Destination Insert Functions

There are two ways you can access destination insert functions from your Segment space:

• From the Connections catalog.
• From the Destinations tab.

To create an insert function from Segment's catalog:

1. Navigate to Connections > Catalog > Functions and click New Function.
2. From the Select Type screen, select Insert and click Next: Build Function.
3. Write and test your function code. Manually enter a sample event and click Run to test.
4. Click Next: Configure & Create to add a function name, description, and logo.
5. Click Create Function to create your insert function. You'll see the insert function displayed in the Functions tab.

For data to flow to your downstream destinations, you'll need to connect your insert function to a destination:

1. Select the insert function you'd like to connect. From the side pane, you can edit, delete, and connect the insert function.
2. Click Connect a destination.
3. Select the destination you'd like to connect to and click Connect to destination.

To access insert functions through the Destinations tab:

1. Navigate to Connections > Destinations.
2. Select your destination.
3. Select Functions and then select your insert function.

Use this page to edit and manage insert functions in your workspace.

You can also use this page to enable destination insert functions in your workspace.

Segment invokes a separate part of the function (called a "handler") for each event type that you send to your destination insert function.

The default source code template includes handlers for all event types. You don't need to implement all of them - just use the ones you need, and skip the ones you don't. For event types that you want to send through the destination, return the event in the respective event handlers.

Insert functions can define handlers for each message type in the Segment spec:

• onIdentify
• onTrack
• onPage
• onScreen
• onGroup
• onAlias
• onDelete
• onBatch

Each of the functions above accepts two arguments:

• event - Segment event object, where fields and values depend on the event type. For example, in Identify events, Segment formats the object to match the Identify spec.
• settings - Set of settings for this function.

The example below shows a function that listens for "Track" events, and sends some details about them to an external service.

1
async function onTrack(event) {
2
  await fetch('https://example-service.com/api', {
3
    method: 'POST',
4
    headers: {
5
      'Content-Type': 'application/json'
6
    },
7
    body: JSON.stringify({
8
      event_name: event.event,
9
      event_properties: event.properties,
10
      timestamp: event.timestamp
11
    })
12
  })
13

14
  return event;
15
  
16
}

To change which event type the handler listens to, you can rename it to the name of the message type. For example, if you rename this function onIdentify, it listens for Identify events instead.

To ensure the destination processes an event payload modified by the function, return the event object at the handler's end.

When declaring settings variables, make sure to declare them in the function handler rather than globally in your function. This prevents you from leaking the settings values across other function instances.

The handler for insert functions is event-specific. For example, onTrack(), onIdentify(), and so on.

Segment considers a function's execution successful if it finishes without error. You can throw an error to create a failure on purpose. Use these errors to validate event data before processing it to ensure the function works as expected.

You can throw the following pre-defined error types to indicate that the function ran as expected, but the data was not deliverable:

• EventNotSupported
• InvalidEventPayload
• ValidationError
• RetryError
• DropEvent

The following examples show basic uses of these error types.

1
async function onGroup(event) {
2
  if (!event.traits.company) {
3
    throw new InvalidEventPayload('Company name is required')
4
  }
5
}
6

7
async function onPage(event) {
8
  if (!event.properties.pageName) {
9
    throw new ValidationError('Page name is required')
10
  }
11
}
12

13
async function onAlias(event) {
14
  throw new EventNotSupported('Alias event is not supported')
15
}
16

17
async function onTrack(event) {
18
  let res
19
  try {
20
    res = await fetch('http://example-service.com/api', {
21
      method: 'POST',
22
      headers: {
23
        'Content-Type': 'application/json'
24
      },
25
      body: JSON.stringify({ event })
26
    })
27

28
    return event;
29
    
30
  } catch (err) {
31
    // Retry on connection error
32
    throw new RetryError(err.message)
33
  }
34
  if (res.status >= 500 || res.status === 429) {
35
    // Retry on 5xx and 429s (ratelimits)
36
    throw new RetryError(`HTTP Status ${res.status}`)
37
  }
38
}
39

40
async function onIdentify(event) {
41
  if (event.traits.companyName) {
42
    // Drop Event | Do NOT forward event to destination
43
    throw new DropEvent('Company name is required')
44
  }
45
  return event;
46
}
47

If you don't supply a function for an event type, Segment throws an EventNotSupported error by default.

