Purpose

The Python script assigns dimension values to campaigns based on whether they are categorized as “Brand,” “Local,” or “Emergency.”

To Elaborate

The script processes a dataset of marketing campaigns to categorize each campaign into one of three dimensions: “Brand,” “Local,” or “Emergency.” This categorization is based on the presence of these keywords in the campaign names. The script extracts these keywords and assigns them to a new column in the dataset. It then identifies which campaigns have been categorized and prepares a subset of the data containing only these categorized campaigns. If no campaigns are categorized, it prepares an empty dataset. This process helps in organizing and analyzing marketing campaigns based on their strategic focus.

Walking Through the Code

1. Define Constants and Patterns
   • The script begins by defining constants for column names used in the dataset, such as RPT_COL_CAMPAIGN, RPT_COL_ACCOUNT, and RPT_COL_BRAND_LOCAL_EMERGENCY.
   • A regular expression pattern PATTERN1 is defined to match the keywords “Brand,” “Local,” or “Emergency” in the campaign names.
2. Extract and Assign Dimension Values
   • The script uses the str.extract() method to find occurrences of the defined pattern in the Campaign column and assigns these values to the Brand | Local | Emergency column. If no match is found, it fills the column with a space.
3. Identify and Select Changed Campaigns
   • It checks for changes in the Brand | Local | Emergency column to identify campaigns that have been categorized.
   • The script selects only the relevant columns (Campaign, Account, and Brand | Local | Emergency) for further processing.
4. Prepare Output DataFrame
   • If there are any categorized campaigns, they are stored in outputDf, and the data is printed.
   • If no campaigns are categorized, an empty outputDf is prepared.

Vitals

• Script ID : 535
• Client ID / Customer ID: 1306917127 / 60268084
• Action Type: Bulk Upload
• Item Changed: Campaign

• Output Columns: Account, Campaign, ___Brand

  Local

  Emergency___

• Linked Datasource: M1 Report
• Reference Datasource: None
• Owner: ascott@marinsoftware.com (ascott@marinsoftware.com)
• Created by ascott@marinsoftware.com on 2023-11-17 14:53
• Last Updated by ascott@marinsoftware.com 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
28
29
30
31
32
33
34
35
36
#
# Assign Campaign Dimension Values according to:
#  - Brand | Local | Emergency
#
# Author: Jeremy Brown & Adam Scott
# Date: 2023-11-17

RPT_COL_CAMPAIGN = 'Campaign'
RPT_COL_ACCOUNT = 'Account'
RPT_COL_BRAND_LOCAL_EMERGENCY = 'Brand | Local | Emergency'


# define pattern match
PATTERN1 = r'(Brand|Local|Emergency)'


# find all occurrences of the pattern in the Campaign name
inputDf[RPT_COL_BRAND_LOCAL_EMERGENCY] = inputDf[RPT_COL_CAMPAIGN].str.extract(PATTERN1).fillna(' ')


# find changed campaigns
changed = inputDf[RPT_COL_BRAND_LOCAL_EMERGENCY].ne(' ') | inputDf[RPT_COL_BRAND_LOCAL_EMERGENCY].ne(' ')

#selecting only relevant cols
inputDf= inputDf[[RPT_COL_CAMPAIGN,RPT_COL_ACCOUNT,RPT_COL_BRAND_LOCAL_EMERGENCY]]

# put changed campaigns into outputDf; if none, prepare empty outputDf
if sum(changed) > 0:
  #print('changed data',tableize(inputDf[changed]))
  outputDf = inputDf[changed]                 
  print("outputDf", tableize(outputDf))

else:
  print("Empty outputDf")
  outputDf = outputDf.iloc[0:0]##

Post generated on 2024-11-27 06:58:46 GMT