Launching soon
Template-driven PDF extraction for finance & operations teams
Automation you can verify.
Every step on the record.
Tablefire turns PDFs into clean, structured data — and shows its work. Define extraction rules once; documents are detected, extracted, normalized, and exported with a full audit trail.
Tablefire opens soon. When it does, you'll create an account in under a minute and start extracting — no demo call, no sales process.
See how it works·See pricing·How auditing works
1 page = 1 credit · No editing fees · 24-hour auto-deletion by default
Self-serve from day one
Sign up with an email and you're in. No procurement, no onboarding call, no minimum seats — none of what usually sits between you and trying a finance tool.
Free tier, no credit card
10 credits every month plus 20 when you sign up, so you can run real documents before deciding anything.
Subscription when you need more
Monthly or yearly plans that scale with volume. Everything is priced per page — seepricing.
Black-box automation asks for blind trust. Manual entry wastes your week.
The question that keeps finance teams up at night: "What if the system extracts something wrong and I don't notice?"
Manual entry
Slow and error-prone
Hours every week retyping numbers from PDFs into spreadsheets — and every keystroke is a chance to introduce an error you catch at reconciliation time, or never.
Black-box parsers
Fast but unverifiable
One-click automation with no visibility: you cannot see whether a number was read wrong, normalized wrong, or produced without a source. Unverifiable output is compliance risk.
Tablefire
Automated and auditable
You define the rules once, the system applies them consistently, and you can verify every step — from the pixel in the PDF to the cell in your export.
Tablefire is the middle ground between manual entry and black-box automation:automation with transparency, at every stage of the pipeline.
How Tablefire works
A template is not a saved layout — it's a structured set of rules that runs an entire pipeline: detection, extraction, normalization, and export.
Upload
Native PDF, scanned PDF, or image. OCR runs automatically when needed, extracting word-level text with positional coordinates.
Build a template
From one representative document: draw the extraction areas, define text anchors that identify this document type, and set normalization rules and export mapping.
Apply at scale
Upload a batch. Tablefire matches text anchors to select a template, then runs extraction → normalization → export for each document — with the match shown for review.
Audit & export
Review per-field confidence scores and stage-by-stage logs, correct anything flagged, then export clean CSV, Excel, or JSON — every time in the same schema.
Refine a template or its anchors once, and every future run uses the updated rules. Minutes, not hours.
See Tablefire in action
One document through the whole pipeline: detected against a template, extracted field by field, normalized, and exported — with every step recorded.
How the Tablefire pipeline works: you upload a PDF such as a bank statement, the right template is detected from text anchors, the table is extracted with a visual grid, values like dates and amounts are normalized by rule with per-field confidence scores, and the clean rows are exported to CSV, Excel, or JSON in the same schema every time.
The animation above shows the same pipeline, step by step.
The Tablefire difference
Automation with an audit trail
See exactly how every number was detected, extracted, normalized, and exported. Answer "where did this figure come from?" in seconds, not hours.
Per-document status tracking
Follow every document from upload through OCR, template detection, extraction, normalization, and final export.
Per-field confidence scores
Every OCR-detected word carries a confidence score. Low-confidence cells are flagged for review.
Stage-by-stage audit
Inspect exactly how the document progressed: which anchors matched, how the grid was applied, which headers mapped, which normalization rules ran.
Template application transparency
See which template was matched, which text anchors triggered it, how many documents succeeded, and where human attention is needed.
Editable before export
Every extracted table is fully editable before it leaves the system. You confirm — the data ships.
Traceable to the source.
Every cell is traceable to its source location in the original PDF. If a value looks wrong, you can see exactly where it came from — and fix it before export.
Everything you need to turn PDFs into clean, structured data
One platform for the whole pipeline: OCR, template-driven extraction, rule-based normalization, and audited exports.
Template Engine
A template encodes a complete pipeline for a document type: detection, extraction, normalization, and export rules. Set it once — it runs the same way, every time.
