Purpose:

The Python script automates the assignment of “Brand vs NonBrand” dimensions to campaigns based on their names.

To Elaborate

The script is designed to automate the process of tagging campaigns as either “Brand” or “Non-Brand” based on the presence of the word “Brand” in the campaign name. This is particularly useful for marketing and advertising teams who need to categorize their campaigns for reporting and analysis purposes. The script takes input data, checks each campaign name for the keyword “Brand,” and then assigns the appropriate dimension tag. This helps streamline the process of campaign categorization, ensuring consistency and saving time in manual tagging efforts.

Walking Through the Code

1. Setup and Constants Definition:
   • The script begins by defining constants for column names used in the input and output DataFrames. These constants are used to reference specific columns in the data, ensuring consistency and reducing errors in column name usage.
2. Campaign Name Check:
   • A boolean check (bcheck) is performed on the RPT_COL_CAMPAIGN column to identify campaign names containing the word “Brand.” This check is case-sensitive and excludes any missing values.
3. DataFrame Creation:
   • Two separate DataFrames are created based on the results of the bcheck.
   • brandDf is created for campaigns identified as “Brand,” and the BULK_COL_B_NB column is set to “Brand.”
   • nonbrandDf is created for campaigns not identified as “Brand,” and the BULK_COL_B_NB column is set to “Non-Brand.”
4. DataFrame Merging:
   • The script merges the brandDf and nonbrandDf DataFrames using an outer join to create a comprehensive outputDf that includes all campaigns with their respective “Brand vs NonBrand” tags.
5. Output Display:
   • Finally, the script prints the outputDf in a table format without displaying the index, providing a clear view of the tagged campaigns.

Vitals

• Script ID : 147
• Client ID / Customer ID: 200958251 / 5139
• Action Type: Bulk Upload
• Item Changed: Campaign
• Output Columns: Account, Campaign, Brand vs NonBrand
• Linked Datasource: M1 Report
• Reference Datasource: None
• Owner: Jonathan Reichl (jreichl@marinsoftware.com)
• Created by Jonathan Reichl on 2023-05-30 09:51
• Last Updated by Grégory Pantaine on 2023-12-06 04:01

> See it in Action

Python Code

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
#
# Automating "Brand vs NonBrand" Dimension Tagging
# Taken from Byron Porter's Fisher Investment account
# setup 2023-11-15 by Greg Pantaine
#

RPT_COL_CAMPAIGN = 'Campaign'
RPT_COL_ACCOUNT = 'Account'
BULK_COL_ACCOUNT = 'Account'
BULK_COL_CAMPAIGN = 'Campaign'
BULK_COL_B_NB = 'Brand vs NonBrand'

# Check if RPT_COL_CAMPAIGN column has 'Brand' in the campaign name
bcheck = (inputDf[RPT_COL_CAMPAIGN].str.contains('Brand', case=True, na=False))

# Use check to create two DataFrames. One for Brand and one for Non-Brand tagging
brandDf = inputDf.loc[bcheck, [BULK_COL_ACCOUNT, BULK_COL_CAMPAIGN, BULK_COL_B_NB]]
brandDf.loc[:, BULK_COL_B_NB] = 'Brand'

nonbrandDf = inputDf.loc[~bcheck, [BULK_COL_ACCOUNT, BULK_COL_CAMPAIGN, BULK_COL_B_NB]]
nonbrandDf.loc[:, BULK_COL_B_NB] = 'Non-Brand' 

# Merges the Brand and Non-Brand dataframes
outputDf = brandDf.merge(nonbrandDf, how='outer')

# Print the tableized version of the output DataFrame
print(outputDf.to_string(index=False))

Post generated on 2025-03-11 01:25:51 GMT