Derived Attributes

Derived attributes are customer attributes that are generated or calculated by Blueshift based on other customer attributes and event data. These derived attributes are available on the Customer Profile and are refreshed at regular intervals. At the time of building a segment, you can select these in the segment builder. 

Standard derived attributes

Blueshift can compute certain attributes based on all the data (event, campaign) from a customer. 

Attribute	Definition
last_browser_platform	Most recent browser platform used by user (for example, webkit or blink).
last_browser_type	Most recent browser used by user (for example, chrome or safari).
last_browser_version	Version of browser Most recently used by user.
last_device_id	Most recent known device ID from user's mobile app.
last_device_token	Most recent device token sent from user's mobile app.
last_device_token_updated_at	Time when the device token was last updated.
last_utm_medium	The medium the link was used on such as, email, CPC, and so on.
last_utm_source	The source of your traffic. For example, Blueshift.
last_visit_at	Last known page load event. This attribute is updated when Blueshift receives an event for the customer and is updated once during a user session. The attribute is updated for a click event only if the click generates a page load event or any other event.
last_click_at	Last known click event.
last_open_at	Last known open event. Together with last_click_at, this attribute helps you to understand when a customer last engaged with your campaigns.
last_user_open_at	Timestamp of last open by user. This exclude open by prefetch or bots
last_user_click_at	Timestamp of last click by user. This excludes any bot clicks
session_last_activity_at	Last known session date.
unsubscribed_at	The time when a user unsubscribed from an email campaign.

Location attributes

Attribute	Definition
last_location_city	Most recent city for user based on latitude/longitude data.
last_location_country	Most recent country for user based on latitude/longitude data.
last_location_country_code	Most recent country code (ie +44 or +1) for user.
last_location_geo_delta_updated_at	Most recent time the geo delta was updated.
last_location_geo_latitude	Last known geo latitude of the user.
last_location_geo_longitude	Last known geo longitude of the user.
last_location_geo_updated_at	Most recent time the geo was updated.
last_location_state	Most recent state for user based on latitude/longitude data.
last_location_timezone	Last known timezone for the user.

Campaign attributes

Attribute	Definition
last_campaign_uuid	UUID of campaign Most recently received by user.
last_click_at	Most recent time a user clicked on a campaign.
last_click_campaign_uuid	UUID of campaign on which user Most recently clicked.
last_click_creative_uuid	Most recent creative UUID where click occurred.
last_click_experiment_uuid	Most recent experiment UUID where click occurred.
last_click_trigger_type	Type of trigger where last click occurred (for example, email, sms, and so on).
last_click_trigger_uuid	UUID of Most recent trigger where click occurred.
last_creative_uuid	Most recent creative a user received.
last_experiment_uuid	Most recent experiment sent to a user.
last_open_at	Most recent campaign open time.
last_open_campaign_uuid	Most recent campaign UUID tied to open.
last_open_creative_uuid	Most recent creative UUID tied to open.
last_open_experiment_uuid	Most recent experiment UUID tied to open.
last_open_message_uuid	Most recent message UUID tied to open.
last_open_trigger_type	Most recent trigger type tied to open (for example, email, sms, and so on).
last_open_trigger_uuid	Most recent trigger UUID tied to open.
last_purchase_at	Date when a user last purchased.  Derived from purchase event
last_trigger_uuid	UUID of most recent trigger sent to user.
last_utm_campaign	The campaign name. For example, Abandoned Browse.
last_utm_content	Optional parameter for additional details for A/B testing and content-targeted ads.
last_utm_medium	The medium the link was used on such as, email, CPC, and so on.
last_utm_source	The source of your traffic. For example, Blueshift.
last_utm_term	Optional parameter suggested for paid search to identify keywords for your ad.
unsubscribed_at	The time when a user unsubscribed from an email campaign.

Lifetime attributes

Attribute	Definition
last_purchase_at	Date when a user last purchased.  Derived from purchase event
lifetime_revenue	Computed and aggregated from purchase event. Requires revenue as an event attribute.
lifetime_orders	Computed and aggregated from purchase event.
lifetime_visits	Computed from session data. Each session lasts 30 minutes.
lifetime_(custom goal)	Computed from custom event goals.

Custom derived attributes

Using the custom derived attributes feature, you can compute complex attributes like counts, min and max values, averages, percentages, and so on.

For example, you can calculate the total number of orders for customers, find out the last time a person visited a physical store to purchase an item, the unique visitor count for a store and so on. You can also add conditions using event attributes and fine tune the derived attributes. For example, the number of shows watched using the app or the favorite brand of a customer with high lifetime value.

You can then use these derived attributes to create highly focused segments or as filters in campaigns.

Functions

You can use the following functions to calculate a derived attribute in Blueshift.

• Count
• Sum
• Average
• Max/Min
• First/Last

Count

The count for a particular event over a specific period of time.

For example, the number of comedy shows watched. This count is derived from the custom event playback_complete or playback_paused using the event attributes show_name and show_category. This count along with counts of other categories of shows can help you to determine which shows to promote to the customer.

Other examples:

• Number of movies watched in the last year
• Items purchased within past 90 days
• Items returned within past 90 days
• Page views for a particular item within past 7 days

Sum

The computed sum for a particular event over a specific period of time.

For example, the lifetime revenue for a customer. This sum is derived from the revenue attribute that is passed in the order event. You can use this to promote a loyalty program. We can also add conditions to identify favorite brands which will enable you to promote items from the same brand or promote a similar new brand.

Other examples:

• The total revenue from orders in the past 7 days
• The total viewing time in the past 30 days
• Total visits in the past 7 days.
• Total orders in the past 30 days.

Average

The average value for a particular event over a specific period of time.

For example, the average number of minutes watched for a particular show in the past 30 days. This average is derived from the custom event playback_complete or playback_paused using the event attribute minutes_watched. We can also add conditions to identify the channel so that only shows watched through the app will be considered.

Other examples:

• Average order value (Revenue/No. of orders) in the past 90 days
• Average playback time per session

Max/Min

The maximum or minimum value for an attribute over a specific period of time.

For example, the maximum order value for the last 30 days. This amount  is derived from the revenue attribute that is passed in the order event. You can use this value to offer a discount to customers who have spent more than a certain amount.

Other examples:

• The maximum order quantity for a particular item
• The minimum watch time for a show

First/Last

The first or the last occurrence of an event.

For example, Last time media viewed. This time is derived from the custom event playback_complete or playback_paused. You can use this to encourage or incentivize customers who have not visited the app in a while to come back. Another way this could be used is to add conditions to identify the channel so that you can encourage customers who are using the website to use the app instead.

Other examples:

• First purchase date
• Last purchase date
• First time customer purchased In Store
• Last time customer purchased In Store
• First time customer purchased in eCommerce store
• Last time customer purchased in eCommerce store