/
Transaction Level Dataset

Transaction Level Dataset

As part of the BI initiative, we have created datasets that compile logic across our various datasets.
The dataset mentioned below is the transaction-level dataset that would have service mode details for each transaction.

The goal is to map out any transaction-level metric as a column in this dataset.

 Data Dictionary

Here are the columns present in the data.

  • Databricks Table name :

    loyalty.reportbi.weekly_execution_transaction_level

Column Name

Datatype

Description

Source

Notes:

TH_FISCAL_YEAR

string

The fiscal year of the transaction date

STG.DERIVED_MASTER_TABLE_NEW

 

TH_CALENDAR_MONTH

int

The calendar month of the transaction date

STG.DERIVED_MASTER_TABLE_NEW

 

TH_FISCAL_WEEK

string

The fiscal week of the transaction date

STG.DERIVED_MASTER_TABLE_NEW

 

PERIOD_DT

date

Date of the transaction

STG.DERIVED_MASTER_TABLE_NEW

 

DAY_PART_CD

string

Daypart of transaction

STG.DERIVED_MASTER_TABLE_NEW

 

TICKET_ID

bigint

Ticket_id, unique for each transaction

STG.DERIVED_MASTER_TABLE_NEW

 

OPERATOR

string

GM of the restaurant

TLOG_DIM_RESTAURANT_CURRENT

 

AFL

string

Director of the restaurant

TLOG_DIM_RESTAURANT_CURRENT

 

MFP

string

MFP of the restaurant

TLOG_DIM_RESTAURANT_CURRENT

 

LOYALTY_CUSTOMER_ID

string

Loyalty Short ID

STG.DERIVED_MASTER_TABLE_NEW

 

LOYALTY_FLAG

int

Flag, 1 or 0

STG.DERIVED_MASTER_TABLE_NEW

 

REGISTERED_ACCOUNT_ID

string

Registered ID 

STG.DERIVED_MASTER_TABLE_NEW

 

REGISTERED_FLAG

int

Registered flag, 1 or 0

STG.DERIVED_MASTER_TABLE_NEW

 

REST_NO

int

Resturant num

STG.DERIVED_MASTER_TABLE_NEW

 

CITY

string

City

STG.DERIVED_MASTER_TABLE_NEW

 

OPS_DIV_NM

string

OPS_DIV_NM

STG.DERIVED_MASTER_TABLE_NEW

 

REST_TYPE_NM

string

Resturant type

STG.DERIVED_MASTER_TABLE_NEW

 

SERVICE_MODE

string

mode of the purchase (SCAN_REG, MOP_REG , WHITELABEL_REG, KIOSK_REG, CATERING_REG, 3P_NON_LOYALTY, KIOSK_LOYALTY_UNREG, KIOSK_NON_LOYALTY, CATERING_LOYALTY_UNREG, CATERING_NON_LOYALTY, OTHER_REG, OTHER_LOYALTY_UNREG, OTHER_NON_LOYALTY) 

STG.DERIVED_MASTER_TABLE_NEW,
BRINGG REPORT

White_label restaturant count might be incorrect due to POS issue

SUB_SERVICE_MODE

string

sub category of the service mode (MOBILE ORDER DRIVE THRU', MOBILE ORDER EAT IN', MOBILE ORDER TAKE OUT, UBER, DOORDASH, SKIP THE DISHES, APP SCAN , PHYSICAL CARD SCAN, WHITELABEL_REG, KIOSK_REG, CATERING_REG, KIOSK_LOYALTY_UNREG, KIOSK_NON_LOYALTY, CATERING_LOYALTY_UNREG, CATERING_NON_LOYALTY, OTHER_REG, OTHER_LOYALTY_UNREG, OTHER_NON_LOYALTY)

STG.DERIVED_MASTER_TABLE_NEW,
TLOG.TLOG_SALE_TICKET_TENDERS,
DYDB.CARDS

Scan and Pay uses a different source of data as the other metrics, and is not classifed by itself on this table.

Related content

How to create your own charts from BI datasets
How to create your own charts from BI datasets
Read with this
Points Bank - Points Table [DONE]
Points Bank - Points Table [DONE]
Read with this
Weekly Execution - Report
Weekly Execution - Report
Read with this
Weekly Execution - Offers and Challenges
Weekly Execution - Offers and Challenges
Read with this
Business Intelligence - Project Plan
Business Intelligence - Project Plan
Read with this