AI insights and analysis for analytics

Overview

AI insights analyze your analytics data automatically. They help you understand API performance, usage patterns, and possible issues. The feature examines reports and individual API events. It then shows recommendations and key insights.

AI insights are available in the API Manager UI at the provider organization, catalog, and spaces. You can also access them through the REST API and CLI.

Prerequisites

Before you use AI insights and analysis, make sure that you have the following prerequisites:

• The API Agent must be enabled and configured in your provider organization. For more information about configuring the API Agent, see Configuring API Assistant.
• Appropriate permissions to view analytics data in your provider organization, catalog, or space.

Reports that support AI insights and analysis

AI can analyze several analytics reports. It highlights trends, patterns, and recommended actions. For more information about reports, see Analytics reports view. The following reports support AI analysis:

• Call volume trends: Provides a high-level view of API activity for a selected time range. The report shows total API calls, successful and failed calls, and the volume of bytes sent and received.

  

• Consumer trends:Provides an overview of long‑term trends in consumer activity, including application usage, subscription counts, and API call volumes.

  

• AI Platform report: Review AI and LLM API usage, including provider and consumer statistics.

  

• Disk usage report: tracks and monitors storage usage across APIs, products, applications, and consumer organizations.

  

AI insights and analysis for individual events

When you open an API event, AI reviews the full execution path. It highlights findings about function, performance, and policy behavior. For each event, AI analyzes the request, response, gateway behavior, and downstream calls. This analysis helps you diagnose problems and tune performance. For more information about events, see Discover view.

Using AI insights in the API Manager UI

To access AI insights and analysis in the API Manager UI:

1. Click in the navigation panel.

2. You can also access analytics data at the catalog and space levels. The roles and permissions assigned to you determine what data you can view. For more information about catalogs and spaces, see Working with catalogs and Using syndication in API Connect.

   To access the Catalog scoped analytics dashboard, complete the following steps:

   1. In the API Manager UI, click Manage in the navigation panel.
   2. From the dashboard select the catalog.
   3. Optional: If spaces are enabled Click the Spaces tab and select a space.

   4. Click the Analytics tab.

3. To access AI insights and analysis for a report. Open any supported analytics report.
4. Look for the AI insights and analysis section, which displays an AI-Powered badge.
5. Review the AI-generated analysis, which includes:
   • Summary: A high-level overview of the key findings and trends.
   • Key insights: Detailed observations organized by category (Performance, Usage, Data, and so on).
   • Recommendations: Actionable suggestions prioritized by importance (High priority, Medium priority, and so on).
   • Anomalies detected: Highlights unusual patterns or possible issues.
   • Predictions and forecast: Expected growth trends and future projections (where applicable).
6. To access AI insights and analysis for an individual API event. Click Discover.
7. Look for the AI insights and analysis section, which displays an AI-Powered badge.
8. Review the AI-generated analysis, which includes:
   • Summary: Provides a brief overview of the API request outcome, highlighting the error status and key context details.
   • Response analysis: Explains the detected error, its source (for example, gateway or backend), and why the response status differs from the backend status.
   • Latency analysis: Breaks down the time that is spent across request handling stages, showing where most processing delay occurs.
   • Rate limiting status: Indicates current API quota usage, remaining request limits, and whether throttling is affecting requests.

The AI analysis includes a confidence indicator (for example, Confidence: 85%) that shows how reliable the analysis is based on the available data.

Accessing AI insights by using the API and CLI

In addition to the API Manager UI, you can access AI insights programmatically through:

• REST API: Use the Analytics REST API to retrieve AI-generated insights for reports and events. For more information, see Using the Analytics REST API.
• CLI: Use the API Connect CLI to access analytics data and AI insights. For more information, see Using the Analytics CLI.

Best practices

To get the most value from AI insights and analysis:

• Review AI insights regularly to stay informed about trends and potential issues before they become critical.
• Prioritize high-severity recommendations because they can affect API performance or availability.
• Use anomaly detection to identify unusual patterns that might indicate security issues, configuration problems, or unexpected usage.
• Combine AI insights with your knowledge of the API environment to make better decisions.
• Use event-level analysis to find and fix API errors quickly.
• Monitor confidence levels in AI analysis. Higher confidence levels indicate more reliable insights that are based on sufficient data.