See errors and error handling for more information on supported error types and how to troubleshoot them.

Segment Functions run on the Node.js LTS runtime (currently v20). This keeps the runtime current with industry standards and security updates.

Based on the AWS Lambda and Node.js support schedule, production applications should only use Node.js releases that are in Active LTS or Maintenance LTS.

When Segment upgrades the Functions runtime to a new LTS version, existing functions automatically use the new runtime after their next deployment. Segment recommends checking your function after deployment to ensure everything works as expected, since dependency or syntax changes between Node.js versions might affect your function's behavior.

Functions don't support importing dependencies, but you can contact Segment Support to request that one be added.

The following dependencies are installed in the function environment by default:

• atob v2.1.2 exposed as atob

• aws-sdk v2.488.0 exposed as AWS

• btoa v1.2.1 exposed as btoa

• fetch-retry exposed as fetchretrylib.fetchretry

• form-data v2.4.0 exposed as FormData

• @google-cloud/automl v2.2.0 exposed as google.cloud.automl

• @google-cloud/bigquery v5.3.0 exposed as google.cloud.bigquery

• @google-cloud/datastore v6.2.0 exposed as google.cloud.datastore

• @google-cloud/firestore v4.4.0 exposed as google.cloud.firestore

• @google-cloud/functions v1.1.0 exposed as google.cloud.functions

• @google-cloud/pubsub v2.6.0 exposed as google.cloud.pubsub

• @google-cloud/storage v5.3.0 exposed as google.cloud.storage

• @google-cloud/tasks v2.6.0 exposed as google.cloud.tasks

• hubspot-api-nodejs exposed as hubspotlib.hubspot

• jsforce v1.11.0 exposed as jsforce

• jsonwebtoken v8.5.1 exposed as jsonwebtoken

• libphonenumber-js exposed as libphonenumberjslib.libphonenumberjs

• lodash v4.17.19 exposed as _

• mailchimp marketing exposed as mailchimplib.mailchimp

• mailjet exposed as const mailJet = nodemailjet.nodemailjet;

• moment-timezone v0.5.31 exposed as moment

• node-fetch v2.6.0 exposed as fetch

• oauth v0.9.15 exposed as OAuth

• @sendgrid/client v7.4.7 exposed as sendgrid.client

• @sendgrid/mail v7.4.7 exposed as sendgrid.mail

• skyflow exposed as skyflowlib.skyflow

• stripe v8.115.0 exposed as stripe

• twilio v3.68.0 exposed as twilio

• uuidv5 v1.0.0 exposed as uuidv5.uuidv5

• winston v2.4.6 exposed as const winston = winstonlib.winston

• xml v1.0.1 exposed as xml

• xml2js v0.4.23 exposed as xml2js

• zlib v1.0.5 exposed as zlib.zlib

  
  uuidv5 is exposed as an object. Use uuidv5.uuidv5 to access its functions. For example:

  Copy code block

  1
  async function onRequest(request, settings) {
  2
       uuidv5 = uuidv5.uuidv5;
  3
       console.log(typeof uuidv5);
  4
  
  5
        //Generate a UUID in the default URL namespace
  6
        var urlUUID = uuidv5('url', 'http://google/com/page');
  7
        console.log(urlUUID);
  8
  
  9
        //Default DNS namespace
  10
        var dnsUUID = uuidv5('dns', 'google.com');
  11
        console.log(dnsUUID);
  12
    }

  zlib's asynchronous methods inflate and deflate must be used with async or await. For example:

  Copy code block

  1
  zlib = zlib.zlib;  // Required to access zlib objects and associated functions
  2
  async function onRequest(request, settings) {
  3
  const body = request.json();
  4
  
  5
  const input = 'something';
  6
  
  7
  // Calling inflateSync method
  8
  var deflated = zlib.deflateSync(input);
  9
  
  10
  console.log(deflated.toString('base64'));
  11
  
  12
  // Calling inflateSync method
  13
  var inflated = zlib.inflateSync(new Buffer.from(deflated)).toString();
  14
  
  15
  console.log(inflated);
  16
  
  17
  console.log('Done');
  18
  }

The following Node.js modules are available:

• crypto Node.js module exposed as crypto.
• https Node.js module exposed as https.

