Document extraction answers
Short, checkable answers to the questions people actually ask about getting data out of technical documents — each one linked to the page that covers it in full.
Questions and answers
Can AI extract data from an engineering drawing?
Yes. AxioExtract reads engineering drawings against a schema for that drawing type and returns the title block, general notes, revision history and dimensional callouts as spreadsheet rows. Complex spatial sheets such as drawings, P&IDs, schematics and isometrics are routed to a frontier vision model; every value can be checked against the highlighted region on the page before export.
How do I get a parts list out of a PDF?
Upload the PDF with the exploded view and parts list profile. The diagram and its table are read together, so item and balloon numbers stay joined to part number, description and quantity, and the reviewed bill of materials exports to Excel, Google Sheets, JSON or XML.
What file formats can be converted to structured data?
PDF, PNG, JPG/JPEG, DOCX, XLSX, CSV, DXF (AutoCAD exchange) and IFC (BIM models, up to 60 MB). Up to 12 pages are processed per file. DWG, RVT, SolidWorks, Inventor, CATIA and STEP/IGES must be exported to PDF, DXF or IFC first.
What can the extracted data be exported to?
An Excel workbook (.xlsx), tab-separated values for pasting into Google Sheets, JSON, or XML. SPIR reads additionally offer a tag register CSV download.
Are CAD and BIM files read by a vision model?
No. DXF and IFC files carry their text and properties as data, so they are parsed programmatically and no image is sent to a vision model. A lightweight model is used only to map the parsed entities onto the output schema.
Can it read scanned or photographed documents?
Yes. Pages are rasterised and read as images, so scans, screenshots and photographs of printed sheets work the same way as native PDFs.
How is a SPIR workbook handled?
A SPIR is read deterministically using label and position matching rather than a vision model, including multi-tab workbooks where only the relevant sheet is read. The tag-versus-part matrix is flattened into tag group, part and quantity triples, from which a tag register and a per-tag or consolidated bill of materials can be generated.
How is it priced?
Prepaid credits, charged per page inside each file at a rate set by document complexity: invoices and office tables 1 credit per page, engineering drawings 4, SPIR workbooks 4, exploded views 6, and P&IDs, electrical schematics and piping isometrics 8. CAD and BIM files are charged per file — 6 credits for DXF and 10 for IFC. Prices are in pounds sterling and exclude UK VAT, which is added at checkout where applicable.
How accurate is the extraction?
No extraction tool should be trusted unreviewed, which is why the workflow puts the document beside the results grid. Selecting a row highlights where the value was read, low-confidence readings are flagged, any region can be boxed and re-read, and every export reflects the reviewed values rather than the raw model output.
What happens to uploaded files?
They are stored in a private bucket scoped to your workspace with row-level security and per-user history. A stored file can be deleted from the server while retaining its history record.
Can a workspace define its own document type?
Yes. A workspace user can create a custom document type with its own fields and columns, choose whether it is read as text or visually, and then select it during extraction alongside the built-in profiles.
Is two-factor authentication available?
Yes. Authenticator-app (TOTP) two-factor authentication with backup codes, trusted devices and an emailed one-time-code fallback is available to every workspace, and workspace owners or admins can make it mandatory for all members.
Go deeper
- Engineering drawing data extraction — drawings, schematics and isometrics
- Extract a BOM from a PDF — exploded views, parts lists and consolidated BOMs
- P&ID data extraction — equipment, lines, instruments and valves
- SPIR spare parts extraction — deterministic template reading and tag registers
- DXF to Excel and IFC to Excel — CAD and BIM parsing
- PDF to Excel and the step-by-step guide
- Supported formats and security
Try it on your own document
Credits are included with every new workspace — no card required to start.
Start extracting