preserving their context.Read 15 Oct 2026 · aws.amazon.com
- Identity documents
- AnalyzeID extracts information such as expiry date, birth date, names and addresses from supported identity documents without templates.Read 15 Oct 2026 · aws.amazon.com
- Independent review tradeoff
- OCR Vendors rates Textract 7.4/10, praising low per-page cost and AWS integrations while citing AWS lock-in, required expertise, extra-cost support and fewer prebuilt processors than Google.Read 15 Oct 2026 · ocrvendors.com
- Independent test limitation
- An independent review found Textract’s unstructured text extraction accurate and inexpensive but reported poor forms accuracy and slower asynchronous PDF workflows requiring S3.Read 15 Oct 2026 · crosstab.io
- Input
- A document can be supplied as image bytes or as an Amazon S3 object.Read 2 Oct 2026 · docs.aws.amazon.com
- Language support limit
- Text detection supports English, French, German, Italian, Portuguese and Spanish; handwriting, invoices, receipts, identity documents and Queries are English-only.Read 15 Oct 2026 · docs.aws.amazon.com
- Lending
- Analyze Lending classifies, splits and extracts information from mortgage application document packages and can summarize signature detection.Read 15 Oct 2026 · aws.amazon.com
- Licence
- Textract is a proprietary AWS cloud service rather than an open-source package with a redistributable software licence.Read 15 Oct 2026 · aws.amazon.com
- Limit: cost and AWS complexity
- G2 reviewers repeatedly complain that large-volume processing is expensive and that integration requires AWS infrastructure knowledge.Read 15 Oct 2026 · g2.com
- Limit: Cyrillic support
- A Gartner Peer Insights dislike states that Textract does not support Cyrillic characters.Read 15 Oct 2026 · gartner.com
- Limit: document layout
- Textract does not support vertically aligned text, XFA PDFs or password-protected PDFs.Read 15 Oct 2026 · docs.aws.amazon.com
- Limit: handwriting and training
- G2 reviewers report weaker handwriting accuracy and say some desired formats cannot be trained or identified reliably.Read 15 Oct 2026 · g2.com
- Limit: query volume
- Queries are limited to 15 per page synchronously and 30 per page asynchronously.Read 15 Oct 2026 · aws.amazon.com
- Maker founded
- Amazon Web Services launched in 2006, according to AWS’s official LinkedIn page.Read 15 Oct 2026 · linkedin.com
- Maker headquarters and size
- AWS’s LinkedIn page identifies Seattle, Washington as its location and shows more than 148,000 employees associated with the company page.Read 15 Oct 2026 · linkedin.com
- Maker legal name
- Amazon.com, Inc. is the legal corporate name shown in Amazon investor materials.Read 15 Oct 2026 · ir.aboutamazon.com
- Normalized fields
- The API can return normalized key names and column headers so downstream applications can compare data across invoices and receipts.Read 2 Oct 2026 · aws.amazon.com
- Notable outage
- The October 20, 2025 AWS outage disrupted many online services and AWS reported normal operations only later that day.Read 15 Oct 2026 · apnews.com
- OCR
- Textract detects printed and handwritten text, including noisy or distorted text and varied fonts.Read 15 Oct 2026 · aws.amazon.com
- Praise: accuracy and automation
- G2 reviewers repeatedly praise accurate OCR, fast extraction, straightforward API calls and reduced manual document handling.Read 15 Oct 2026 · g2.com
- Praise: table processing
- Gartner reviewers praise Textract for handling complex table structures and processing large document volumes quickly.Read 15 Oct 2026 · gartner.com
- Privacy and encryption
- AWS says inputs may be used to maintain and improve Textract and related AI technologies; it describes encryption at rest and in transit and offers an Organizations opt-out policy.Read 2 Oct 2026 · aws.amazon.com
- Private connectivity
- Textract supports VPC endpoints through AWS PrivateLink for API calls from within a VPC without using the public internet.Read 2 Oct 2026 · aws.amazon.com
- Processing modes
- AnalyzeExpense is synchronous; asynchronous invoice and receipt analysis uses StartExpenseAnalysis and GetExpenseAnalysis.Read 2 Oct 2026 · docs.aws.amazon.com
- Queries
- Users can ask natural-language questions such as a customer-name query and receive the requested value without knowing document layout.Read 15 Oct 2026 · aws.amazon.com
- Rating on G2
- 4.3 out of 5 from 27 reviews.Read 15 Oct 2026 · g2.com
- Rating on Gartner Peer Insights
- 4.4 out of 5 from 37 ratings.Read 15 Oct 2026 · gartner.com
- Rating on GetApp
- 0.0 out of 5 from 0 reviews.Read 15 Oct 2026 · getapp.com
- Rating on TrustRadius
- 9.9 out of 10 from 3 reviews.Read 15 Oct 2026 · trustradius.com
- Regional storage
- Processed content is encrypted and stored at rest in the AWS region in use, though some content may be stored in another region for service improvement unless the customer opts out.Read 2 Oct 2026 · aws.amazon.com
- SDKs
- AWS provides SDK documentation and examples for C++, Go, Java, JavaScript, .NET, PHP, Python/Boto3 and Ruby.Read 15 Oct 2026 · docs.aws.amazon.com
- Security encryption
- Input data can use Amazon S3-managed or AWS KMS server-side encryption at rest, with TLS protecting data in transit.Read 15 Oct 2026 · docs.aws.amazon.com
- Signatures
- Analyze Document can detect handwritten signatures, electronic signatures and initials and return their locations and confidence scores.Read 15 Oct 2026 · aws.amazon.com
- Supported formats and language
- AWS advises using supported PDF, JPG, or PNG files; invoice and receipt processing is English only.Read 2 Oct 2026 · aws.amazon.com
- Tables
- Textract identifies table rows, columns and cells and returns bounding geometry and confidence scores.Read 15 Oct 2026 · aws.amazon.com
- Unlabeled data
- Analyze Expense can identify a vendor from a logo and extract line item, quantity, and price even when those fields lack labels or column headers.Read 2 Oct 2026 · aws.amazon.com
Plans and prices
Strong points and limits
Strong points
- Extracts line items and summary fields from receipts and invoices
- Can return normalized field names and column headers
- Accepts image bytes or an Amazon S3 object
- Free Tier offers up to 100 pages monthly for three months
Limits
- Invoice and receipt processing is English only
- No free plan is listed
- AWS may use inputs to improve Textract unless opted out
Who it suits
AnalyzeExpense suits developers and organizations that need to extract invoice or receipt data for downstream applications. It is a fit for English-language PDF, JPG, or PNG processing through an API or web service.
Verdict
AnalyzeExpense provides structured invoice and receipt extraction, with synchronous and asynchronous processing options. Check the language, input format, per-page pricing, and data-use terms against your needs.
Where Specifiction places it
- Intelligent Document Processing Software#1 of 319.2
- AI OCR Software#1 of 309.2
- Data Mining And Extraction Software#1 of 299.2
- OCR API Software#1 of 299.3
- AI PDF Tools#2 of 309.3
- AI Document Processing Software#2 of 279.2