Other built-in Node.js modules aren't available.

For more information on using the aws-sdk module, see how to set up functions for calling AWS APIs.

Basic cache storage is available through the cache object, which has the following methods defined:

• cache.load(key: string, ttl: number, fn: async () => any): Promise<any>
  • Obtains a cached value for the provided key, invoking the callback if the value is missing or has expired. The ttl is the maximum duration in milliseconds the value can be cached. If omitted or set to -1, the value will have no expiry.
• cache.delete(key: string): void
  • Immediately remove the value associated with the key.

Some important notes about the cache:

• When testing functions in the code editor, the cache will be empty because each test temporarily deploys a new instance of the function.
• Values in the cache are not shared between concurrently-running function instances; they are process-local which means that high-volume functions will have many separate caches.
• Values may be expunged at any time, even before the configured TTL is reached. This can happen due to memory pressure or normal scaling activity. Minimizing the size of cached values can improve your hit/miss ratio.
• Functions that receive a low volume of traffic may be temporarily suspended, during which their caches will be emptied. In general, caches are best used for high-volume functions and with long TTLs. The following example gets a JSON value through the cache, only invoking the callback as needed:

1
const ttl = 5 * 60 * 1000 // 5 minutes
2
const val = await cache.load("mycachekey", ttl, async () => {
3
    const res = await fetch("http://echo.jsontest.com/key/value/one/two")
4
    const data = await res.json()
5
    return data
6
})

A payload must come into the pipeline with the attributes that allow it to match your mapping triggers. You can't use an insert function to change the event to match your mapping triggers. If an event comes into an actions destination and already matches a mapping trigger, that mapping subscription will fire. If a payload doesn't come to the actions destination matching a mapping trigger, even if an insert function is meant to alter the event to allow it to match a trigger, it won't fire that mapping subscription. Segment sees the mapping trigger first in the pipeline, so a payload won't make it to the insert function at all if it doesn't come into the pipeline matching a mapping trigger.

Unlike Source functions and Destination functions, which return multiple events, an insert function only returns one event. When the insert function receives an event, it sends the event to be handled by its configured mappings.

If you would like multiple mappings triggered by the same event:

1. Create different types of mappings (Identify, Track, Page, etc) or multiple mappings of the same type.
2. Configure the mapping's trigger conditions to look for that event name/type or other available field within the payload.
3. Configure the mapped fields to send different data.

You can also configure the insert function to add additional data to the event's payload before it's handled by the mappings and configure the mapping's available fields to reference the payload's available fields.

You may want to consider the context object's available fields when adding new data to the event's payload.

Settings allow you to pass configurable variables to your function, which is the best way to pass sensitive information such as security tokens. For example, you might use settings as placeholders to use information such as an API endpoint and API key. This way, you can use the same code with different settings for different purposes. When you deploy a function in your workspace, you are prompted to fill out these settings to configure the function.

To use settings:

1. Add a setting in the Settings tab in the code editor:

Expand image

2. Click Add Setting to add your new setting.
3. Configure the details about this setting, which change how it's displayed to anyone using your function:

• Label - Name of the setting, which users see when configuring the function.
• Name - Auto-generated name of the setting to use in function's source code.
• Type - Type of the setting's value.
• Description - Optional description, which appears below the setting name.
• Required - Enable this to ensure that the setting cannot be saved without a value.
• Secret - Enable this to ensure that sensitive data, like API key values or passwords, are not displayed in the Segment UI.

As you change the values, a preview to the right updates to show how your setting will look and work.

4. Click Add Setting to save the new setting.

Once you save a setting, it appears in the Settings tab for the function. You can edit or delete settings from this tab.

