Research & documents

Pulling data out of PDFs and scans, and filing them

SI reads PDFs, scans and photos, extracts the fields you need into a spreadsheet, and renames and files the documents consistently.

Worth knowing: Medical records and other personal data must be handled under the privacy rules that apply to you (for example HIPAA). Use approved, secure tools only.

The problem

Information is trapped in PDFs, scans and phone photos: statements, forms, delivery notes, certificates. Someone retypes it, and the files end up in a messy downloads folder.

What SI does

SI reads each document, extracts the fields you specify (dates, names, amounts, reference numbers, line items), and writes them to a spreadsheet with a link to the source file. It renames files with a consistent pattern and moves them into the right folders, and lists documents it could not read clearly.

Examples

The same job looks different in different businesses. These are illustrations, not client results.

  • Property manager

    A year of scanned utility bills across buildings becomes a sheet of usage and cost by building and month.

  • Freight broker

    Proof-of-delivery scans are renamed by load number and the delivery dates extracted for billing.

  • Medical office

    Faxed referral forms are filed by referring practice and the key referral details logged for the scheduling team.

What it takes to set up

A source folder, the list of fields, and a naming pattern. Setup is often a day.

What a human still checks

  • Extracted values are spot-checked against the documents.
  • Unreadable or uncertain fields are flagged, not guessed.
  • Original files are kept unchanged.

Try it yourself

Paste this into your SI tool of choice and swap the [bracketed] parts for your own details. It asks SI to stop and show you before it does anything.

Copy-ready prompt
For each document in [folder], extract [fields] into [sheet] with a link to the file. Rename each file as [pattern] and move it to [folder rules]. List any document or field you could not read clearly. Do not change the original contents. Pause and show me before sending, submitting, posting or changing anything.

Next step

Want this built? Book a free 30-minute consult

Bring one real example, such as an export, an inbox, or a spreadsheet. We will map the steps and decide what SI drafts and what you approve.