JSON data in SQL Server

THIS TOPIC APPLIES TO: yesSQL Server (starting with 2016)yesAzure SQL DatabasenoAzure SQL Data Warehouse noParallel Data Warehouse

JSON is a popular textual data format that's used for exchanging data in modern web and mobile applications. JSON is also used for storing unstructured data in log files or NoSQL databases such as Microsoft Azure Cosmos DB. Many REST web services return results that are formatted as JSON text or accept data that's formatted as JSON. For example, most Azure services, such as Azure Search, Azure Storage, and Azure Cosmos DB, have REST endpoints that return or consume JSON. JSON is also the main format for exchanging data between webpages and web servers by using AJAX calls.

JSON functions in SQL Server enable you to combine NoSQL and relational concepts in the same database. Now you can combine classic relational columns with columns that contain documents formatted as JSON text in the same table, parse and import JSON documents in relational structures, or format relational data to JSON text. You see see how JSON functions connect relational and NoSQL concepts in SQL Server and Azure SQL Database in the following video:

JSON as a bridge between NoSQL and relational worlds

Here's an example of JSON text:

    "name": "John",
    "skills": ["SQL", "C#", "Azure"]
}, {
    "name": "Jane",
    "surname": "Doe"

By using SQL Server built-in functions and operators, you can do the following things with JSON text:

  • Parse JSON text and read or modify values.
  • Transform arrays of JSON objects into table format.
  • Run any Transact-SQL query on the converted JSON objects.
  • Format the results of Transact-SQL queries in JSON format.

Overview of built-in JSON support

Key JSON capabilities of SQL Server and SQL Database

The next sections discuss the key capabilities that SQL Server provides with its built-in JSON support. You can see how to use JSON functions and operators in the following video:

SQL Server 2016 and JSON Support

Extract values from JSON text and use them in queries

If you have JSON text that's stored in database tables, you can read or modify values in the JSON text by using the following built-in functions:

  • JSON_VALUE extracts a scalar value from a JSON string.
  • JSON_QUERY extracts an object or an array from a JSON string.
  • ISJSON tests whether a string contains valid JSON.
  • JSON_MODIFY changes a value in a JSON string.


In the following example, the query uses both relational and JSON data (stored in a column named jsonCol) from a table:

SELECT Name,Surname,
 JSON_VALUE(jsonCol,'$.info.address.PostCode') AS PostCode,
 JSON_VALUE(jsonCol,'$.info.address."Address Line 1"')+' '
  +JSON_VALUE(jsonCol,'$.info.address."Address Line 2"') AS Address,
 JSON_QUERY(jsonCol,'$.info.skills') AS Skills
FROM People
 AND JSON_VALUE(jsonCol,'$.info.address.Town')='Belgrade'
 AND Status='Active'
ORDER BY JSON_VALUE(jsonCol,'$.info.address.PostCode')

Applications and tools see no difference between the values taken from scalar table columns and the values taken from JSON columns. You can use values from JSON text in any part of a Transact-SQL query (including WHERE, ORDER BY, or GROUP BY clauses, window aggregates, and so on). JSON functions use JavaScript-like syntax for referencing values inside JSON text.

For more information, see Validate, query, and change JSON data with built-in functions (SQL Server), JSON_VALUE (Transact-SQL), and JSON_QUERY (Transact-SQL).

Change JSON values

If you must modify parts of JSON text, you can use the JSON_MODIFY function to update the value of a property in a JSON string and return the updated JSON string. The following example updates the value of a property in a variable that contains JSON:


SET @jsonInfo=JSON_MODIFY(@jsonInfo,'$.info.address[0].town','London') 

Convert JSON collections to a rowset

You don't need a custom query language to query JSON in SQL Server. To query JSON data, you can use standard T-SQL. If you must create a query or report on JSON data, you can easily convert JSON data to rows and columns by calling the OPENJSON rowset function. For more information, see Convert JSON Data to Rows and Columns with OPENJSON (SQL Server).

