Bing Ads API report request (Python)

Question

Hello everyone,

I am facing some problems in the data extraction process using the Bing API with Python. I followed the Overview, but I still don't understand exactly how to request and extract the report for the following metrics:

Metrics:

Investment (Cost)
Impressions
Clicks
Conversions

Can anyone suggest me some solution or path to follow ?

Answer 1

I was faced with the same question and here is what finally fixed it for me:

(Note that I assume that you have a working authentication and that you know what I mean when I refer to certain authentication constructs. If your problem lies there consult the official documentation)

The API you want to use is the reporting api so you have to

import bingads
from bingads.v13 import reporting
rep_man = reporting.ReportingServiceManager(authorization_data=authorization_data, working_directory="tmp")

I don't know if the working_directory parameter is strictly necessary, I set it just to be safe

Now you need to construct your report request you can do this like this:

rep_serv = bingads.ServiceClient(service = 'ReportingService',
version=13,
authorization_data=authorization_data)

rep_req = rep_serv.factory.create("CampaignPerformanceReportRequest")
rep_req.Aggregation = 'Daily'
time = rep_serv.factory.create('ReportTime')
time.PredefinedTime='ThisMonth'
time.CustomDateRangeStart = None
time.CustomDateRangeEnd = None
rep_req.Time = time
rep_req.Filter=None
scope = rep_serv.factory.create('AccountThroughCampaignReportScope')
scope.AccountIds = {'long': [account_id]}
scope.Campaigns = None
rep_req.Scope=scope
rep_col = rep_serv.factory.create('ArrayOfCampaignPerformanceReportColumn')
rep_col.CampaignPerformanceReportColumn.append(['TimePeriod', 'CampaignName', 'Clicks', 'Conversions', 'Impressions', 'Spend'])
rep_req.Columns = rep_col
rep_req.Format = "Csv"
rep_param = reporting.ReportingDownloadParameters(report_request = rep_req, result_file_directory = None, result_file_name = 'report.csv')
rep_man.download_report(rep_param)

with this very ugly function I finally got a result. I hope this is not too late to be useful