PDF paperwork on autopilot

By Arjun Mehta · Updated 2026-07-05

Merging and splitting PDFs

The most common PDF chore is combining documents that arrived separately or extracting the pages you actually need from a larger file. BackBrain handles both with plain-language instructions.

"Merge all the statements in this folder into one PDF, newest first" — BackBrain finds the files, sorts them by date, and produces a single merged document. "Extract pages 3 to 7 from this contract and save them as a separate file" — it splits the specified range without touching the original. You can chain these operations: "split each chapter into its own file, then watermark each one."

OCR: reading scanned pages

Many important documents — older bank statements, government letters, signed contracts — exist as scanned images inside a PDF, with no text layer a computer can search. BackBrain runs OCR on these pages to extract the underlying text, then treats the document the same as a native-text PDF for any subsequent operation.

This matters for organisation and search: a scanned utility bill becomes a searchable, classifiable document instead of an opaque image. OCR also enables form-filling on scanned forms — BackBrain identifies the field regions, extracts the text labels, and fills the correct positions.

Filling PDF forms

Government forms, loan applications, and compliance documents often arrive as PDF forms with fillable fields. BackBrain can fill these fields from information you provide or from data it has already gathered during the same task — such as details extracted from a statement it downloaded earlier in the run.

Before submitting or saving a filled form, BackBrain presents a review summary showing each field and its value. Nothing is written to the final file until you approve the completed form.

Watermarking and annotating

Adding a "CONFIDENTIAL" watermark across a batch of documents before sharing them, stamping each page with a date, or adding a reference number to a set of invoices — these are repetitive tasks that BackBrain handles in bulk. Specify the text, position, opacity, and which files to apply it to, and it processes the entire batch headlessly.

Watermarks can be added or removed. If you receive a watermarked document you need to work from, BackBrain can strip the watermark layer and produce a clean working copy.

Extracting text and data

Rather than reading a PDF yourself to find the numbers you need, BackBrain can extract specific values and compile them. "Pull the total amount from each invoice in this folder and put them in a spreadsheet" — it reads each PDF, finds the relevant field, and writes the results to Excel without manual copying.

For statement PDFs with tabular data — transaction lists, balance summaries — it can extract the full table and output it as a structured CSV or Excel file, ready for analysis or import into accounting tools.

PDFs inside longer workflows

PDF operations rarely happen in isolation. A typical task might be: download a statement from an online portal, rename it by its contents, merge it with last month's statement, watermark both pages with "CONFIDENTIAL," and email the merged file to your accountant. BackBrain executes this as a single task — no manual steps between the stages.

Because BackBrain verifies each step before moving to the next, the email is never sent until the merge and watermark are confirmed complete and the file exists on disk. The task either finishes correctly or stops and tells you exactly where it stalled.

For the upstream step — getting statements and files correctly named and filed before processing — see how smart-organize handles an entire messy Downloads folder in one command.