Free online JSON to MDB converter

JSON data often needs conversion when an older Microsoft Access application, reporting workflow, or database archive requires an .mdb file. The conversion creates Access tables from JSON records, but nested objects and arrays must be flattened or stored as JSON text first.

What the JSON format is

JSON is a text-based data-interchange format built from objects, arrays, strings, numbers, Boolean values, and null. Web APIs, configuration tools, JavaScript applications, logging systems, and export features in databases commonly produce .json files.

A JSON file may contain one object, an array of records, or a wrapper such as {"data":[...]}. Access tables require a fixed set of columns, so nested objects such as customer.address.city need separate columns or serialized JSON text, and arrays usually need a related child table.

What the MDB format is

MDB is the Microsoft Access database format used primarily by Access 2003 and earlier. It stores tables, indexes, queries, forms, reports, relationships, and other database objects in one file.

MDB is useful when a legacy application accepts only Access databases, when users need to edit records in Microsoft Access, or when an existing workflow depends on Jet/ACE database drivers. It supports typed columns, primary keys, indexes, and relationships that JSON does not provide. Its main limitations are the older format’s size and compatibility constraints: an MDB database is limited to approximately 2 GB, and newer Access features may require .accdb instead.

How to actually convert JSON to MDB

Use the built-in online converter

You can convert the file on 101convert.com for free without installing software. Upload the JSON file, select MDB as the output format, start the conversion, and download the resulting .mdb file. Inspect the tables and column types in Access afterward, especially if the JSON contains nested objects, arrays, mixed data types, or very long strings.

Convert locally with Microsoft Access and a database driver

1. Install Microsoft Access or the Microsoft Access Database Engine matching the bitness required by your Python installation.
2. Create a blank MDB database in Access through File → New → Blank database, and save it as output.mdb.
3. Inspect the JSON structure and decide which properties become columns. Use a separate table for repeated child objects instead of placing multiple values in one field.
4. Import the records with Python, json, and pyodbc. The connection string normally uses DRIVER={Microsoft Access Driver (*.mdb, *.accdb)};DBQ=C:\path\output.mdb;.
5. Create the destination table with SQL such as CREATE TABLE Records (ID LONG, Name TEXT(255), Amount DOUBLE, Active BIT), then insert values with parameterized statements such as INSERT INTO Records (ID, Name) VALUES (?, ?).
6. Open output.mdb in Access, verify the imported row count, assign a primary key, create indexes, and define relationships if the source contained related records.

A minimal local workflow for a flat JSON array is:

import json
import pyodbc

with open("input.json", encoding="utf-8") as f:
    records = json.load(f)

if isinstance(records, dict):
    records = records.get("data", records.get("records", []))

connection = pyodbc.connect(
    r"DRIVER={Microsoft Access Driver (*.mdb, *.accdb)};"
    r"DBQ=C:\path\output.mdb;"
)
cursor = connection.cursor()
cursor.execute("CREATE TABLE Records (Name TEXT(255), Amount DOUBLE, Active BIT)")

for row in records:
    cursor.execute(
        "INSERT INTO Records (Name, Amount, Active) VALUES (?, ?, ?)",
        row.get("name"), row.get("amount"), row.get("active")
    )

connection.commit()
connection.close()

This example assumes that output.mdb already exists and that every record has compatible scalar properties. A production script should inspect all records, quote Access identifiers, infer or explicitly define column types, convert ISO date strings to Python datetime values, and escape or serialize nested values before insertion.

Quality, compatibility, and limitations

JSON has no enforced schema, while MDB requires column definitions. Values that are numbers in some records and strings in others need a deliberate target type; otherwise the driver may reject rows or coerce values incorrectly. JSON null should normally become an Access NULL, not an empty string or zero.

Access text fields have defined sizes, and long values may require LONGTEXT rather than TEXT(255). JSON dates are strings unless the converter parses them, so confirm date and time formats after import. Property names containing spaces, punctuation, or reserved Access words need bracketed identifiers, for example [Order].

Test the resulting MDB with the exact legacy application and driver that will use it. A converter can preserve values while still losing JSON nesting, array order, schema metadata, duplicate object keys, or numeric precision. Keep the original JSON as the authoritative archive.