The following example calls OPENJSON and transforms the array of objects that is stored in the @json variable to a rowset that can be queried with a standard SQL SELECT statement:

SET @json =  
       { "id" : 2,"info": { "name": "John", "surname": "Smith" }, "age": 25 },  
       { "id" : 5,"info": { "name": "Jane", "surname": "Smith" }, "dob": "2005-11-04T12:00:00" }  

  WITH (id int 'strict $.id',  
        firstName nvarchar(50) '$.info.name', lastName nvarchar(50) '$.info.surname',  
        age int, dateOfBirth datetime2 '$.dob')  


id firstName lastName age dateOfBirth
2 John Smith 25
5 Jane Smith 2005-11-04T12:00:00

OPENJSON transforms the array of JSON objects into a table in which each object is represented as one row, and key/value pairs are returned as cells. The output observes the following rules:

  • OPENJSON converts JSON values to the types that are specified in the WITH clause.
  • OPENJSON can handle both flat key/value pairs and nested, hierarchically organized objects.
  • You don't have to return all the fields that are contained in the JSON text.
  • If JSON values don't exist, OPENJSON returns NULL values.
  • You can optionally specify a path after the type specification to reference a nested property or to reference a property by a different name.
  • The optional strict prefix in the path specifies that values for the specified properties must exist in the JSON text.

For more information, see Convert JSON Data to Rows and Columns with OPENJSON (SQL Server) and OPENJSON (Transact-SQL).

Convert SQL Server data to JSON or export JSON

Format SQL Server data or the results of SQL queries as JSON by adding the FOR JSON clause to a SELECT statement. Use FOR JSON to delegate the formatting of JSON output from your client applications to SQL Server. For more information, see Format Query Results as JSON with FOR JSON (SQL Server).

The following example uses PATH mode with the FOR JSON clause:

SELECT id, firstName AS "info.name", lastName AS "info.surname", age, dateOfBirth as dob  
FROM People  

The FOR JSON clause formats SQL results as JSON text that can be provided to any app that understands JSON. The PATH option uses dot-separated aliases in the SELECT clause to nest objects in the query results.


    "id": 2,
    "info": {
        "name": "John",
        "surname": "Smith"
    "age": 25
}, {
    "id": 5,
    "info": {
        "name": "Jane",
        "surname": "Smith"
    "dob": "2005-11-04T12:00:00"

For more information, see Format query results as JSON with FOR JSON (SQL Server) and FOR Clause (Transact-SQL).

Combine relational and JSON data

SQL Server provides a hybrid model for storing and processing both relational and JSON data by using standard Transact-SQL language. You can organize collections of your JSON documents in tables, establish relationships between them, combine strongly typed scalar columns stored in tables with flexible key/value pairs stored in JSON columns, and query both scalar and JSON values in one or more tables by using full Transact-SQL.

JSON text is stored in varchar or nvarchar columns and is indexed as plain text. Any SQL Server feature or component that supports text supports JSON, so there are almost no constraints on interaction between JSON and other SQL Server features. You can store JSON in In-memory or Temporal tables, apply Row-Level Security predicates on JSON text, and so on.

If you have pure JSON workloads where you want to use a query language that's customized for the processing of JSON documents, consider Microsoft Azure Cosmos DB.

Here are some use cases that show how you can use the built-in JSON support in SQL Server.

Store and index JSON data in SQL Server

To learn more about your options for storing, indexing, and optimizing JSON data in SQL Server, see the following articles:

Load JSON files into SQL Server

You can format information that's stored in files as standard JSON or line-delimited JSON. SQL Server can import the contents of JSON files, parse it by using the OPENJSON or JSON_VALUE functions, and load it into tables.

  • If your JSON documents are stored in local files, on shared network drives, or in Azure Files locations that can be accessed by SQL Server, you can use bulk import to load your JSON data into SQL Server. For more information about this scenario, see Importing JSON files into SQL Server using OPENROWSET (BULK).

  • If your line-delimited JSON files are stored in Azure Blob storage or the Hadoop file system, you can use PolyBase to load JSON text, parse it in Transact-SQL code, and load it into tables.

Import JSON data into SQL Server tables

If you must load JSON data from an external service into SQL Server, you can use OPENJSON to import the data into SQL Server instead of parsing the data in the application layer.


SET @jsonVariable = N'[  
          "Order": {  
          "Item": {  
          "Order": {  
          "Item": {  

INSERT INTO SalesReport  
SELECT SalesOrderJsonData.*  
FROM OPENJSON (@jsonVariable, N'$.Orders.OrdersArray')  
           WITH (  
              Number   varchar(200) N'$.Order.Number',   
              Date     datetime     N'$.Order.Date',  
              Customer varchar(200) N'$.AccountNumber',   
              Quantity int          N'$.Item.Quantity'  
  AS SalesOrderJsonData;  

You can provide the content of the JSON variable by an external REST service, send it as a parameter from a client-side JavaScript framework, or load it from external files. You can easily insert, update, or merge results from JSON text into a SQL Server table. For more information about this scenario, see the following blog posts:

Analyze JSON data with SQL queries

If you must filter or aggregate JSON data for reporting purposes, you can use OPENJSON to transform JSON to relational format. You can then use standard Transact-SQL and built-in functions to prepare the reports.

