Google Analyst

Google Analyst gives you quick, clear summaries of your website performance based on Google Analytics 4 (GA4).

It focuses on the most important metrics — like sessions, new users, bounce rate — so you can see at a glance how your site is performing without digging through GA’s dashboard.

👉 How the Template Works

Google Analyst is an agent that connects to your Google Analytics property with your own service-account credentials. It understands natural-language requests such as “last week’s traffic”, converts relative dates into exact ranges, calls the official Analytics Data API, and returns a neatly formatted report that includes:

• Overview

• Key metrics (sessions, users, bounce rate, engagement time, and more)

• Trends and insights

• Optional recommendations

All you have to do is chat with the agent; no SQL, dashboards, or spreadsheets required.

🎯 Steps to Use

1. Confirm you have a Google Analytics 4 property

• Sign in at https://analytics.google.com.

• Make sure a Google Analytics property is listed for your website and note its numeric Property ID (e.g. 1234567890).

2. Create a service account on Google Cloud

1. Go to https://console.cloud.google.com and select (or create) a project.

2. Open IAM & Admin ▸ Service Accounts.

3. Click Create service account → name it ga-bika-service → assign role Viewer → Done.

4. In that account, open Keys ▸ Add key ▸ JSON → download service-account.json.

5. Open the JSON file and copy the full JSON contents:

   Open the downloaded service-account.json, select all and copy the entire JSON text (including client_email and private_key).

   In Bika, paste the complete JSON file into the credentials input — do not extract only individual fields.

6. Enable the Analytics Data API

• Google Cloud Console ▸ APIs & Services ▸ Library.

• Search Analytics Data API ➜

  • If the button says Enable, click it.

  • If the button says Manage, the API is already enabled.

3. Grant the service account Viewer access inside Google Analytics

1. Back in https://analytics.google.com click the gear Admin.

2. Under the centre Property column choose Property Access Management.

3. Click ＋ Add users → paste the client_email from step 2-3 → Viewer → Add.

4. If you did not record the numeric Property ID earlier, copy it now (e.g. 1234567890).

4. Connect the agent in Bika.ai

1. Open the Google Analytics skillset configuration page in Bika (the Google Analytics skillset).

2. Paste the entire JSON credential content you downloaded (service-account.json) into the Google Credentials field.

3. Click Connect to complete credential configuration.

4. Provide your Google Analytics Property ID when prompted (or include it in the System Prompt during conversations).

5. Start chatting

Ask things like:

“How was the new users in July 2025?”
“What was the engagement rate and pageviews last week?”
“Show me yesterday’s sessions and bounce rate.”

👉 Supported Metrics

• Sessions
• New Users
• Active Users
• Bounce Rate
• Avg Session Duration
• Engagement Rate
• Pageviews
• Views per Session
• Conversions
• Event Count

👉 Suitable Users

• Website owners who want quick traffic snapshots every day. • Marketing or product teams that need plain-English GA4 insights. • Anyone comfortable creating a Google Cloud project but not keen on coding dashboards.

FAQ

1. Is it safe to share the Service Account JSON key?

Yes, within the context of this agent. You are not sharing your personal Google password. You are creating a specific "robot" account (Service Account) with limited "Viewer" access to your Analytics property. You can revoke this access at any time in the Google Cloud Console.

2. Where do I find the Property ID?

The Property ID is a numeric string (e.g., 123456789). You can find it by going to Google Analytics Admin > Property Settings (in the middle column) > Property Details. It is displayed in the top right corner of that pane.

3. Can I ask for custom date ranges?

Yes. The agent understands natural language date references. You can ask for "last week," "yesterday," "Q3 2025," or specific dates like "from Jan 1 to Jan 31, 2025."