Document tools

Drop in a document. Get structured data back – checked, not just converted.

Start free

Not sure which tool you need?

Drop a document in and it will tell you what it is – invoice, statement, receipt, CV – and point you at the tool that reads it. Naming a document is always free: your one free extraction stays unused until you run it.

No document to hand? See it run on a sample – this does not use your free document.

Your document is read and not stored – it never reaches our file storage. Upload only documents you are entitled to share. Privacy Policy · Trust Center

The tools

Each one reads a document, pulls out the fields that matter, and checks its own work before you export.

Bank statement to Excel, CSV or JSON

Drop in a PDF or a photograph of a bank statement and get every transaction back as a spreadsheet – dates, descriptions, debits, credits and the running balance.

Also asCSVJSONAPI

Open the tool

Invoice to Excel, CSV or JSON

Drop in a supplier invoice and get the header and every line item back as structured data – with the VAT and the totals checked against each other.

Also asCSVJSONAPI

Open the tool

Receipt to Excel, CSV or JSON

Photograph a receipt or drop in a stack of them, and get one expense table back – merchant, date, VAT and total, ready to claim or reconcile.

Open the tool

CV to Excel, CSV or JSON

Turn a stack of CVs into one comparable table – candidate, contact details, qualifications, employment history and skills, in the order you want to read them.

Open the tool

Validate three months of bank statements

Three statements in, one answer out: whether they are complete, whether they belong together, what the income actually is, and what will not reconcile.

Open the tool

Every statement is laid out differently. These already know how.

A South African bank statement is not a standard format – the columns, the date style and where the balance sits change from bank to bank. Each page below is the same extraction and the same checks, tuned to one bank's layout.

FNB Standard Bank Absa Nedbank Capitec Investec Discovery Bank GoTyme Bank African Bank Bank Zero Sasfin Bidvest Bank

Not listed? The general statement tool reads it anyway

One pipeline, whatever you drop in

The document type decides which fields come out and which checks run. Everything else is the same path – which is why a new document type is a schema here rather than a new product.

The pipeline: read, classify, extract, check, export – with a failed check routed to a person for review. 1 Read 2 Classify 3 Extract 4 Check 5 Export anything unresolved goes to a person, not a guess

Converted is not the same as correct

Every converter can turn a PDF into rows. The question that decides whether you can trust the file is what happened after that.

The arithmetic is re-run

Lines are re-multiplied, totals re-added, VAT recalculated, balances walked from opening to closing figure. Not sampled – every row.

Disagreement is reported

Where a check fails, you are told which one and by how much. Nothing is adjusted to make a difference disappear.

Uncertainty is visible

A field the extractor was not sure about comes back marked, beside the page it came from, so a person corrects a reading instead of re-typing a document.

The same engine that runs the automations

These tools are the front door to the document pipeline Automize already runs in production – the one a Creditors Clerk uses to read every supplier invoice, and a Bank & Reconciliation Clerk uses to read every statement. Start by converting one document by hand; when it turns out you do it every month, the process that does it for you is the same extraction with a schedule on it.

An Automize work queue: supplier statements with priority, SLA state, and the reason the failed one failed.
The same extraction running as a worklist: what arrived, what is waiting, and why the failed one failed.

See the Digital Workers

Questions

A converter gives you the data. These give you the data and tell you whether it is right – the totals are re-added, the balances walked, the VAT recalculated, and anything that does not agree is reported instead of being quietly corrected. On a financial document that second half is the part that matters, because a number that is wrong and confident is worse than no number at all.
To run a document, yes. The tool pages are free to read and the pipeline they describe runs inside your own company on the platform, under your access control and retention settings, which is not something an anonymous upload box can offer.
PDF, scanned PDF, JPG and PNG. A PDF downloaded from the source system reads most accurately because the text is already in the file; a photograph or a scan is OCR’d first and is more likely to return fields flagged for review.
Yes – Excel, CSV or JSON. JSON is the one to ask for if the data is going into another system, because it keeps the structure a spreadsheet has to flatten: an invoice’s line items stay nested inside the invoice rather than becoming rows that have to be joined back up.
It comes back with the fields it could read and the ones it could not marked as such, beside the original page. Nothing is invented to fill a gap. A document that fails a check is surfaced with the check that failed, so the fix is a correction rather than a re-key.
Yes, and that is the point of them being here rather than in a standalone app. The same extraction runs inside a process – watching a mailbox or a folder, reading what arrives, checking it, posting it, and asking a person only about the exceptions.

Try it on your own document

Create a free Automize company and upload a document. Extraction, the checks and the exports are all in the platform – and when the same document arrives every month, the process that reads it can post it too.

Start free See pricing