5. Fill out this setting's value in the Test tab, so you can run the function and verify that the correct setting value is passed. (This value is only for testing your function.)
6. Add code to read its value and run the function, as in the example below:

1
async function onTrack(request, settings) {
2
  const apiKey = settings.apiKey
3
  //=> "super_secret_string"
4
}

When you deploy your destination insert function in your workspace, you fill out the settings on the destination configuration page, similar to how you would configure a normal destination.

You can manually test your code from the functions editor:

1. From the Test tab, click customize the event yourself and manually input your own JSON payload.
2. If your test fails, you can check the error details and logs in the Output section.

• Error messages display errors surfaced from your function.
• Logs display any messages to console.log() from the function.

Once you finish building your insert function, click Next: Configure & Create to name it, then click Create Function to save it.

Once you do that, you'll see the insert function from the Functions page in your catalog.

If you're editing an existing function, you can save changes without updating the instances of the function that are already deployed and running.

You can also choose to Save & Deploy to save the changes, then choose which already-deployed functions to update with your changes.

You need to enable your insert function for it to process your data.

To enable your insert function:

1. Navigate to Connections > Destinations.
2. Select your destination, then select the Functions tab.
3. Select the Enable Function toggle, and click Enable on the pop-out window.

To prevent your insert function from processing data, toggle Enable Function off.

Batch handlers are an extension of insert functions. When you define an onBatch handler alongside the handler functions for single events (for example, onTrack or onIdentity), you're telling Segment that the insert function can accept and handle batches of events.

Consider creating a batch handler if:

• You have a high-throughput function and want to reduce cost. When you define a batch handler, Segment invokes the function once per batch, rather than once per event. As long as the function's execution time isn't adversely affected, the reduction in invocations should lead to a reduction in cost.
• Your destination supports batching. When your downstream destination supports sending data downstream in batches you can define a batch handler to avoid throttling. Batching for functions is independent of batch size supported by the destination. Segment automatically handles batch formation for destinations.

Segment collects the events over a short period of time and combines them into a batch. The system flushes them when the batch reaches a certain number of events, or when the batch has been waiting for a specified wait time.

To create a batch handler, define an onBatch function within your destination insert function. You can also use the "Default Batch" template found in the Functions editor to get started quickly.

You must preserve the original order in the onBatch implementation, as Segment's function invoker service relies on positional consistency in the onBatch handler between the input and output arrays. When a function returns a transformed batch, Segment pairs each output event with its corresponding input using array index positions, not event IDs or timestamps.

1
async function onBatch(events, settings){
2
  // handle the batch of events
3
  return events
4
}

The handler function receives an array of events. The events can be of any supported type and a single batch may contain more than one event type. Handler functions can also receive function settings. Here is an example of what a batch can look like:

1
[
2
    {
3
      "type": "identify",
4
      "userId": "019mr8mf4r",
5
      "messageId": "ajs-next-1757955997356-c029ce21-9034-45a6-ac62-b8b23b655df5",
6
      "traits": {
7
        "email": "jake@yahoo.com",
8
        "name": "Jake Peterson",
9
        "age": 26
10
      }
11
    },
12
    {
13
      "type": "track",
14
      "userId": "019mr8mf4r",
15
      "messageId": "ajs-next-1757955997356-bde2ad8e-2af9-45f5-837b-993671f6b1b0",
16
      "event": "Song Played",
17
      "properties": {
18
        "name": "Fallin for You",
19
        "artist": "Dierks Bentley"
20
      }
21
    },
22
    {
23
      "type": "track",
24
      "userId": "971mj8mk7p",
25
      "messageId": "ajs-next-1757955997356-9e2ba07c-0085-4a5b-8928-e4450a417f74",
26
      "event": "Song Played",
27
      "properties": {
28
        "name": "Get Right",
29
        "artist": "Jennifer Lopez"
30
      }
31
    }
32
]

Segment batches together any event of any type that it sees over a short period of time to increase batching efficiency and give you the flexibility to decide how batches are created. If you want to split batches by event type, you can implement this in your functions code by writing a handler.