SELECT Tab.Id, SalesOrderJsonData.Customer, SalesOrderJsonData.Date  
FROM   SalesOrderRecord AS Tab  
          CROSS APPLY  
     OPENJSON (Tab.json, N'$.Orders.OrdersArray')  
           WITH (  
              Number   varchar(200) N'$.Order.Number',   
              Date     datetime     N'$.Order.Date',  
              Customer varchar(200) N'$.AccountNumber',   
              Quantity int          N'$.Item.Quantity'  
  AS SalesOrderJsonData  
WHERE JSON_VALUE(Tab.json, '$.Status') = N'Closed'  
ORDER BY JSON_VALUE(Tab.json, '$.Group'), Tab.DateModified  

You can use both standard table columns and values from JSON text in the same query. You can add indexes on the JSON_VALUE(Tab.json, '$.Status') expression to improve the performance of the query. For more information, see Index JSON data.

Return data from a SQL Server table formatted as JSON

If you have a web service that takes data from the database layer and returns it in JSON format, or if you have JavaScript frameworks or libraries that accept data formatted as JSON, you can format JSON output directly in a SQL query. Instead of writing code or including a library to convert tabular query results and then serialize objects to JSON format, you can use FOR JSON to delegate the JSON formatting to SQL Server.

For example, you might want to generate JSON output that's compliant with the OData specification. The web service expects a request and response in the following format:

  • Request: /Northwind/Northwind.svc/Products(1)?$select=ProductID,ProductName

  • Response: {"@odata.context":"http://services.odata.org/V4/Northwind/Northwind.svc/$metadata#Products(ProductID,ProductName)/$entity","ProductID":1,"ProductName":"Chai"}

This OData URL represents a request for the ProductID and ProductName columns for the product with id 1. You can use FOR JSON to format the output as expected in SQL Server.

SELECT 'http://services.odata.org/V4/Northwind/Northwind.svc/$metadata#Products(ProductID,ProductName)/$entity'
 AS '@odata.context',   
 ProductID, Name as ProductName   
FROM Production.Product  
WHERE ProductID = 1  

The output of this query is JSON text that's fully compliant with the OData spec. Formatting and escaping are handled by SQL Server. SQL Server can also format query results in any format, such as OData JSON or GeoJSON. For more information, see Returning spatial data in GeoJSON format.

Test drive built-in JSON support with the AdventureWorks sample database

To get the AdventureWorks sample database, download at least the database file and the samples and scripts file from Microsoft Download Center.

After you restore the sample database to an instance of SQL Server 2016, extract the samples file, and then open the JSON Sample Queries procedures views and indexes.sql file from the JSON folder. Run the scripts in this file to reformat some existing data as JSON data, test sample queries and reports over the JSON data, index the JSON data, and import and export JSON.

Here's what you can do with the scripts that are included in the file:

  • Denormalize the existing schema to create columns of JSON data.

    • Store information from SalesReasons, SalesOrderDetails, SalesPerson, Customer, and other tables that contain information related to sales order into JSON columns in the SalesOrder_json table.

    • Store information from EmailAddresses/PersonPhone tables in the Person_json table as arrays of JSON objects.

  • Create procedures and views that query JSON data.

  • Index JSON data. Create indexes on JSON properties and full-text indexes.

  • Import and export JSON. Create and run procedures that export the content of the Person and the SalesOrder tables as JSON results, and import and update the Person and the SalesOrder tables by using JSON input.

  • Run query examples. Run some queries that call the stored procedures and views that you created in steps 2 and 4.

  • Clean up scripts. Don't run this part if you want to keep the stored procedures and views that you created in steps 2 and 4.

Learn more about JSON in SQL Server and Azure SQL Database

Microsoft blog posts

For specific solutions, use cases, and recommendations, see these blog posts about the built-in JSON support in SQL Server and Azure SQL Database.

Microsoft videos

For a visual introduction to the built-in JSON support in SQL Server and Azure SQL Database, see the following video:

Using JSON in SQL Server 2016 and Azure SQL Database

Building REST API with SQL Server using JSON functions

Reference articles