Script 269: script good performing keyword 2

Purpose:

The Python script adjusts search bid values and bid override statuses for keywords in advertising campaigns based on specific performance criteria.

To Elaborate

The script is designed to optimize search bid values and bid override statuses for keywords within advertising campaigns. It evaluates the performance of keywords based on conversion rates, cost per conversion, and impression share. Depending on these metrics, the script adjusts the bid values and determines whether the bid override should be turned on or off. The script categorizes campaigns into ‘Brand’, ‘Generic/Competitor’, and ‘Title’ types, applying different rules for bid adjustments based on these categories. The goal is to enhance the efficiency of advertising spend by increasing bids for high-performing keywords and managing bid overrides to ensure optimal visibility and cost-effectiveness.

Walking Through the Code

  1. Initialization:
    • The script begins by defining several constants related to column names used in the input and output data frames.
    • Two new columns are initialized in the input data frame to store updated bid values and bid override statuses.
  2. Condition Setup:
    • Conditions are established to categorize campaigns into ‘Brand’, ‘Generic/Competitor’, and ‘Title’ based on the campaign name.
    • These conditions are used to apply different rules for bid adjustments.
  3. Bid Adjustment Logic:
    • For ‘Brand’ campaigns, if conversions exceed 20 and cost per conversion is below $10, bids are increased by 15% unless the impression share is 100%, in which case the bid remains unchanged.
    • For ‘Generic/Competitor’ and ‘Title’ campaigns, similar logic is applied with different thresholds and bid increase percentages.
    • The bid override is set to ‘On’ for high-performing keywords based on the defined conditions.
  4. Final Adjustments:
    • If the bid override is not ‘On’, it is set to ‘Off’, and the bid is reset to the original search bid value.
    • The script ensures that the updated bid is at least the minimum bid required.
  5. Output Preparation:
    • The updated bid values and override statuses are transferred to the output data frame.
    • The output data frame is filtered to include only rows where bid updates are applicable.

Vitals

  • Script ID : 269
  • Client ID / Customer ID: 1306912115 / 69058
  • Action Type: Bulk Upload (Preview)
  • Item Changed: Keyword
  • Output Columns: Account, Campaign, Group, Keyword, Match Type, Bid Override, Search Bid
  • Linked Datasource: M1 Report
  • Reference Datasource: None
  • Owner: Jonathan Reichl (jreichl@marinsoftware.com)
  • Created by Jonathan Reichl on 2023-08-03 09:51
  • Last Updated by Jeremy Brown on 2024-07-09 13:00
> 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

RPT_COL_KEYWORD = 'Keyword'
RPT_COL_CAMPAIGN_LANGUAGE = 'Campaign Language'
RPT_COL_MATCH_TYPE = 'Match Type'
RPT_COL_TITLE = 'Title'
RPT_COL_GROUP = 'Group'
RPT_COL_CAMPAIGN = 'Campaign'
RPT_COL_CAMPAIGN_STATUS = 'Campaign Status'
RPT_COL_GROUP_STATUS = 'Group Status'
RPT_COL_ACCOUNT = 'Account'
RPT_COL_IMPR = 'Impr.'
RPT_COL_CONV = 'Conv.'
RPT_COL_COST_PER_CONV = 'Cost/Conv. $'
RPT_COL_IMPR_SHARE = 'Impr. share %'
RPT_COL_SEARCH_BID = 'Search Bid'
RPT_COL_MIN_BID = 'Min Bid $'
BULK_COL_ACCOUNT = 'Account'
BULK_COL_CAMPAIGN = 'Campaign'
BULK_COL_GROUP = 'Group'
BULK_COL_KEYWORD = 'Keyword'
BULK_COL_MATCH_TYPE = 'Match Type'
BULK_COL_SEARCH_BID = 'Search Bid'
BULK_COL_BID_OVERRIDE = 'Bid Override'
RPT_COL_BID_OVERRIDE = 'Bid Override'

outputDf[BULK_COL_SEARCH_BID] = "<<YOUR VALUE>>"
outputDf[BULK_COL_BID_OVERRIDE] = "<<YOUR VALUE>>"

today = datetime.datetime.now(CLIENT_TIMEZONE).date()
print(tableize(inputDf))



UPDATE_BID = RPT_COL_SEARCH_BID + '_new'
UPDATE_BID_OV = BULK_COL_BID_OVERRIDE + '_new'
# blank out tmp field
inputDf[UPDATE_BID] = np.nan
inputDf[UPDATE_BID_OV] = np.nan

