searchlight-api-client-python
The Searchlight API Client is a Python package that makes it easy to authenticate with and retrieve data from the Conductor Searchlight API.
Getting Started
Installation
The latest released version can be installed through the Python package index.
pip install searchlight-api
Dependencies
- Requests: 2.19.1 or higher
- Simplejson: 3.11.1 or higher
Authentication
The Searchlight API Client needs to know your API Key and API Secret to authenticate with Searchlight. It can find these in one of two ways:
- Add the credentials to your environmental variables (preferred)
- MacOS / Linux: Edit your .bash_profile to include the lines
- export SEARCHLIGHT_API_KEY=xxxxx
- export SEARCHLIGHT_SHARED_SECRET=xxxxxx
- Windows: Add SEARCHLIGHT_API_KEY and SEARCHLIGHT_SHARED_SECRET in the Environment Variables in the Advanced system settings
- MacOS / Linux: Edit your .bash_profile to include the lines
- Pass the credentials when instantiating the API Client from searchlight_api.client import AccountService account_service = AccountService(api_key=xxxxx, secret=xxxxx)
If you don't have an API Key and Secret, request one from http://developers.conductor.com/. You should receive an email within a week containing your credentials.
Using searchlight_api
The Searchlight API Package has two main functionalities: client and analysis
- client provides an SDK around the Searchlight REST API, making it easy to request the data that you need from your accounts.
- analysis includes functions for aggregating data (Rank and Search Volume) using the Searchlight API.
Examples
Client
SearchlightService
The SearchlightService client provides wrappers for getting Searchlight and profile configuration data
from searchlight_api import client
# instantiate Searchlight client
ss = client.SearchlightService()
# Retrieve all Searchlight accounts you have access to.
# You can use this to retrieve Account IDs to instantiate the AccountService client
my_accounts = ss.get_accounts()
print(my_accounts)
[{'accountId': 'XXXX',
'isActive': False,
'name': 'Account 1',
'webProperties': 'https://api.conductor.com/v3/accounts/XXXX/web-properties'},
{'accountId': 'YYYY',
'isActive': True,
'name': 'Account 2',
'webProperties': 'https://api.conductor.com/v3/accounts/YYYY/web-properties'}]
# Account IDs can also be found in Searchlight URLs: https://searchlight.conductor.com/YYYY/insight-stream
In Searchlight, keywords can be tracked across different Rank Sources (or Search Engines), Locations and Devices. You can retrieve what's supported with the following:
rank_sources = ss.get_rank_sources()
locations = ss.get_locations()
devices = ss.get_devices()
AccountService
The AccountService client provides functionality for retrieving reporting data from Searchlight accounts, while retaining all the functionality found in the SearchlightService client. A Searchlight account id is required upon instantiation.
from searchlight_api import client
# instantiate the AccountService client
account_service = client.AccountService(YYYY)
# retrieve all web properties tracked in the account
web_properties = account_service.get_web_properties()
web_property_id = web_properties[0]['webPropertyId']
# the web_property_id can be used to get the name of the corresponding domain
domain_name = account_service.get_domain_name(web_property_id)
# use the web_property_id to get all tracked searches in the account
tracked_searches = account_service.get_tracked_searches(web_property_id)
# get a rank_source_id that the web_property is tracked against
rank_source_id = web_properties[0]['rankSourceInfo'][0]['rankSourceId']
# retrieve rank data for searches tracked against a web property and rank source for a given date within a reporting interval.
# by default the current date is used.
ranks = account_service.get_ranks(web_property_id, rank_source_id, date='2019-01-12')
# retrieve search volume data for searches tracked against a web property and rank source for a given date within a reporting interval.
# by default the current date is used.
search_volume = account_service.get_volume(web_property_id, rank_source_id, date='2019-01-12')
Analysis
Analysis allows you to use the AccountService client to aggregate reporting data across an entire account for a given date.
from searchlight_api.client import AccountService
from searchlight_api import analysis
account_service = AccountService(YYYY)
rank_data = analysis.rank_data(account_service, date='2019-01-12')
search_volume = analysis.search_volume(account_service)
To request additional aggregation methods, reach out to dgoodman@conductor.com or pmurphy@conductor.com
Using Pandas with Searchlight API Client
from searchlight_api import client, analysis
import pandas as pd
# instantiate AccountService object
account_service = client.AccountService(YYYY)
# get rank sources, locations and devices and turn to dfs
rank_source_df = pd.DataFrame(account_service.get_rank_sources())
location_df = pd.DataFrame(account_service.get_locations())
device_df = pd.DataFrame(account_service.get_devices())
# get the tracked searches and turn into a df
tracked_search_df = pd.DataFrame(analysis.all_tracked_searches(account_service))
# turn search_volume and rank_data to Data Frames
search_volume_df = pd.DataFrame(analysis.search_volume(account_service))
rank_data_df = pd.DataFrame(analysis.rank_data(account_service))
report = pd.merge(rank_data_df, search_volume_df, on=['trackedSearchId', 'rankSourceId', 'webPropertyId'])
report = report.merge(rank_source_df, on='rankSourceId', how='left')
print(report.head())
itemType rankSourceId standardRank target targetDomainName \
0 ANSWER_BOX 1 3.0 example.com
1 IMAGE_RESULT 1 NaN example.com
2 LOCAL_RESULT 1 NaN www.organiccompetitor.com
3 LOCAL_RESULT 1 NaN www.organiccompetitor.com
4 LOCAL_RESULT 1 NaN www.organiccompetitor.com
targetUrl targetWebPropetyId \
0 https://www.example.com/subfolder1/ 43162.0
1 https://www.example.com/subfolder2/... 43162.0
2 http://www.organiccompetitor.com/subfolder... NaN
3 http://www.organiccompetitor.com/subfolder/... NaN
4 http://www.organiccompetitor.com/subfolder/... NaN
trackedSearchId trueRank webPropertyId averageVolume \
0 7188291 6 43162 135000
1 7188291 62 43162 135000
2 7188291 4 43162 135000
3 7188291 2 43162 135000
4 7188291 3 43162 135000
volumeItems baseDomain \
0 [{'volume': 165000, 'month': 11, 'year': 2018}... google.com
1 [{'volume': 165000, 'month': 11, 'year': 2018}... google.com
2 [{'volume': 165000, 'month': 11, 'year': 2018}... google.com
3 [{'volume': 165000, 'month': 11, 'year': 2018}... google.com
4 [{'volume': 165000, 'month': 11, 'year': 2018}... google.com
description name deviceId isActive locationId \
0 Google (US / English) GOOGLE_EN_US 1 True 1
1 Google (US / English) GOOGLE_EN_US 1 True 1
2 Google (US / English) GOOGLE_EN_US 1 True 1
3 Google (US / English) GOOGLE_EN_US 1 True 1
4 Google (US / English) GOOGLE_EN_US 1 True 1
preferredUrl queryPhrase
0 http://www.example.com/ example phrase
1 http://www.example.com/ example phrase
2 http://www.example.com/ example phrase
3 http://www.example.com/ example phrase
4 http://www.example.com/ example phrase
