How to Integrate IBM Watson with Google Analytics

Setup Prerequisites

• Create an IBM Cloud account if you don't have one.

 

• Ensure you have access to a Google Analytics account with the property you want to integrate with IBM Watson.

 

• Set up an API key for Google Analytics and IBM Watson. Both services require authentication using their API keys.

Access IBM Watson and Google Analytics APIs

• Navigate to IBM Cloud Dashboard and create an instance of IBM Watson that you intend to use, e.g., Watson Assistant, Natural Language Understanding, etc.

 

• Go to your Google Cloud Platform, select the appropriate project, and enable the Google Analytics API under “APIs & Services.”

Install Necessary Libraries

• To interact with these services programmatically, you can use Python. First, ensure Python is installed on your system.

 

• Install the `ibm-watson` and `google-api-python-client` libraries:

pip install ibm-watson google-api-python-client

Authenticate with IBM Watson

• Save the Watson API key and your instance URL.

 

• Use the following code snippet to authenticate with IBM Watson:

from ibm_watson import AssistantV2
from ibm_cloud_sdk_core.authenticators import IAMAuthenticator

authenticator = IAMAuthenticator('your-ibm-watson-api-key')
assistant = AssistantV2(
    version='2021-06-14',
    authenticator=authenticator
)
assistant.set_service_url('your-ibm-watson-url')

Authenticate with Google Analytics

• Save the credentials to a JSON file recommended by Google when you create an API key. Use this file for authentication.

 

• Use the snippet below to authenticate:

from google.oauth2 import service_account
from googleapiclient.discovery import build

credentials = service_account.Credentials.from_service_account_file(
    'path/to/your/credentials.json',
    scopes=['https://www.googleapis.com/auth/analytics.readonly']
)

analytics = build('analyticsreporting', 'v4', credentials=credentials)

Fetch Data from Google Analytics

• Query Google Analytics to fetch the data you need:

 

• Replace the placeholder values with actual values that suit your needs:

response = analytics.reports().batchGet(
    body={
        'reportRequests': [
        {
          'viewId': 'your-view-id',
          'dateRanges': [{'startDate': '7daysAgo', 'endDate': 'today'}],
          'metrics': [{'expression': 'ga:sessions'}]
        }]
    }
).execute()

Process Data using IBM Watson

• Send this data to IBM Watson for processing. Here's an example that forwards data to Watson Assistant:

session_response = assistant.create_session(
    assistant_id='your-assistant-id'
).get_result()

session_id = session_response['session_id']

message_response = assistant.message(
    assistant_id='your-assistant-id',
    session_id=session_id,
    input={
        'message_type': 'text',
        'text': 'User had {} sessions this week'.format(response['reports'][0]['data']['totals'][0]['values'][0])
    }
).get_result()

print(message_response['output']['generic'][0]['text'])

Considerations for Integration

• Data Volume: Be mindful of API limits and perform data processing in batches if necessary.

 

• Security: Ensure you store your API keys securely and rotate them periodically.

 

• Error Handling: Include error handling to manage potential API failures gracefully.

Test Your Integration

• After setting up both services, run the complete code to ensure data flows as expected.

 

• Check both your Google Analytics reports and IBM Watson response to confirm successful integration.