Derived Events

Derived events are synthetic events generated by Blueshift when certain criteria are met. You can use a combination of customer metadata, transactions, browsing behavior, and catalog/content changes to specify the criteria to trigger such events. 

You can then use these derived events in event triggered campaigns or use them as filters to segment customers for segment triggered campaigns.

Use Cases

Here are some examples of use cases for derived events:

Catalog driven

To generate catalog driven events, you must first upload a catalog of items (e.g. products, content etc.) to Blueshift.

• Price drop alert - Notify customers about a price drop for an item they have viewed but not purchased.
• Back in stock notification - Notify customers that an out of stock item that they viewed within a preset time frame is now available.
• Promotion for frequently viewed items - Send promotional material to customers who frequently visit particular items in a catalog.
• New items in collection - Send promotional material to customers if there is a new item in a collection in which they have expressed interest.

Transaction data driven

To generate transaction driven events, you must first set up a transaction model in Blueshift.

• Subscription renewal - Send subscription renewal messages to customers who have subscriptions that are expiring in the next X days.
• Feedback reminder - Send feedback reminders to customers after a specific event if they haven't submitted a feedback in X days.

Include items from derived events in templates

For catalog driven alerts, you would need to include the catalog item in the notification that you send out to your customers. You can do this either by 

• Using a recommendation scheme to include the item and/or other related items. (This works for all types of campaigns.)

OR

• Directly using the item in the derived event context. (This works only for event triggered campaigns.)

Use recommendations

Create recommendation schemes for the required scenario. Blueshift’s recommendation studio includes recommendation recipes that you can use to configure the required schemes.

In the following example, the Retail - Price Drop recipe is used to create a price_drop_items block in the Price_drop_10 recommendation scheme to find catalog items for which there was a price change derived event for a viewed product and exclude those items which were purchased in the past 30 days.

In the following example, the previously created recommendation scheme Price_drop_10 is used in a template to recommend items with a price drop. For more information about using recommendations in templates, see Activate recommendations.

Use item information from event

To use data from the derived event, complete the following steps:

1. Go to the Data tab of the template and go to the Preview Data section.
2. Select Event and then select the required derived event. For example, the price_change.view event in the screenshot below. You can see the event JSON in the Data panel on the right.
   Note: You must select the correct preview user to be able to see the event data for the user.
3. In the data section, expand bsft_event_context and then products. This section contains the details for the items that are included in the derived event. You can use this item information in your templates. 

For more information, see Personalization in messages.

Create segments based on derived events

You can create segments based on derived events and then use the segments in either one-time, recurring, or segment triggered campaigns.

You can find the derived events in the Recent Activity tab.

Example 1: The following example shows a segment of users for whom the price of an item they viewed dropped (i.e. there was a price drop derived event) by 10% or more in the last 7 days.

Create campaigns based on derived events

Segment Based Campaigns

You can use a segment that uses a derived event (see above) to create a segment triggered/ one time or recurring campaign. 

Event Triggered Campaigns

You can use the derived event to trigger a campaign. You can use the Triggering event attributes to fine tune the audience for the campaign.

Example 1: This campaign is triggered when the price change event occurs for a viewed product. The Triggering event attribute for percentage price change is used to further filter customers down to only those who experienced a price change of at least 10%.