# Conditions for bid adjustments
brand_condition = inputDf[RPT_COL_CAMPAIGN].str.contains('Brand', case=False)
generic_competitor_condition = inputDf[RPT_COL_CAMPAIGN].str.contains('Generic|Competitor', case=False) & ~brand_condition
title_condition = inputDf[RPT_COL_CAMPAIGN].str.contains('Title', case=False) & ~brand_condition & ~generic_competitor_condition


# Apply bid adjustments
inputDf.loc[brand_condition & (inputDf[RPT_COL_CONV] > 20) & (inputDf[RPT_COL_COST_PER_CONV] < 10) & (inputDf[RPT_COL_IMPR_SHARE] <= 100), UPDATE_BID_OV] = 'On'
inputDf.loc[brand_condition & (inputDf[RPT_COL_CONV] > 20) & (inputDf[RPT_COL_COST_PER_CONV] < 10) & (inputDf[RPT_COL_IMPR_SHARE] < 100), UPDATE_BID] = inputDf[RPT_COL_SEARCH_BID]*1.15
inputDf.loc[brand_condition & (inputDf[RPT_COL_CONV] > 20) & (inputDf[RPT_COL_COST_PER_CONV] < 10) & (inputDf[RPT_COL_IMPR_SHARE] == 100), UPDATE_BID] = inputDf[RPT_COL_SEARCH_BID]


inputDf.loc[generic_competitor_condition & (inputDf[RPT_COL_CONV] > 10) & (inputDf[RPT_COL_COST_PER_CONV] < 15) & (inputDf[RPT_COL_IMPR_SHARE] <= 100), UPDATE_BID_OV] = 'On'
inputDf.loc[generic_competitor_condition & (inputDf[RPT_COL_CONV] > 10) & (inputDf[RPT_COL_COST_PER_CONV] < 15) & (inputDf[RPT_COL_IMPR_SHARE] < 100), UPDATE_BID] =   inputDf[RPT_COL_SEARCH_BID]*1.10
inputDf.loc[generic_competitor_condition & (inputDf[RPT_COL_CONV] > 10) & (inputDf[RPT_COL_COST_PER_CONV] < 15) & (inputDf[RPT_COL_IMPR_SHARE] == 100), UPDATE_BID] =   inputDf[RPT_COL_SEARCH_BID]


inputDf.loc[title_condition & (inputDf[RPT_COL_CONV] > 10) & (inputDf[RPT_COL_COST_PER_CONV] < 15) & (inputDf[RPT_COL_IMPR_SHARE] <= 100), UPDATE_BID_OV] = 'On'
inputDf.loc[title_condition & (inputDf[RPT_COL_CONV] > 10) & (inputDf[RPT_COL_COST_PER_CONV] < 15) & (inputDf[RPT_COL_IMPR_SHARE] < 100), UPDATE_BID] =  inputDf[RPT_COL_SEARCH_BID]*1.15
inputDf.loc[title_condition & (inputDf[RPT_COL_CONV] > 10) & (inputDf[RPT_COL_COST_PER_CONV] < 15) & (inputDf[RPT_COL_IMPR_SHARE] == 100), UPDATE_BID] =  inputDf[RPT_COL_SEARCH_BID]

inputDf.loc[(inputDf[UPDATE_BID_OV] != 'On'), UPDATE_BID_OV] = 'Off'
inputDf.loc[(inputDf[UPDATE_BID_OV] != 'On'), UPDATE_BID] = inputDf[RPT_COL_SEARCH_BID]

inputDf[UPDATE_BID] = inputDf.apply(
    lambda row: max(row[UPDATE_BID], row[RPT_COL_MIN_BID]) if row[UPDATE_BID_OV] == 'On' else row[UPDATE_BID],
    axis=1
)

#add new values to outputDf

outputDf.loc[:,BULK_COL_SEARCH_BID] = inputDf.loc[:, UPDATE_BID]
outputDf.loc[:,BULK_COL_BID_OVERRIDE] = inputDf.loc[:, UPDATE_BID_OV]


outputDf = outputDf[
    (inputDf[UPDATE_BID].notnull()) & 
    ((inputDf[UPDATE_BID_OV] == 'On') | (inputDf[UPDATE_BID_OV] != inputDf[RPT_COL_BID_OVERRIDE]))
]

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

comments powered by Disqus