Publish

Publish sends your transformed data directly to an external API — row by row — as structured JSON HTTP requests. Instead of downloading a file and importing it yourself, WeTransform delivers each row of your transformed data to the endpoint of your choice.

How It Works

Publish operates at the Template level. Once configured, every transformation using that template will automatically push results after completion.

Use Cases

Setup

Publish is configured on a per-template basis. Navigate to your template's settings and enable the Publish feature.

JSON Builder

The JSON Builder lets you visually design the JSON payload that will be sent for each row. It works as a tree of nodes, where each node represents either a structural element (object, array) or a value (column reference, literal string, number, etc.).

Example: if your template has columns product_name (text), price (number), and in_stock (boolean), you can build:

Root (object)
  |-- "name"       -> Column: product_name
  |-- "price"      -> Column: price
  |-- "available"  -> Column: in_stock
  |-- "source"     -> String: "WeTransform"

For a row where product_name = "Widget A", price = 19.99, in_stock = true, the resulting JSON body sent to your API would be:

{
  "name": "Widget A",
  "price": 19.99,
  "available": true,
  "source": "WeTransform"
}

Node Types

Each node in the JSON tree has a type that determines how its value is resolved:

Column type inference

When using column nodes, the JSON type is determined by the template column's data type:

Omit if empty

Each node has an "omit if empty" option. When enabled, the node is excluded from the output if its resolved value is null, an empty string, or an empty array. This keeps your payloads clean by avoiding unnecessary null fields.

HTTP Settings

Configure how each request is sent to your API:

{
  "url": "https://api.example.com/products",
  "method": "POST",
  "headers": {
    "Authorization": "Bearer sk-abc123...",
    "Content-Type": "application/json",
    "X-Custom-Header": "my-value"
  },
  "cookies": {}
}

Rate Limiting

Many APIs enforce rate limits (e.g., "max 100 requests per minute"). Publish includes a built-in sliding-window rate limiter that automatically throttles requests to stay within your configured limits.

Examples

When the limit is reached, Publish automatically pauses and resumes once the window allows more requests. No rows are dropped — they are simply delayed.

Reports & Email Notifications

After every push execution, a detailed CSV report is generated and stored. Each row in the report corresponds to one HTTP request.

Report columns

Email notifications

If you provide a notification email, WeTransform sends a summary after each push execution. The email includes:

• Transformation statistics: rows in source, rows processed, rows in error, success percentage.
• Push statistics: total requests, successful requests, failed requests, success percentage.
• Attachments: the push report CSV, and (if there were transformation errors) the transformation error report CSV.

The push report is also available for download for 30 days from the submission details page.

Examples

Simple product sync

Push product data to an external catalog API:

Root (object)
  |-- "sku"         -> Column: product_sku
  |-- "name"        -> Column: product_name
  |-- "price"       -> Column: unit_price
  |-- "currency"    -> String: "EUR"
  |-- "in_stock"    -> Column: available

{
  "sku": "PROD-001",
  "name": "Ergonomic Chair",
  "price": 299.99,
  "currency": "EUR",
  "in_stock": true
}

Nested structure with arrays

Build complex payloads with nested objects and arrays:

Root (object)
  |-- "id"          -> Column: employee_id
  |-- "person"      -> Object
  |     |-- "first" -> Column: first_name
  |     |-- "last"  -> Column: last_name
  |     |-- "email" -> Column: email
  |-- "tags"        -> Array
  |     |-- [0]     -> Column: department
  |     |-- [1]     -> String: "imported"
  |-- "active"      -> Boolean: true

{
  "id": "EMP-042",
  "person": {
    "first": "Alice",
    "last": "Martin",
    "email": "alice.martin@example.com"
  },
  "tags": ["Engineering", "imported"],
  "active": true
}

Using "omit if empty"

Keep your API payloads clean by omitting fields when values are missing:

Root (object)
  |-- "name"   -> Column: contact_name
  |-- "email"  -> Column: contact_email
  |-- "phone"  -> Column: contact_phone    (omit if empty)
  |-- "notes"  -> Column: contact_notes    (omit if empty)

If contact_phone and contact_notes are empty for a row, the output will be:

{
  "name": "Bob Smith",
  "email": "bob@example.com"
}