Text-Anchor Detection
Define keywords that uniquely identify a document type. Incoming documents are checked against those anchors and routed to the template that matches — no manual sorting.
Visual Extraction Editor
Draw rows, columns, and table regions directly on the PDF page. Human-guided extraction for granular control instead of black-box parsing.
OCR for Scanned PDFs & Images
Built-in OCR extracts word-level text with positional coordinates from scanned bank statements, receipts, and image-based PDFs.
Normalization Engine
Standardize extracted values by rule: dates, currencies, vendor names, account numbers, and more — consistently across every document.
Per-Field Confidence Scores
Every OCR-detected word carries a confidence score. Low-confidence cells are flagged for review.
Stage-by-Stage Audit
Inspect how each document moved through the pipeline: anchors matched, grid applied, headers mapped, rules executed, table produced.
Exports & Integration-Ready Schemas
One-click export to CSV, Excel, and JSON with template-defined columns — ready for Google Sheets, QuickBooks, and Xero workflows.
Secure by Default
Authenticated access, row-level security, 24-hour auto-deletion by default, and no model training on your data. An extraction tool, not a storage system.
Send clean data where your team already works
Exports land in integration-ready schemas — direct push connectors are on the way.
Google Sheets
Collaborative review, client-shared reconciliations, and financial models. Templates enforce consistent columns and data types, so every export lands in the same Sheet structure.
QuickBooks
Post invoices, bills, expenses, and payments into AP/AR. Templates map extracted fields — vendor, date, line items, tax, totals — to QuickBooks fields with an audit trail on every batch.
Xero
Sync invoice and statement data for bookkeeping and cash-flow tracking. Templates standardize recurring suppliers and export Xero-ready schemas with predictable headers every time.
Templates make integrations repeatable: define the rules once, and matching documents are detected, extracted, normalized, and exported in the same schema.
Built for teams that answer for the numbers
If someone downstream reconciles, reports, or audits your data, Tablefire was designed for you.
Accountants & bookkeepers
Bank statements, trial balances, general-ledger reports, client PDFs
Hours of manual retyping — and compliance risk if a number lands wrong.
SMB finance & operations
Invoices, receipts, expense reports, vendor statements
Manual entry errors that surface later as reconciliation problems.
Analysts
Operational reports, inventory lists, logistics manifests
Recurring PDFs that need to land in spreadsheets or BI tools consistently.
Audit & compliance
Any workflow where data lineage matters
Must prove exactly how each figure was extracted, transformed, and exported.
Why Tablefire is built differently
| Instead of… | Tablefire gives you |
|---|---|
| Black-box parsers | Field-level visibility, correction, and template-level audit at every stage — not just a finished file. |
| Manual entry / copy-paste in Excel | Reusable templates and confidence scoring replace document-by-document retyping. |
| Rigid OCR pipelines | Built for real-world documents: scanned, skewed, multi-page, multi-table — and improvable over time. |
| No-code automation scripts | No-code template rules with stage-by-stage logs — you can see which step failed and why. |
Planned for launch
Simple page-based pricing.
Every credit will process one PDF page, including OCR when needed. No extra charge for editing, reviewing, or exporting your data.
Build a template once, apply it to hundreds of documents — no charges for editing, re-exporting, or template creation.
Free
Built for getting started
Starter
For individuals & freelancers
Professional
PopularFor growing teams
Business
For larger operations
Enterprise
For high-volume needs
Planned pricing — final plans and prices may change before launch.
Need to extract more pages?
Contact us for a custom plan tailored to your high-volume needs.
Contact UsSet the rules once. Apply them at scale. Audit every step.
Tablefire opens soon, and getting in will take a minute: sign up, upload a PDF, build a template. Nothing to schedule and nobody to talk to first.
Questions before then? Send us a message — we read every one.
Questions before you commit?
Tell us what you'd need Tablefire to handle. We read every message and reply within 24-48 hours.
You can also reach us directly at support@tablefire.io