Campaign A/B Test Sends — /v5/campaign/mail/sends_abtest

Overview

The A/B Test Campaigns API lets you split a contact list into test groups, send different email variants to each group, and automatically deliver the winning version — determined by open rate or click rate — to the remaining contacts after the test period. Base URL: https://api.effilink.co
Endpoint: POST /v5/campaign/mail/sends_abtest You can define up to 4 additional test versions (5 total including the main version). Each version can vary its subject line, content, or both.

Request Parameters

string

required

The name of the A/B test task. Maximum 200 bytes. Must be unique within your team.

string

required

The subject line for the main (control) version of the email.

string

required

The HTML body content for the main (control) version of the email.

string

required

The display name shown to recipients in the “From” field.

string

required

The verified sender email address. Must be an address already verified in your EffiLink account.

string

The reply-to email address. If omitted, replies go to senderMail.

string

Scheduled send time in ISO 8601 UTC format (e.g., 2025-06-01T10:00:00Z). Must not be more than 1 hour in the past or more than 30 days in the future. Omit to send immediately.

array[string]

required

An array of target contact list names. The combined total of contacts across all lists must be at least 10.

["All Customers"]

array[string]

An array of exclusion contact list names. Recipients on these lists will be excluded from the send.

array[object]

required

An array of test variant objects, defining 1 to 4 additional versions beyond the main version (5 total maximum). At least one version must be provided.

Show abTestVersions[] properties

string

The subject line for this variant. Cannot be empty if content is also empty.

string

The HTML body content for this variant. Cannot be empty if subject is also empty.

Each version must have at least one of subject or content set. A version with both fields empty is invalid.

integer

required

The duration of the A/B test in hours, after which the winning version is determined and sent to remaining contacts. Accepted range: 1–100.

integer

required

The percentage of the total contact list assigned to each individual test version. All test versions share the same percentage value, and the total allocation (percent × number of versions) must not exceed 100%. The remaining contacts receive the winning version after the test period.For example, with 2 versions (main + 1 variant) at percent: 20, each version reaches 20% of the list (40% total). The winning version is then sent to the remaining 60%.

string

The metric used to determine the winning version after the test period. Defaults to open.

Value	Winning metric
`open`	Highest unique open rate
`click`	Highest unique click rate

Response

integer

HTTP-style status code. 200 on success.

string

A human-readable status message. Empty string on success.

Error Codes

Code	Cause
`400`	Missing a required field (`mailName`, `subject`, `content`, `senderName`, `senderMail`, `sendListNames`, `abTestVersions`, `timeLimit`, or `percent`)
`400`	`mailName` exceeds 200 bytes
`400`	`sendDate` is not valid ISO 8601 UTC format
`400`	`sendListNames` is empty or combined contacts total fewer than 10
`400`	`abTestVersions` is empty
`400`	Total versions (main + variants) exceed 5
`400`	All provided `abTestVersions` have both `subject` and `content` empty
`400`	`timeLimit` is outside the valid range of 1–100 hours
`400`	`percent` × total number of versions (main + variants) exceeds 100%
`400`	`abTestType` is not `open` or `click`
`403`	`sendDate` is more than 1 hour in the past
`403`	`sendDate` is more than 30 days in the future
`500`	Internal server error during task creation

How A/B Testing Works

Divide — EffiLink splits your contact list into test groups. Each version (including the main) receives percent% of the total list.
Test — All versions are sent simultaneously at sendDate. The test runs for timeLimit hours.
Winner — After the test period, the version with the best abTestType metric (open rate or click rate) is declared the winner.
Deliver — The winning version is sent automatically to all remaining contacts who did not receive a test version.

For statistically meaningful results, ensure your test group sizes are large enough. With percent: 10 and a list of 1,000 contacts, each variant receives only 100 recipients — consider using larger lists or higher percentages for low-volume tests.

Example Request

curl --request POST \
  --url https://api.effilink.co/v5/campaign/mail/sends_abtest \
  --header 'Content-Type: application/json' \
  --header 'ApiKey: YOUR_API_KEY' \
  --data '{
    "mailName": "Subject Line A/B Test - June",
    "subject": "Check out our latest products",
    "content": "<p>Browse our new collection...</p>",
    "senderName": "Your Company",
    "senderMail": "marketing@yourdomain.com",
    "sendListNames": ["All Customers"],
    "abTestVersions": [
      {"subject": "New arrivals you'\''ll love this June"}
    ],
    "timeLimit": 24,
    "percent": 20,
    "abTestType": "open"
  }'

In this example:

Main version (subject: "Check out our latest products") → sent to 20% of the list.
Variant A (subject: "New arrivals you'll love this June") → sent to another 20%.
After 24 hours, the version with the higher open rate is sent to the remaining 60%.

Example Response

{
  "code": 200,
  "message": ""
}

Transactional Email

Campaign Email

SMS

Contacts & Lists

Webhooks

Authentication

Campaign A/B Test Sends — /v5/campaign/mail/sends_abtest

Overview

Request Parameters

Response

Error Codes

How A/B Testing Works

Example Request

Example Response

​Overview

​Request Parameters

​Response

​Error Codes

​How A/B Testing Works

​Example Request

​Example Response

Overview

Request Parameters

Response

Error Codes

How A/B Testing Works

Example Request

Example Response