1
async function onBatch(events, settings) {
2
  // store original order
3
  const originalOrder = events.map(event => event.messageId);
4

5
  // group events by type
6
  const eventsByType = {}
7
  for (const event of events) {
8
    if (!(event.type in eventsByType)) {
9
      eventsByType[event.type] = []
10
    }
11
    eventsByType[event.type].push(event)
12
  }
13

14
  // concurrently process sub-batches of a specific event type
15
  const promises = Object.entries(eventsByType).map(([type, events]) => {
16
    switch (type) {
17
    case 'track':
18
      return onTrackBatch(events, settings)
19
    case 'identify':
20
      return onIdentifyBatch(events, settings)
21
    // ...handle other event types here...
22
    }
23
  })
24
  try {
25
    const results = await Promise.all(promises);
26
    const batchResult = [].concat(...results); // Combine arrays into a single array
27
    
28
    // restore original order
29
    const resultMap = new Map(batchResult.map(e => [e.messageId, e]));
30
    return originalOrder.map(id => resultMap.get(id));
31
  } catch (error) {
32
    throw new RetryError(error.message);
33
  }
34
}
35

36
async function onTrackBatch(events, settings) {
37
  // handle a batch of track events
38
  return events
39
}
40

41
async function onIdentifyBatch(events, settings) {
42
  // handle a batch of identify events
43
  return events
44
}

By default, Functions waits up to 10 seconds to form a batch of 20 events. You can increase the number of events included in each batch (up to 400 events per batch) by contacting Segment support. Segment recommends users who wish to include fewer than 20 events per batch use destination insert functions without the onBatch handler.

The Functions editing environment supports testing batch handlers.

To test the batch handler:

1. In the right panel of the Functions editor, click customize the event yourself to enter Manual Mode.
2. Add events as a JSON array, with one event per element.
3. Click Run to preview the batch handler with the specified events.

The editor displays logs and request traces from the batch handler.

The Public API Functions/Preview endpoint also supports testing batch handlers. The payload must be a batch of events as a JSON array.

Events in a batch can be filtered out using custom logic. The filtered events will be surfaced in the Event Delivery page with reason as Filtered at insert function

1
async function onBatch(events, settings) {
2
  let response = [];
3
  try {
4
    for (const event of events) {
5
      // some business logic to filter event. Here filtering out all the events with name `drop`
6
      if (event.properties.name === 'drop') {
7
        continue;
8
      }
9

10
      // some enrichments if needed
11
      event.properties.message = "Enriched from insert function";
12

13
      // Enriched events are pushed to response
14
      response.push(event);
15
    }
16
  } catch (error) {
17
    console.log(error)
18
    throw new RetryError('Failed function', error);
19
  }
20

21
  // return a subset of transformed event
22
  return response;
23
}

Standard function error types apply to batch handlers. Segment attempts to retry the batch in the case of Timeout or Retry errors. For all other error types, Segment discards the batch.

• Bad Request - Any error thrown by the function code that is not covered by the other errors.
• Invalid Settings - A configuration error prevented Segment from executing your code. If this error persists for more than an hour, contact Segment Support.
• Message Rejected - Your code threw InvalidEventPayload or ValidationError due to invalid input.
• Unsupported Event Type - Your code doesn't implement a specific event type (for example, onTrack()) or threw an EventNotSupported error.
• Retry - Your code threw RetryError indicating that the function should be retried.

Segment only attempts to send the event to your destination insert function again if a Retry error occurs.

You can view Segment's list of Integration Error Codes for more information about what might cause an error.

If your function throws an error, execution halts immediately. Segment captures the event, any outgoing requests/responses, any logs the function might have printed, as well as the error itself.

Segment then displays the captured error information in the Event Delivery page for your destination. You can use this information to find and fix unexpected errors.

You can throw an error or a custom error and you can also add helpful context in logs using the console API. For example:

1
async function onTrack(event, settings) {
2
  const userId = event.userId
3

4
  console.log('User ID is', userId)
5

6
  if (typeof userId !== 'string' || userId.length < 8) {
7
    throw new ValidationError('User ID is invalid')
8
  }
9

10
  console.log('User ID is valid')
11
}

Functions execute only in response to incoming data, but the environments that functions run in are generally long-running. Because of this, you can use global variables to cache small amounts of information between invocations. For example, you can reduce the number of access tokens you generate by caching a token, and regenerating it only after it expires. Segment cannot make any guarantees about the longevity of environments, but by using this strategy, you can improve the performance and reliability of your Functions by reducing the need for redundant API requests.

