Guides
Field-tested walkthroughs for getting handwritten lab data off paper and into a spreadsheet you can actually analyze — written for the person doing it by hand today.
How to Turn a Handwritten Lab Notebook into a Spreadsheet
The manual method for transcribing paper lab notebooks into Excel, why the hard part is reconciliation rather than typing, and how to keep an audit trail.
Read the guideWhy Generic OCR Fails on Handwritten Lab Data
Why OCR that scores well in a vendor demo still produces wrong spreadsheets from real notebooks — and a checklist for testing a tool honestly before you rely on it.
Read the guideReconciliation vs. OCR: Why Reading Handwriting Is Not the Hard Part
The distinction between reading a page and reconciling a document — and why picking a tool that only does the first one still leaves you doing the hard part by hand.
Read the guideHow Continuation Pages Break Structured Data Extraction (and How to Fix It)
Why a second or third page that does not repeat its header breaks row-position-based extraction, and the identity-resolution approach that gets it right.
Read the guideWhy We Never Let an LLM Write the Final Number
The architectural line between where an LLM should make a judgment call and where deterministic code should take over — and why blurring it breaks reproducibility.
Read the guideConfidence Scores, Not Guesses: Designing an Extraction Pipeline That Flags Instead of Fabricates
Why extraction tools optimized to always return a clean-looking value are actually hiding their errors, and the flag-don't-guess pattern that fixes it.
Read the guide