Mixpanel dbt Package (Docs)

Macros

analyze_funnel (source)

You can use the analyze_funnel(event_funnel, group_by_column, conversion_criteria) macro to produce a funnel between a given list of event types.

It returns the following:

• The number of events and users at each step
• The overall user and event conversion % between the top of the funnel and each step
• The relative user and event conversion % between subsequent steps

Note: The relative order of the steps is determined by their event volume, not the order in which they are input.

The macro takes the following as arguments:

• event_funnel: List of event types (not case sensitive).
  • Example: '['play_song', 'stop_song', 'exit']
• group_by_column: (Optional) A column by which you want to segment the funnel (this macro pulls data from the mixpanel__event model). The default value is None.
  • Example: group_by_column = 'country_code'.
• conversion_criteria: (Optional) A WHERE clause that will be applied when selecting from mixpanel__event.
  • Example: To limit all events in the funnel to the United States, you'd provide conversion_criteria = 'country_code = "US"'. To limit the events to only song play events to the US, you'd input conversion_criteria = 'country_code = "US"' OR event_type != 'play_song'.

Pivoting Out Event Properties

By default, this package selects the default columns collected by Mixpanel. However, you likely have custom properties or columns that you'd like to include in the mixpanel__event model.

If there are properties in the mixpanel.event.properties JSON blob that you'd like to pivot out into columns, add the following variable to your dbt_project.yml file:

vars:
  mixpanel:
    event_properties_to_pivot: ['the', 'list', 'of', 'property', 'fields'] # Note: this is case-SENSITIVE and must match the casing of the property as it appears in the JSON

Passthrough Columns

Additionally, this package includes all standard source EVENT columns defined in the staging_columns macro. You can add more columns using our passthrough column variables. These variables allow the passthrough fields to be aliased (alias) and casted (transform_sql) if desired, although it is not required. Data type casting is configured via a SQL snippet within the transform_sql key. You may add the desired SQL snippet while omitting the as field_name part of the casting statement - this will be dealt with by the alias attribute - and your custom passthrough fields will be casted accordingly.

Use the following format for declaring the respective passthrough variables:

vars:
  mixpanel:
    event_custom_columns:
      - name:           "property_field_id"
        alias:          "new_name_for_this_field_id"
        transform_sql:  "cast(property_field_id as int64)"
      - name:           "this_other_field"
        transform_sql:  "cast(this_other_field as string)"

Sessions Event Frequency Limit

The event_frequencies field within the mixpanel__sessions model reports all event types and the frequency of those events as a JSON blob via a string aggregation. For some users there can be thousands of different event types that take place. For Redshift and Postgres warehouses there currently exists a limit for string aggregations (up to 65,535). As a result, in order for Redshift and Postgres users to still leverage the event_frequencies field, an artificial limit is applied to this field of 1,000. If you would like to adjust this limit, you may do so by modifying the below variable in your project configuration.

vars:
  mixpanel:
    mixpanel__event_frequency_limit: 500 ## Default is 1000

Event Date Range

Because of the typical volume of event data, you may want to limit this package's models to work with a recent date range of your Mixpanel data (however, note that all final models are materialized as incremental tables).

By default, the package looks at all events since January 1, 2010. To change this start date, add the following variable to your dbt_project.yml file:

vars:
  mixpanel:
    date_range_start: 'yyyy-mm-dd' 

Note: This date range will not affect the number_of_new_users column in the mixpanel__daily_events or mixpanel__monthly_events models. This metric will be true new users.

Global Event Filters

In addition to limiting the date range, you may want to employ other filters to remove noise from your event data.

To apply a global filter to events (and therefore all models in this package), add the following variable to your dbt_project.yml file. It will be applied as a WHERE clause when selecting from the source table, mixpanel.event.

vars:
  mixpanel:
    # Ex: removing internal user
    global_event_filter: 'distinct_id != "1234abcd"'

Session Configurations

Session Inactivity Timeout

This package sessionizes events based on the periods of inactivity between a user's events on a device. By default, the package will denote a new session once the period between events surpasses 30 minutes.

To change this timeout value, add the following variable to your dbt_project.yml file:

vars:
  mixpanel:
    sessionization_inactivity: number_of_minutes # ex: 60

Session Pass-Through Columns

By default, the mixpanel__sessions model will contain the following columns from mixpanel__event:

• people_id: The ID of the user
• device_id: The ID of the device they used in this session
• event_frequencies: A JSON of the frequency of each event_type in the session

To pass through any additional columns from the events table to mixpanel__sessions, add the following variable to your dbt_project.yml file. The value of each field will be pulled from the first event of the session.

vars:
  mixpanel:
    session_passthrough_columns: ['the', 'list', 'of', 'column', 'names'] 

Session Event Criteria

In addition to any global event filters, you may want to disclude events or place filters on them in order to qualify for sessionization.

To apply any filters to the events in the sessions model, add the following variable to your dbt_project.yml file. It will be applied as a WHERE clause when selecting from mixpanel__event.

vars:
  mixpanel:

    # ex: limit sessions to include only these kinds of events
    session_event_criteria: 'event_type in ("play_song", "stop_song", "create_playlist")'

Lookback Window

Events can sometimes arrive late. For example, events triggered on a mobile device that is offline will be sent to Mixpanel once the device reconnects to wifi or a cell network. Since many of the models in this package are incremental, by default we look back 7 days to ensure late arrivals are captured while avoiding requiring a full refresh. To change the default lookback window, add the following variable to your dbt_project.yml file:

vars:
  mixpanel:
    lookback_window: number_of_days # default is 7

Changing the Build Schema

By default this package will build the Mixpanel staging models within a schema titled (<target_schema> + _stg_mixpanel) and Mixpanel final models within a schema titled (<target_schema> + mixpanel) in your target database. If this is not where you would like your modeled Mixpanel data to be written to, add the following configuration to your dbt_project.yml file:

models:
    mixpanel:
      +schema: my_new_schema_name # leave blank for just the target_schema
      staging:
        +schema: my_new_schema_name # leave blank for just the target_schema

Change the source table references

If an individual source table has a different name than the package expects, add the table name as it appears in your destination to the respective variable:

IMPORTANT: See this project's dbt_project.yml variable declarations to see the expected names.

vars:
    mixpanel_<default_source_table_name>_identifier: your_table_name 

Event De-Duplication Logic

Events are considered duplicates and consolidated by the package if they contain the same:

• insert_id (used for de-deuplication internally by Mixpanel)
• people_id (originally named distinct_id)
• type of event
• calendar date of occurrence (event timestamps are set in the timezone the Mixpanel project is configured to)

This is performed in line with Mixpanel's internal de-duplication process, in which events are de-duped at the end of each day. This means that if an event was triggered during an offline session at 11:59 PM and resent when the user came online at 12:01 AM, these records would not be de-duplicated. This is the case in both Mixpanel and the Mixpanel dbt package.

Fivetran offers the ability for you to orchestrate your dbt project through Fivetran Transformations for dbt Core™. Learn how to set up your project for orchestration through Fivetran in our Transformations for dbt Core setup guides.