This example code fetches an access token from an external API and refreshes it every hour:

1
const TOKEN_EXPIRE_MS = 60 * 60 * 1000 // 1 hour
2
let token = null
3
async function getAccessToken () {
4
  const now = new Date().getTime()
5
  if (!token || now - token.ts > TOKEN_EXPIRE_MS) {
6
    const resp = await fetch('https://example.com/tokens', {
7
      method: 'POST'
8
    }).then(resp => resp.json())
9
    token = {
10
      ts: now,
11
      value: resp.token
12
    }
13
  }
14
  return token.value
15
}

Functions have specific roles which can be used for access management in your Segment workspace.

Access to functions is controlled by two permissions roles:

• Functions Admin: Create, edit, and delete all functions, or a subset of specified functions.
• Functions Read-only: View all functions, or a subset of specified functions.

You also need additional Source Admin permissions to enable source functions, connect destination functions to a source, or to deploy changes to existing functions.

If you are a Workspace Owner or Functions Admin, you can manage your function from the Functions page.

Yes, Functions access is logged in the Audit Trail, so user activity related to functions appears in the logs.

Yes, Segment retries invocations that throw RetryError or Timeout errors (temporary errors only). Segment's internal system retries failed functions API calls for four hours with a randomized exponential backoff after each attempt. This substantially improves delivery rates.

Retries work the same for both functions and cloud-mode destinations in Segment.

No, Segment can't guarantee the order in which the events are delivered to an endpoint.

No, specifying an endpoint is not always required for insert functions. If your function is designed to transform or filter data internally—such as adding new properties to events or filtering based on existing properties—you won't need to specify an external endpoint.

However, if your function aims to enrich event data by fetching additional information from an external service, then you must specify the endpoint. This would be the URL of the external service's API where the enriched or captured data is sent.

No, Destination Insert Functions are currently available for use with Cloud Mode (server-side) destinations only. Segment is in the early phases of exploration and discovery for supporting customer web plugins for custom Device Mode destinations and other use cases, but this is unsupported today.

Insert Functions are only supported by Cloud Mode (server-side) destinations and aren't compatible with Storage destinations.

Yes, you can connect an insert function to multiple destinations.

No, you can only connect one insert function to a destination.

Yes, you can have both destination filters and destination insert functions in the same connection.

Segment's data pipeline applies Destination Filters before invoking Insert Functions.

There is an 120-Character limit for the insert function display name.

This error occurs because your insert function code might not be handling all event types (Page, Track, Identify, Alias, Group) that your destination supports. When these unlisted events pass through the function, they are rejected with the "Unsupported Event Type" error.

To resolve this, verify your insert function includes handlers for all expected event types and returns the event object for each. Here's an example of how you can structure your insert function to handle all event types:

1
async function onTrack(event, settings) {
2
    //Return event to handle page event OR Your existing code for track event
3
    return event;
4
}
5

6
async function onPage(event, settings) {
7
    //Return event to handle page event OR Your existing code for track event
8
    return event;
9
}
10

11
async function onIdentify(event, settings) {
12
    //Return event to handle page event OR Your existing code for track event
13
    return event;
14
}
15

16
async function onAlias(event, settings) {
17
    //Return event to handle page event OR Your existing code for track event
18
    return event;
19
}
20

21
async function onGroup(event, settings) {
22
  //Return event to handle page event OR Your existing code for track event
23
    return event;
24
}
25

26
// Ensure that all expected event types are included in your function

By including handlers for all the major event types, you ensure that all supported events are processed correctly, preventing the "Unsupported Event Type" error. Always test your updated code before implementing it in production.

Only cloud-mode event destinations support destination Insert Functions. List destinations aren't compatible.

The test function interface has a 4KB console logging limit. Outputs surpassing this limit won't be visible in the user interface.