LedgerLens is a native desktop app that turns messy PDF bank statements into clean, spreadsheet-ready rows. It auto-detects different bank layouts, extracts every transaction — date, description, debit, credit, balance — and reconciles to the penny. Because it runs entirely on your device, your financial data is never uploaded, and you don't even need an account to start.
Hey Product Hunt 👋
I built LedgerLens to kill one small, deeply annoying job: getting transactions out of PDF bank statements and into a spreadsheet without hand-typing them.
The problem: anyone who does bookkeeping — accountants, founders, small-business owners — knows the pain of copy-pasting statements into Excel, or fighting with a bank export that doesn't exist. There are tools that do this… but almost all of them ask you to upload your bank statements to their server. For your most sensitive financial data, that's a hard no for a lot of people.
The evolution: I actually started this as a web app. Then it hit me — the "upload your statements to us" model is the problem, not the solution. So I scrapped that and rebuilt LedgerLens as a native desktop app for macOS and Windows that does everything 100% on your device. Your statements never leave your computer. No account needed to start. (The Mac build is Apple-notarized, so it just opens.)
What it does:
📄 Drop in PDF statements → clean CSV / XLSX rows
🏦 Auto-detects different bank layouts and reconciles every balance to the penny
🔑 Handles password-protected PDFs + batch conversions
🔒 Fully local — nothing uploaded, ever
Fun bit of backstory: the idea came from a Starter Story list of micro-SaaS ideas. Turning "someone should build this" into a shipped, notarized app with a landing page and payments was the whole fun of it.
Your first 5 statements are free — I'd genuinely love your feedback. And if there's a specific bank or statement format you want supported next, drop it in the comments and I'll add a template. 🙏
Report
Auto-detecting bank layouts and reconciling to the penny is the hard part, most tools stop at "extracted something that looks like a number." Question from someone who's built PDF statement parsing before: how does it handle statements that are scanned images rather than text-layer PDFs, since that's usually where auto-detection breaks down and you're relying on OCR instead of just reading the text? And when a reconciliation doesn't add up, does it flag the specific row that's off, or just tell you the total is wrong?
Report
Reviews
No reviews yetBe the first to leave a review for LedgerLens
Auto-detecting bank layouts and reconciling to the penny is the hard part, most tools stop at "extracted something that looks like a number." Question from someone who's built PDF statement parsing before: how does it handle statements that are scanned images rather than text-layer PDFs, since that's usually where auto-detection breaks down and you're relying on OCR instead of just reading the text? And when a reconciliation doesn't add up, does it flag the specific row that's off, or just tell you the total is wrong?