Data-Driven Testing

Data-driven testing lets Orbittest Client run the same request many times — once per row of your data. You prepare a small data file (a spreadsheet saved as CSV, or a JSON list), reference its columns as {{variables}} in your request, and Orbittest Client swaps in each row as it runs.

The two formats behave identically — pick whichever is easier for you.

📊 Use a spreadsheet (CSV)

The easiest path. Build a table in Excel or Google Sheets and save it as CSV. Jump to the CSV guide ↓

🧩 Use JSON

Best when your data is already JSON or comes from code. Jump to the JSON guide ↓

The easy way — with Excel (CSV)

The simplest way to make a data file is in Excel (or Google Sheets), then save it as a CSV file.

1. Set up your sheet. Put your column names in the first row — these names become your variables. Each row below is one test run. For example, to test fetching different users:

   userId	expectedName
   1	Leanne Graham
   2	Ervin Howell
   3	Clementine Bauch

   Tip

   Keep column names simple — letters and numbers, no spaces (use userId, not user id).

2. Save it as CSV. In Excel: File → Save As → “CSV (Comma delimited) (*.csv)”, and save it as e.g. users.csv. (In Google Sheets: File → Download → Comma-separated values (.csv).) Your file now looks like this inside:

   users.csv

   userId,expectedName
   1,Leanne Graham
   2,Ervin Howell
   3,Clementine Bauch

3. Use the columns in your request. Reference any column with {{columnName}} — in the URL, headers, or body. For example, in the URL:

   https://api.example.com/users/{{userId}}

4. Run it. Right-click your collection → Run → Attach data file → pick users.csv → Run. Orbittest Client runs the request once per row, swapping {{userId}} each time (/users/1, /users/2, /users/3).

In Post-Run scripts you can also read the current row with ot.data.columnName and check the results:

ot.test("name matches", () => {
  ot.expect(ot.response.json.name).toBe(ot.data.expectedName);
});

Example — creating records (POST)

Spreadsheet:

name	email
Jane Doe	jane@example.com
John Smith	john@example.com

Request body:

{ "name": "{{name}}", "email": "{{email}}" }

Each row sends one new record.

Sample file

Grab a ready-made users.csv to try it immediately.

JSON format

Prefer JSON? Orbittest Client accepts that too. A JSON data file is simply an array (a list) of objects, where each object is one test run and its keys are your variable names.

[
  { "column1": "value", "column2": "value" },
  { "column1": "value", "column2": "value" }
]

• The square brackets [ ] wrap the whole list.
• Each { } block is one iteration (one run of your request).
• The keys (userId, expectedName…) become your {{variables}}.

1. Write your data. Using the same “fetch users” example:

   users.json

   [
     { "userId": 1, "expectedName": "Leanne Graham" },
     { "userId": 2, "expectedName": "Ervin Howell" },
     { "userId": 3, "expectedName": "Clementine Bauch" }
   ]

   Tip

   You don’t have to type JSON by hand. Build the table in Excel exactly like the CSV example, then use any free “CSV to JSON” converter — it produces the array above automatically.

2. Save it with a .json extension, e.g. users.json. Make sure:

   • Text values are wrapped in double quotes ("Leanne Graham").
   • Numbers don’t need quotes (1), but quoting them is fine too.
   • Every object is separated by a comma, and there’s no comma after the last one.

3. Use the keys in your request — exactly like CSV, reference any key with {{keyName}}:

   https://api.example.com/users/{{userId}}

4. Run it. Right-click your collection → Run → Attach data file → pick users.json → Run. It runs once per object in the list.

Example — creating records (POST)

Data file:

new-users.json

[
  { "name": "Jane Doe",   "email": "jane@example.com" },
  { "name": "John Smith", "email": "john@example.com" }
]

Request body:

{ "name": "{{name}}", "email": "{{email}}" }

Sample files

Try them right away: users.json · new-users.json

CSV vs JSON — which should I use?

	CSV	JSON
Easiest to make	✅ Excel / Sheets → Save as CSV	Best when data is already JSON, or has nested values
Looks like	a spreadsheet	a list of objects
Good when	simple rows of text / numbers	you’re comfortable with JSON or exporting from code

Both behave identically in Orbittest Client — pick whichever is easier for you. 🎉

Where the data shows up

Each row’s columns are exposed as {{variables}} in the request and as ot.data.<column> inside your Pre-Run and Post-Run scripts, alongside {{$iteration}} for the current row number. See Pre-Run and Post-Run scripts for validation patterns.