How to Query Nested JSON with JSONPath (Without Writing Loops)

You just got back a 300-line API response. Somewhere inside three levels of nesting is the email field you actually need. So you write a loop, then another loop, then a conditional — and now you're maintaining brittle traversal code that breaks every time the API schema shifts.

There's a better way: JSONPath.

JSONPath is a query language for JSON, similar to how XPath works for XML. Instead of writing code to traverse a structure, you write a short expression that reads like a path. It works across languages, and once you learn the syntax, you'll reach for it constantly.

The Basics: What JSONPath Looks Like

Here's a typical API response — a list of orders, each with nested customer and item data:

{

  "store": {

    "orders": [

      {

        "id": 1,

        "customer": { "name": "Alice", "email": "[email protected]" },

        "items": [{"sku": "A1", "qty": 2}, {"sku": "B3", "qty": 1}]

      },

      {

        "id": 2,

        "customer": { "name": "Bob", "email": "[email protected]" },

        "items": [{"sku": "C7", "qty": 5}]

      }

    ]

  }

}

To get every customer email without writing any loops:

$.store.orders[*].customer.email

Result:

["[email protected]", "[email protected]"]

The $ represents the root. The . navigates into an object key. [*] means "all array elements." Three characters replace a full nested loop.

Filtering: The Real Power

JSONPath shines when you need to filter by a condition. Say you only want items where qty is greater than 1:

$.store.orders[*].items[?(@.qty > 1)]

Result:

[

  {"sku": "A1", "qty": 2},

  {"sku": "C7", "qty": 5}

]

The ?() is a filter expression. @ refers to the current node. You can combine conditions with && and ||, check for key existence with @.key, and do string comparisons too.

Here's the equivalent Python — same outcome, but compare the cognitive load:

# Without JSONPath — five lines, easy to get wrong

results = []

for order in data["store"]["orders"]:

    for item in order["items"]:

        if item["qty"] > 1:

            results.append(item)

The JSONPath version is one line and self-documenting. Anyone reading it immediately knows what data you're after.

Recursive Descent: Finding a Key Anywhere in the Tree

Sometimes you don't know exactly where a key lives — you just know it exists somewhere in a deeply nested structure. The .. (recursive descent) operator handles this:

$..email

This finds every email field at any depth in the document. It's invaluable for exploring unfamiliar API schemas or debugging payloads where nesting is inconsistent across records.

Try it live with the JSONPath Evaluator on jsonindenter.com — paste your JSON, type any expression, and see matched results highlighted instantly. No installation, nothing leaves your browser.

A Practical Workflow for Real API Responses

When you're dealing with a large, unfamiliar JSON blob, this three-step routine saves a lot of "why does my query return nothing?" debugging time:

1. Paste the response into the JSON Beautifier to get a clean indented view and understand the shape of the data.


2. Run it through the JSON Validator to confirm it's well-formed — a malformed payload will silently produce empty results and send you chasing the wrong problem.


3. Switch to the JSONPath tool and iterate on your expression until you're extracting exactly what you need.

Quick Reference: Syntax That Covers 80% of Cases

• 
  $ — root element


• 
  .key — child key


• 
  ..key — recursive search for key at any depth


• 
  [*] — all array elements


• 
  [0] — first element; [-1] — last element


• 
  [0,2] — elements at index 0 and 2


• 
  [?(@.price < 10)] — filter where price is less than 10


• 
  @.key — current node's key (used inside filter expressions)

For edge cases and implementation differences across languages, the JSONPath guide on jsonindenter.com covers the full spec with annotated examples worth bookmarking.

When JSONPath Is the Wrong Tool

JSONPath is read-only — it queries and extracts, it doesn't mutate. If you need to patch a JSON document in place (add a field, replace a value, remove a key), that's what JSON Patch is designed for. The two tools are complementary: JSONPath to find, JSON Patch to modify.

Also worth knowing: JSONPath implementations vary across languages. jsonpath-ng in Python, jsonpath-plus in JavaScript, and Jayway in Java each have subtle differences. RFC 9535 (published in 2024) is working to standardize the spec, but for production code it's worth testing your expressions against the specific library you're targeting.

JSONPath Is Worth Adding to Your Toolkit

The argument for JSONPath isn't that it's magic — it's that it's composable and readable. A well-written JSONPath expression communicates intent at a glance. Nested loops require reading to the end before you understand what's being extracted.

It's also portable: the same expression works in JavaScript, Python, Java, and tools like Postman, AWS CloudFormation, and Kubernetes JSON patches. Learning it once pays off across your entire stack.

What's the most complex JSON query you've had to write? Did you reach for JSONPath, or did you end up with custom traversal logic you later regretted?

Free tools used in this post:

• 
  JSONPath Evaluator — test JSONPath expressions against any JSON payload, results highlighted live


• 
  JSON Beautifier — format and indent raw JSON for easy reading


• 
  JSON Validator — confirm your JSON is well-formed before querying it


• 
  JSON Patch — apply RFC 6902 patches to modify JSON structures in place


• 
  All tools — client-side, no sign-up, nothing leaves your browser