Your documents, read by AI. Your follow-up, on autopilot.
DOXBOX classifies every upload, pulls out the data inside it, and turns that data into scheduled reports, alerts and webhooks. Every step stays governed, audited and on cloud storage you control.
Classified as
Invoice
Confidence
94%
Extracted fields
Invoice no.
INV-2044
Vendor
Acme Corp
Total
$12,450
Due date
14 Sep 2026
Automation ran: “Unpaid invoices over $10k”
Report sent to finance@acme.com as PDF + XLSX
Illustrative: upload, classify, extract, automate.
22
document types recognised automatically
3
report formats: PDF, XLSX, CSV
3
bring-your-own storage backends
Every
action recorded in the audit trail
The AI Pipeline
From a raw upload to structured, searchable data. Automatically.
No templates to build and no manual data entry. DOXBOX runs every file through OCR, classification and extraction the moment it lands.
1. Read
Multi-language OCR turns scans, photos and Office files into searchable text you can also download.
2. Classify
Each document is identified with a confidence score, from invoices and contracts to passports and bills of lading.
3. Extract
Type-aware extractors pull the fields that matter: amounts, dates, parties, ID numbers, line items.
4. Verify
Low-confidence fields are flagged for a human to review and correct, so you can trust what you automate.
Recognised out of the box
Describe the report you want. DOXBOX builds the rule.
Turn extracted data into action. Set a condition on any extracted field, choose a schedule or an event, and DOXBOX generates the report and emails it, with an optional webhook to your other systems. Or just say what you need in plain English and review the draft it writes for you.
auto_awesomeDescribe it in plain English
“Every Monday at 9am, email finance a PDF of unpaid invoices over $10,000 due this week.”
Draft: Unpaid invoices over $10k
Awaiting your reviewIllustrative. AI drafting is an optional add-on your admin enables per organisation.
Webhooks for your other systems
Push document and automation events to your ERP, chat or workflow tools, with signed payloads and a delivery log to see what was sent and what failed.
Run history you can trust
Every execution is recorded: success, no matches or failure, so nothing quietly stops working. Failures surface on the dashboard.
Time-zone aware
Each automation runs on its own time zone, with daylight-saving handled for you, and shows run times in that zone.
Search that understands
Ask the way you'd ask a colleague. Filter by what's actually inside.
Semantic search finds the right document even when the exact words never appear. Structured field search filters on extracted values, such as every invoice from one vendor, over an amount, due this month.
Illustrative results
Dashboard & Insights
See what needs attention before it becomes a problem.
A live dashboard surfaces expiring retention, failed OCR, shares about to lapse, low-confidence extractions and broken automations, and every row links straight to the documents behind it.
notifications_activeNeeds attention
categoryClassification confidence
Weakest types link to their documents so you can fix them fast.
trending_upExtracted value trend
$184,320
sum of invoice total, last 30 days
grid_onAudit activity heatmap
Day × hourcloudStorage & pipeline
1
Queued
14s
Avg. job
0
Failed
Illustrative dashboard with sample data.
Built For Real Workflows
One platform, wherever paperwork piles up.
Finance & AP
Capture invoices, receipts and bank statements. Get a weekly digest of what's unpaid and due.
Legal & Contracts
Compare revisions line by line, put matters on legal hold, and be alerted before contracts expire.
Trade & Logistics
Bills of lading, packing lists and certificates of origin, read and searchable as they arrive.
HR & Onboarding
Résumés, salary slips, passports, national IDs and visas, with fields extracted and access restricted.
Case-Based Governance
Every case, governed from upload to archive.
A Case groups documents into a matter with its own approval workflow, permissions and activity history.
Approve, reject, archive or restore, gated to admins and logged automatically.
Granular permissions
Owner, visibility (Organisation / Restricted / Private) and shared users: real access controls per case, not just per document.
Colour-coded activity timeline
Every workflow change, upload, share and security event is logged and colour-coded. Drag documents straight into a case, with no separate upload step.
Document Version Comparison
Every revision, line by line.
Compare any two versions in File Preview mode, side by side, or drop into OCR Diff mode for a scroll-synced, line-level diff. Additions and removals are colour-coded, and nothing is silently overwritten.
MSA_Contract_2026.pdf - V2
MSA_Contract_2026.pdf - V3
Illustrative diff view
Also Included
Smart Auto-Tagging
Retention Policies
Legal Hold
Compliance & Audit Reports
Multi-Organisation Support
Role-Based Access
Two-Factor Auth
Light & Dark Themes
Questions, Answered
What people ask before they book a demo.
Where are my documents stored?expand_more
In storage your organisation controls. Point DOXBOX at your own AWS S3, DigitalOcean Spaces or on-premise MinIO instead of a shared bucket.
Does the AI act on its own?expand_more
Not without you. Low-confidence extracted fields are flagged for human review, and automations drafted from plain English are pre-filled for you to check. Nothing is saved until a person approves it.
What kinds of documents does it understand?expand_more
Twenty-two document types today, including invoices, purchase orders, contracts, bank statements, bills of lading, passports, national IDs and résumés. Anything else is still OCR'd, tagged and searchable.
Can it connect to the rest of our systems?expand_more
Yes. Outbound webhooks push document and automation events to your own tools, and every delivery is logged so you can see what succeeded and what didn't.
How do we get started?expand_more
Request a demo and we'll walk through DOXBOX live, ideally with a few of your own documents.
Security & Compliance
Governance isn't bolted on. It's the foundation.
Your documents are encrypted before they're stored.
Built for banks, insurers and financial institutions: every file your organisation stores is encrypted with AES-256-GCM, so the stored object is unreadable ciphertext without the keys.
-
enhanced_encryption
AES-256-GCM
Authenticated encryption on every stored object. -
key
A unique key per file
Each key is wrapped by a managed master key, never stored in the clear. -
description
Covers every copy
Originals, previews, thumbnails and generated reports. -
sync_lock
Existing files included
Encrypt your current archive with one click. New uploads are protected automatically. -
travel_explore
No trade-off on search or OCR
Preview, download, OCR, search and AI extraction keep working on encrypted documents. Downloads are served only through authenticated links, never raw storage URLs.
Per-organisation storage isolation
Role-based + two-factor access control
Full audit trail on every action
Retention policies & legal hold
See DOXBOX turn documents into decisions.
AI classification, structured extraction, plain-English automations and governance on infrastructure you control. Walk through it live with your own documents.