{% docs STG_AE_EVENTS_DBT %}

This table contains AE events. It is used to make dependant queries
more streamlined

{% enddocs %}

{% docs STG_ALL_FANS_ARTISTS_DBT %}

This table contains fan's first event date grouped by vendor, 
subaccount and artist(global participant)

{% enddocs %}

{% docs STG_AE_FEATURES_FOR_PREREQUISITES_DBT %}

This table collects attributes related to AE per artist

{% enddocs %}

{% docs STG_SEGMENTATION_PREREQUISITES_DBT %}

This table collects attributes related to determine if artist has enough 
high quality features to be considered for segmentation. This table
combines data from non AE related attributes and AE related 
attributes (STG_AE_FEATURES_FOR_PREREQUISITES_DBT)

{% enddocs %}

{% docs STG_FANS_FOR_SEGMENTATION_DBT %}

This table contains fans that are available for segmentation. This table excludes 
fans that are new(recently had first interaction with artist)

{% enddocs %}

{% docs STG_FANS_ENGINEERED_FEATURES_DBT %}

This table contains engineered feature based on Apple Music's AE activity with id 13703

{% enddocs %}


{% docs STG_NEW_FANS_DBT %}

This table contains fans and indicator if they have been recently acquired or not. 
Recently acquired fans will be segmented as new fans and excluded from segmentation.

{% enddocs %}

{% docs STG_FANS_POINT_CALCULATION_DBT %}

This table contains points scored per fan and feature.

{% enddocs %}

{% docs STG_ARTIST_BIN_INTERVALS_DBT %}

This table contains bin limits for each artist and segment. This is based on the points
scored from the fan related features.  Bin values are relevant so we can later assign
fans into proper segments based on their total score per artist. 
First approach uses logarithmic preprocessing and second approach doesn't use.
Bins are created by dividing maximum score - minimum score into 4 equal lengths.

{% enddocs %}

{% docs STG_UPDATED_ARTIST_BIN_INTERVALS_DBT %}

This table contains modifies bin limits.

{% enddocs %}

{% docs STG_FANS_LOGARITHMIC_BINNING_DBT %}

This table assigns fans into segments using logarithmically created bin limits.

{% enddocs %}

{% docs STG_FANS_PLAYLIST_FOLLOW_FEATURE_DBT %}

This table contains nr of AE activities with id 4830 or 4210 for fan.

{% enddocs %}

{% docs STG_FANS_TOTAL_POINTS_DBT %}

Table containing total points (with and without logarithmic transformation)
fan scored per vendor, subaccount and global participant

{% enddocs %}

{% docs STG_SEGMENTED_FANS_DBT %}

Table containing fans and related segments based on the proper segmentation approach

{% enddocs %}

{% docs STG_FANS_SHOPIFY_FEATURES_DBT %}

Table containing fan merchandise features for segmentation

{% enddocs %}

{% docs STG_POTENTIAL_WINBACK_INDICATOR_DBT %}

Table containing indicators if fans should be part of potential
or win back segments

{% enddocs %}

{% docs STG_SECONDARY_FANS_DBT %}

Table contains makes sure that fans with enough who still were exclude from segmentation will be
defined as secondary fans

{% enddocs %}

{% docs STG_FAN_POINT_OUTLIERS_DBT %}

Table contains fans whose points exceed 99ntile of total points per artist

{% enddocs %}

{% docs STG_SUPERFANS_FROM_OUTLIERS_DBT %}

This model creates super fan indicators for those fans who were selected during
outlier detection from table STG_FAN_POINT_OUTLIERS_DBT

{% enddocs %}

{% docs STG_SEGMENTATION_DECAY_INTERVALS_DBT %}

This model specifies date intervals used to determine what decay factor to apply to fan 
fan activities when calculating total points for segments.

{% enddocs %}

{% docs STG_FANS_WITHOUT_PROFILING_CONSENT_DBT %}

This model contains fans that were excluded from segmentation
because they hadn't givcen consent for analytics profiling.

{% enddocs %}

{% docs STG_FANS_BIT_ENGINEERED_FEATURES_DBT %}

This model contains engineered features for Bandsintown features
called rsvp and ticket click. We only store unique artist/user id
 value in Fansifter event table, that's why we need to collect all
relevant rows separately and apply extra logic.

{% enddocs %}