Core Tools
query_analytics (Google) & bing_analytics_query (Bing)
These are your primary tools for getting raw keyword data. They support filtering and dimension grouping.
- When to use: When you need a specific list of keywords, pages, or country-level data.
- Best for: “List the top 10 keywords for the last 30 days.”
compare_periods (Google) & bing_analytics_compare_periods (Bing)
Calculates the delta between two time windows.
- When to use: When you need to see if you are growing or shrinking.
- Best for: “Compare this week’s performance to last week.”
analytics_page_performance (GA4)
Detailed page metrics from Google Analytics 4, including sessions, engagement rates, and unique views.
- When to use: When you need to understand how users behave after they click from search results.
- Best for: “Show me the engagement rate for my top landing pages.”
Advanced Analysis
analytics_time_series (Google) & bing_analytics_time_series (Bing)
The “brain” of our analytics suite. It calculates rolling averages (to smooth out weekend dips), detects seasonality, and provides simple forecasting.
- When to use: To understand long-term trends or identify the exact day a drop started.
- What problem it solves: Distinguishing between a “bad day” and a “downward trend.”
analytics_drop_attribution (Google) & bing_analytics_drop_attribution (Bing)
Correlates traffic drops with known algorithm updates or device-specific issues.
- When to use: When you see a sudden decline.
- What problem it solves: Instantly tells you if a drop coincides with a Core Update.
Example Agent Prompts
1. Checking specific performance
“Run a performance report for https://example.com comparing January to December. Segment the data by device (mobile vs. desktop).“
2. Visualizing trends
“Analyze the click trend for my top 5 keywords over the last 90 days. Use a 7-day rolling average to clean up the data.”
3. Regional analysis
“Which country had the highest growth in CTR for keywords containing ‘premium’ in the last 6 months?”