Excel to JSON

In-Depth Guide

An Excel-to-JSON converter reads an .xlsx file — the OOXML (ECMA-376) format Microsoft adopted in Office 2007 — and emits a JSON array of objects where each row becomes an object keyed by the header row values. Under the surface, an .xlsx is a ZIP archive of XML parts: xl/worksheets/sheet1.xml contains cell data, xl/sharedStrings.xml deduplicates repeated strings, and xl/styles.xml carries formatting. The converter parses those parts client-side using a WebAssembly-compiled xlsx library, extracts cell values with their correct JavaScript types (string, number, boolean, Date), and serialises the result as valid RFC 8259 JSON. FastTool runs the entire conversion in the browser, so your customer lists, financial trial balances, and HR exports never touch a server. One spreadsheet in, one JSON file out, no upload, no cloud pipeline, no login.

Why This Matters

Data engineers, analysts, and backend developers receive Excel files from every non-technical counterparty in the business — finance ledgers from accounting, customer lists from sales, SKU inventories from operations — and need them as JSON to load into a Node/Python ETL, feed a REST API, or seed a database. Commercial ETL tools charge per-seat; ad-hoc conversions that happen twice a month do not justify that. A free, local converter that handles multi-sheet files, typed values, and date serialisation deletes an hour of scripting every time it runs.

Real-World Case Studies

• Database seeding. A data engineer inherits a 50,000-row customer export from sales in .xlsx format. She converts it to JSON, pipes the result through a Node script that maps columns to her PostgreSQL schema, and bulk-inserts with a single COPY command. The alternative — asking sales to re-export as CSV and manually fixing the quoting — would have cost a half-day. Instead, the seed is live on staging in thirty minutes and the PM can QA the migration the same afternoon.
• API client generation. A frontend developer needs sample product data to stub out a React prototype while the backend is still in design. Marketing hands her the product catalogue as .xlsx. She converts to JSON, pastes it into a local db.json for json-server, and has a working mock API in ten minutes — no waiting for the backend team's first sprint to complete. When the real API lands, the contract already matches because the JSON shape was the design artifact everybody worked against.
• Config migration. A DevOps engineer is replacing a legacy internal tool whose entire configuration lives in an Excel sheet curated by a non-technical ops lead. He converts the sheet to JSON, commits it as config.json in the new repo, and wires the ops lead up with a simple GUI that writes to the JSON. The conversion preserves every row and column, giving the team confidence that the migration loses nothing from the old spreadsheet.

Technical Deep Dive

The converter parses .xlsx (OOXML, ECMA-376) and legacy .xls (BIFF) formats. Each workbook is a ZIP of XML parts; the sheet XML references values either inline or as integer indexes into sharedStrings.xml. Cell types follow the XML attribute t=: s is a shared string, n is a number, b is boolean, d is an ISO 8601 date, and blank cells default to number. Dates in Excel are serial numbers: the default epoch is 1900-01-01 with the infamous 29-Feb-1900 off-by-one bug (Microsoft preserved the bug for Lotus compatibility). The converter corrects for this and emits dates as ISO 8601 strings. Formulas are NOT evaluated — the tool reads the cached formula result that Excel writes alongside the formula expression. Numbers are IEEE 754 double-precision floats limited to 15 significant decimal digits, which is why a 16-digit credit-card number stored without quoting loses its last digit. The first row is treated as the header by default, producing an array of objects; alternate modes produce arrays of arrays or nested per-sheet dictionaries. All processing is in-browser using WebAssembly; no upload, no server-side parse.