Script 121: Strategy Assignment

Purpose

The Python script assigns marketing campaigns to specific strategies based on campaign name, creation date, and accumulated clicks.

To Elaborate

The script is designed to automate the assignment of marketing campaigns to predefined strategies. It uses specific criteria such as the campaign name, the date the campaign was created, and the number of clicks the campaign has accumulated. The script applies a set of business rules to determine the appropriate strategy for each campaign. These rules include checking for specific tokens in the campaign name to assign initial strategies, evaluating the age of the campaign and its performance in terms of clicks to potentially reassign it to a different strategy. The goal is to ensure that campaigns are aligned with the most suitable strategy based on their characteristics and performance metrics.

Walking Through the Code

  1. Initialization and Setup
    • The script begins by defining constants for column names and strategies. These constants are used throughout the script to reference specific data fields and strategy values.
    • Strategies are predefined and mapped to specific values, such as “SVOD - VN - Title_Exc-Launch - TIS” for exclusive campaigns.
  2. Strategy Assignment Rules
    • Rule 1: Campaigns containing specific tokens in their names (“_ Exc” or “_ NonExc”) are initially assigned to corresponding strategies. This is done using case-insensitive string matching.
    • Rule 2: Campaigns that have been active for more than 14 days and have accumulated at least 100 clicks are reassigned to the “CPA BAU” strategy. This rule applies to both exclusive and non-exclusive campaigns.
    • Rule 3: Campaigns older than 60 days are assigned to the “CPA Evergreen” strategy, regardless of their initial assignment.
  3. Output Preparation
    • The script identifies campaigns whose strategy has changed based on the applied rules.
    • It prepares an output DataFrame containing only the campaigns with changed strategies, ensuring that the output is ready for further processing or reporting. If no changes are detected, an empty output DataFrame is prepared.

Vitals

  • Script ID : 121
  • Client ID / Customer ID: 1306924441 / 69058
  • Action Type: Bulk Upload
  • Item Changed: Campaign
  • Output Columns: Account, Campaign, Strategy
  • Linked Datasource: M1 Report
  • Reference Datasource: None
  • Owner: Jeremy Brown (jbrown@marinsoftware.com)
  • Created by Jeremy Brown on 2023-05-19 11:13
  • Last Updated by Jonathan Reichl on 2023-12-13 11:21
> 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
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90

#
# Assign Campaign to Strategy according to:
#  - Campaign Name
#  - Campaign Creation Date
#  - accumulated clicks
#
# Author: Michael S. Huang
# Date: 2023-05-18
#


RPT_COL_CAMPAIGN = 'Campaign'
RPT_COL_ACCOUNT = 'Account'
RPT_COL_STRATEGY = 'Strategy'
RPT_COL_CAMPAIGN_CREATIONDATE = 'Campaign Creation Date'
RPT_COL_CLICKS = 'Clicks'
BULK_COL_ACCOUNT = 'Account'
BULK_COL_CAMPAIGN = 'Campaign'
BULK_COL_STRATEGY = 'Strategy'

outputDf[BULK_COL_STRATEGY] = "<<YOUR VALUE>>"

# define Strategies to map to
VAL_STRATEGY_IS_EXC = "SVOD - VN - Title_Exc-Launch - TIS"
VAL_STRATEGY_IS_NON_EXC = "SVOD - VN - Title_NonExc-Launch - TIS"
VAL_STRATEGY_CPA_BAU = "SVOD - VN - Title-BAU - CPS"
VAL_STRATEGY_CPA_EVERGREEN = "SVOD - VN - Title-Evergreen - CPS"

# define campaign tokens to match
VAL_CAMPAIGN_NAME_TOKEN_EXC = "_ Exc"
VAL_CAMPAIGN_NAME_TOKEN_NON_EXC = "_ NonExc"

# defines dates to check
print("timezone", CLIENT_TIMEZONE)
today = datetime.datetime.now(CLIENT_TIMEZONE).date()
date_14_days_ago = pd.to_datetime(today - datetime.timedelta(days=14))
date_60_days_ago = pd.to_datetime(today - datetime.timedelta(days=60))
print(today, date_14_days_ago, date_60_days_ago)

# define tmp column for new Strategy and set to empty
TMP_STRATEGY = RPT_COL_STRATEGY + '_'
inputDf[TMP_STRATEGY] = np.nan

### Rule 1: Assign to Exc/NonExc Strategy based on token in campaign name
###  case insensitive

inputDf.loc[ inputDf[RPT_COL_CAMPAIGN].str.contains(VAL_CAMPAIGN_NAME_TOKEN_EXC, case=False), \
             TMP_STRATEGY \
           ] = VAL_STRATEGY_IS_EXC

inputDf.loc[ inputDf[RPT_COL_CAMPAIGN].str.contains(VAL_CAMPAIGN_NAME_TOKEN_NON_EXC, case=False), \
             TMP_STRATEGY \
           ] = VAL_STRATEGY_IS_NON_EXC

### Rule 2: if Exc/NonExc campaign created more than 14 days ago and clicks >= 100,
###    assign to CPA BAU

inputDf.loc[ ( inputDf[RPT_COL_CAMPAIGN].str.contains(VAL_CAMPAIGN_NAME_TOKEN_EXC, case=False) | \
               inputDf[RPT_COL_CAMPAIGN].str.contains(VAL_CAMPAIGN_NAME_TOKEN_NON_EXC, case=False) ) & \
             (inputDf[RPT_COL_CAMPAIGN_CREATIONDATE] <= date_14_days_ago) & \
             (inputDf[RPT_COL_CLICKS] >= 100), \
             TMP_STRATEGY \
           ] = VAL_STRATEGY_CPA_BAU

### Rule 3: if campaign created more than 60 days ago,
###    assign to CPA Evergreen

inputDf.loc[ (inputDf[RPT_COL_CAMPAIGN_CREATIONDATE] <= date_60_days_ago), \
             TMP_STRATEGY \
           ] = VAL_STRATEGY_CPA_EVERGREEN

# find changed campaigns
changed = inputDf[TMP_STRATEGY].notnull() & (inputDf[RPT_COL_STRATEGY] != inputDf[TMP_STRATEGY])

# put changed campaigns into outputDf; if none, prepare empty outputDf
if sum(changed) > 0:
  print("== Campaigns with Changed Strategy ==", tableize(inputDf.loc[changed]))

  # only select changed rows
  cols = [RPT_COL_ACCOUNT, RPT_COL_CAMPAIGN, TMP_STRATEGY]
  outputDf = inputDf.loc[ changed, cols ].copy() \
                    .rename(columns = { \
                      TMP_STRATEGY: BULK_COL_STRATEGY \
                    })
  print("outputDf", tableize(outputDf))

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

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

comments powered by Disqus