Introducing PubPerf MCP: Query Your Analytics with AI

Publisher analytics teams routinely navigate multiple dashboards, apply filters across dimensions, and cross-reference metrics to answer operational questions. PubPerf MCP streamlines this workflow by connecting AI assistants directly to your analytics data.

PubPerf now supports the Model Context Protocol (MCP) — an open standard that enables AI assistants like Claude, ChatGPT, and Cursor to query structured data sources. With PubPerf MCP, teams can retrieve performance insights through natural language instead of manual dashboard navigation.

Natural Language Analytics

Once connected, AI assistants can query any PubPerf metric on your behalf:

• "Show Core Web Vitals across all properties for the past week."
• "Which pages have the highest LCP on example.com?"
• "Compare prebid revenue by bidder across all sites this month."
• "Identify bidders with the highest timeout rates."
• "Rank content authors by revenue contribution."

The AI assistant queries PubPerf directly, retrieves the relevant data, and returns a structured response — reducing the time from question to insight.

60+ Analytics Tools in One Interface

PubPerf MCP exposes over 60 analytics tools covering the full scope of publisher operations:

Portfolio Overview — Aggregated summaries of users, pageviews, sessions, performance scores, impressions, and viewability across all properties. Monitor portfolio health from a single query.

Core Web Vitals — LCP, CLS, INP, and FCP with p50, p75, and p90 percentiles. Identify worst-performing pages, view score distributions, and track trends over time — segmented by device, browser, or geography.

Prebid and Revenue — Bidder-level analysis of CPM, win rates, timeouts, and revenue. Revenue forecasting with confidence bands. Surface underperforming demand partners and inform optimization decisions.

Ad Performance — Impressions, auction latency, render times, and viewability metrics. Detect performance regressions early and quantify their revenue impact.

Content Analytics — Top-performing pages, authors, and categories ranked by traffic and revenue. Connect editorial strategy to measurable business outcomes.

Designed for Publisher Operations

Immediate answers. Bypass report builders and dashboard navigation. Describe the insight you need and receive the data directly.

Accessible to every team. Ad ops, editorial, product, and leadership can all query performance data without specialized dashboard training.

Flexible time ranges and filters. Query from the last 3 hours to a full year of history. Filter by device type, drill into individual properties, or compare across the entire portfolio.

Enterprise-grade security. Read-only access ensures no data can be modified. API key authentication, organization-scoped permissions, and rate limiting protect your data at every layer. Keys can be revoked or regenerated at any time.

Get Started

PubPerf MCP is available for all accounts. Setup takes minutes — create an API key, configure your AI client, and begin querying.

Visit the MCP page to learn more, or refer to the integration guide for detailed instructions.