Your data already exists — it's just spread across exports that don't match. Different column names, inconsistent date formats, duplicates, numbers stored as text, half-empty rows.
I normalize all of it into one clean dataset you can import anywhere: Excel, CSV, JSON, or straight into a SQL database. Deduplication, validation, type fixing, and a summary of what changed so you can see exactly what happened to your data.
Recent example: six brokerage exports (Robinhood, Fidelity, Schwab, Webull, Computershare, Fundrise) with six different formats, unified into a single normalized table — see my portfolio.
For recurring work, I wrap the pipeline in Docker on a schedule so new files get processed automatically.
What I need to start: a sample of your current files and the output format you want.