Automations

Automations is an extension module currently in beta phase. Get in touch with your mediarithmics contact if you want further information on how to enable it.

Triggered by user activity, automations let you launch predefined sequences of actions.

Required changes to the schema

Using automations might require some changes to your schema.

You need to make sure your UserPoint object has a scenarios property referring to a UserScenario object configured as follows:

type UserPoint @TreeIndexRoot(index:"USER_INDEX"){
  id: ID!
  creation_ts: Timestamp! @TreeIndex(index:"USER_INDEX")
  agents: [UserAgent!]!
  accounts: [UserAccount!]!
  emails: [UserEmail!]!
  activities: [UserActivity!]!
  profiles: [UserProfile!]!
  segments: [UserSegment!]!
  choices: [UserChoice!]!
  scenarios: [UserScenario!]!
}

type UserScenario {
  id: ID! @TreeIndex(index:"USER_INDEX")
  scenario_id: String! @TreeIndex(index:"USER_INDEX")
  execution_id: String! @TreeIndex(index:"USER_INDEX")
  node_id: String! @TreeIndex(index:"USER_INDEX")
  callback_ts: Timestamp @TreeIndex(index:"USER_INDEX")
  start_ts: Timestamp! @TreeIndex(index:"USER_INDEX")
  node_start_ts: Timestamp! @TreeIndex(index:"USER_INDEX")
  active: Boolean @TreeIndex(index:"USER_INDEX")
}

Types of automation nodes

Automations are composed of nodes, linked by edges or branches. During an automation, a user will go from node to node along edges until the scenario is over. According to their type, nodes can have 0 to 1 edges in and 0 to n edges out.

Input nodes

Input nodes are always at the start of an automation (each automation must and can have only one entry node). They define what will trigger the entry of a user in an automation.

On segment entry input nodes

If the scenario starts with this type of node, the scenario will be a On segment entry type of scenario (ON_SEGMENT_ENTRY_INPUT_NODE), where the user point enters the scenario if he enters a specified segment.

The node configuration is associated with a User Query or User List segment.

On segment exit input nodes

The scenario will be a On segment exit type of scenario (ON_SEGMENT_EXIT_INPUT_NODE), where the user point enters the scenario if he leaves a specified User Query segment.

The node configuration is associated with a User Query segment.

Query input nodes

The scenario will have a React to an event trigger (QUERY_INPUT), associated with a specified query. If an activity received matches the defined query, then the user point will enter the scenario.

The node configuration is associated with a query.

Only fields that have the directive @TreeIndex in the runtime schema can be used for these queries.

These queries is evaluated in the ingestion pipeline onto incoming activities, after session aggregation and processing by activity analyzers, before storage within the datamart.

If a user enters a scenario, any attempt to make the user enter the scenario again will fail until the user exits the scenario.

Action nodes

Automations support an array of action nodes. Once a user goes through an action node during a scenario, the action will be triggered asynchronously and the user will move on to the next node.

Add to segment nodes

If a user goes through an Add-to-segment node (ADD_TO_SEGMENT_NODE), he/she will be added to the configured segment. Add-to-segment nodes are associated with a User List segment.

Remove from segment nodes

If a user goes through a Remove from segment node (REMOVE_FROM_SEGMENT_NODE), he/she will be removed from the configured segment (if he/she is in segment, if not, nothing will happen). Remove from segment nodes are associated with a User List segment.

Send email nodes

If a user goes through a Send email node (EMAIL_CAMPAIGN), the platform will attempt to send an email using the information contained in the node configuration:

• Email router. The router which will be used to send your email.
• Provider. The opt-in provider helps you target users who have given you explicit consent to email targeting.
• Email renderer (optional). The renderer can be used to customize the email template.

You can manage frequency capping settings on the send email nodes so that users do not receive too many emails. This is done by updating the capping, capping_period and min_period_between_shoots properties in the email_blast resource created by automations.

Update email blast

PUT https://api.mediarithmics.com/v1/email_campaigns/:emailCampaignId/email_blasts/:emailBlastId

If you just want to update the capping information, look for the GET /v1/email_campaigns/:emailCampaignId/email_blasts in the inspector of your browser when opening an automations scenario, and copy the emailCampaignId (for the URL), emailBlastId (from the payload), as well as all other properties that do not need changing. Here only the properties related to frequency capping are shown:

Path Parameters

Name	Type	Description
emailBlastId	string	The ID for the email blast.
emailCampaignId	string	The ID for the email campaign.

Request Body

Name	Type	Description
min_period_between_shoots	string	The minimum period of time between two emails in the same capping period. Using ISO 8601 duration specification (for instance, "P1D" = 1 day, "P2M" = 2 months).
capping_period	string	The period to consider for capping. Using ISO 8601 duration specification.
capping	string	Maximum number of emails sent in this blast for the defined capping period.

200

{
    "status": "ok",
    "data": {
        "id": "string",
        "blast_name": "string",
        "subject_line": "string",
        "from_email": "string",
        "from_name": "string",
        "reply_to": "string",
        "send_date": 0,
        "status": "SCENARIO_ACTIVATED",
        "capping": 1,
        "capping_period": "P7D",
        "min_period_between_shoots": "P1D",
        "batch_size": 0,
        "number_mail_not_send": 0,
        "campaign_id": "string"
    }
}

Send email nodes are created with a default capping value of one email every 13 months.

Custom action nodes

If a user goes through a Custom action node (CUSTOM_ACTION_NODE), the platform will trigger an execution of the Custom action plugin set in the node configuration.

To make the custom action plugin available in the automations UI, you need to create an instance of your plugin attached to your organisation.

Custom actions for automations are currently in alpha phase. Get in touch with your mediarithmics contact if you want further information on how to enable them.

Feed nodes

If a user goes through a Feed node (SCENARIO_AUDIENCE_SEGMENT_FEED_NODE), the platform will trigger an execution of the Audience segment external feed plugin set in the node configuration.

Feed nodes rely on implicit audience segments, which results in basic frequency capping being implemented out of the box: if a same user point enters twice a feed node, it will be only activated once, unless it has exited the segment before the second passage.

Feed components are only visible when you have presets configured.

Feeds for automations are currently in alpha phase. Get in touch with your mediarithmics contact if you want further information on how to enable them.

Flow control nodes

Flow control nodes modify the normal flow of an automation.

Wait nodes

If a user goes through a Wait node (WAIT_NODE), he/she will be put on hold until wait time is over. Wake up is stored in the callback_ts property of UserScenario.

If nodes

If a user goes through a If node (IF_NODE), he/she will be checked against the associated query and will be sent into the "true" branch or the "false" branch accordingly.

Split nodes

If a user goes through a Split node (ABN_NODE), he/she will be sent in one of the outgoing branches on a random basis.

When sent several times through the same scenario, a same user will always go through the same split node branch.

Exit nodes

Exit nodes mark the exit of an automation. All branches must end with an exit node.

Exit conditions

Exit conditions let you define events which -when received for a user who is currently going through a scenario- will cause the user to exit at once. Exit conditions are stored as queries.

Only fields that have the directive @TreeIndex in the runtime schema can be used for these queries.

These queries are evaluated in the ingestion pipeline onto incoming activities, after session aggregation and processing by activity analyzers, before storage within the datamart.