Script 113: 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 accumulated by the campaign to determine the appropriate strategy. The script applies a set of business rules to categorize campaigns into different strategies, ensuring that campaigns are managed effectively and resources are allocated efficiently. This process helps in maintaining a structured budget allocation (SBA) by aligning campaigns with the most suitable strategies based on their performance and characteristics.

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 include options like “SVOD - MY - Title_Exc-Launch - TIS” and “SVOD - MY - Title-Evergreen - CPS”.
  2. Rule Application
    • Rule 1: Campaigns are assigned to either an “Exc” or “NonExc” strategy based on tokens found in their names. This is done using a case-insensitive search for specific substrings in the campaign name.
    • Rule 2: If a campaign with an “Exc” or “NonExc” token was created more than 14 days ago and has accumulated at least 100 clicks, it is reassigned to the “CPA BAU” strategy.
    • Rule 3: Campaigns created more than 60 days ago are assigned to the “CPA Evergreen” strategy.
  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 updated strategies, ensuring that only relevant changes are captured for further processing. If no changes are detected, an empty DataFrame is prepared.

Vitals

  • Script ID : 113
  • Client ID / Customer ID: 1306924439 / 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 11:07
  • Last Updated by Jonathan Reichl on 2023-12-13 11:19
> 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 - MY - Title_Exc-Launch - TIS"
VAL_STRATEGY_IS_NON_EXC = "SVOD - MY - Title_NonExc-Launch - TIS"
VAL_STRATEGY_CPA_BAU = "SVOD - MY - Title-BAU - CPS"
VAL_STRATEGY_CPA_EVERGREEN = "SVOD - MY - 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