Purpose:

The Python script evaluates advertising campaigns to determine if they meet specific criteria for switching to a new publisher bidding strategy based on minimum ROAS and spend thresholds.

To Elaborate

The Python script is designed to optimize advertising campaigns by promoting them to a new publisher bidding strategy if they meet certain performance criteria. Specifically, it checks if the campaigns have achieved a minimum Return on Advertising Spend (ROAS) threshold. If the ROAS is equal to or greater than the specified minimum, the campaign’s bidding strategy is updated to a target strategy, in this case, ‘TargetCpa’. The script also updates the target CPA value based on the cost per conversion. This process helps in ensuring that only campaigns that are performing well are promoted to more aggressive bidding strategies, potentially improving overall advertising efficiency and effectiveness.

Walking Through the Code

1. Configurable Parameters:
   • The script begins by defining configurable parameters, including MIN_ROAS, which sets the minimum ROAS threshold, and PUB_STRATEGY_TARGET, which specifies the new bidding strategy to be applied if the criteria are met.
2. Data Preparation:
   • Temporary columns are created in the input DataFrame to store new values for the publisher strategy and target CPA. These columns are initialized with np.nan.
3. Criteria Evaluation:
   • The script evaluates each campaign’s ROAS against the MIN_ROAS threshold. If a campaign meets or exceeds this threshold, its publisher strategy is updated to the PUB_STRATEGY_TARGET.
4. Target CPA Update:
   • The target CPA value is updated based on the campaign’s cost per conversion.
5. Campaign Change Detection:
   • The script identifies campaigns where the publisher strategy has changed by comparing the original and updated strategy values.
6. Output Preparation:
   • If any campaigns have changed, the script prepares an output DataFrame containing only the changed campaigns with relevant columns. This DataFrame is then printed for review.
7. Empty Output Handling:
   • If no campaigns meet the criteria for strategy change, an empty DataFrame is prepared and printed, indicating no changes were made.

Vitals

• Script ID : 237
• Client ID / Customer ID: 1306922587 / 2
• Action Type: Bulk Upload (Preview)
• Item Changed: Campaign
• Output Columns: Account, Campaign, Publisher Target CPA, Publisher Bid Strategy
• Linked Datasource: M1 Report
• Reference Datasource: None
• Owner: Byron Porter (bporter@marinsoftware.com)
• Created by Byron Porter on 2023-07-13 11:21
• Last Updated by Byron Porter on 2023-12-07 21:03

> 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
#
# Promote to new Publisher Bidding Strategy if meet minimum:
#  - Spend
#  - ROAS
#
#
# Author: Byron Porter
# Date: 2023-07-13
#

########### Configurable Params - START ##########

MIN_ROAS = 1
PUB_STRATEGY_TARGET = 'TargetCpa'

########### Configurable Params - END ###########


RPT_COL_CAMPAIGN = 'Campaign'
RPT_COL_ACCOUNT = 'Account'
RPT_COL_PUBLISHER_BIDSTRATEGY = 'Publisher Bid Strategy'
RPT_COL_PUBLISHER_TARGETCPA = 'Publisher Target CPA'
RPT_COL_ROAS = 'ROAS'
RPT_COL_CPC = 'Cost/Conv. $'
BULK_COL_ACCOUNT = 'Account'
BULK_COL_CAMPAIGN = 'Campaign'
BULK_COL_PUBLISHER_BIDSTRATEGY = 'Publisher Bid Strategy'
BULK_COL_PUBLISHER_TARGETCPA = 'Publisher Target CPA'

# temp columns for new values
TMP_PUB_STRATEGY = RPT_COL_PUBLISHER_BIDSTRATEGY + '_'
inputDf[TMP_PUB_STRATEGY] = np.nan

TMP_TARGETCPA = RPT_COL_PUBLISHER_TARGETCPA + '_'
inputDf[TMP_TARGETCPA] = np.nan

# if both Spend and ROAS meet mininum threshold, switch to new pub strategy

inputDf.loc[(inputDf[RPT_COL_ROAS] >= MIN_ROAS), \
             TMP_PUB_STRATEGY \
           ] = PUB_STRATEGY_TARGET

inputDf[TMP_TARGETCPA] = inputDf[RPT_COL_CPC]

# find changed campaigns
changed = (inputDf[TMP_PUB_STRATEGY].notnull()) & (inputDf[RPT_COL_PUBLISHER_BIDSTRATEGY] != inputDf[TMP_PUB_STRATEGY])

if sum(changed) > 0:

    print("== Campaigns with Changed Pub Strategy ==", tableize(inputDf.loc[changed]))

    # only select changed rows
    cols = [RPT_COL_ACCOUNT, RPT_COL_CAMPAIGN, TMP_PUB_STRATEGY, TMP_TARGETCPA]
    outputDf = inputDf.loc[ changed, cols ].copy() \
                    .rename(columns = { \
                        TMP_PUB_STRATEGY: BULK_COL_PUBLISHER_BIDSTRATEGY, \
                        TMP_TARGETCPA: BULK_COL_PUBLISHER_TARGETCPA \
                    })

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

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