26 research tools that give Claude the ability to analyze your customer data via natural language. No SQL. No dashboards. Answers in seconds.
customer-intel-mcp sits between Claude and your BigQuery tables. It generates SQL, routes it through your BigQuery MCP, and renders structured findings.
All table paths, column names, and segment definitions come from config.yaml — a single file you fill in once. No hardcoded schema assumptions.
Every tool follows the same delivery standard: structured output, signal emojis, key finding callout, and next steps. Never raw JSON.
Complete behavioral + profile snapshot for any customer from your configured tables.
Plain-English narrative portrait tailored to designer, copywriter, PM, or exec audience.
Lightweight behavioral profile optimized for anomaly scoring.
Aggregate behavioral patterns across a cohort. Answers "why" questions at scale.
Diff two cohorts side-by-side, surfaces signals that predict success or failure.
Translate a plain-English description into real customers with matching profiles.
Sample N customers from any named segment or SQL filter.
Measure how a cohort's signals changed between two dates.
Base-wide scan of unrealized value across all segments.
Signal fingerprint of your best customers vs. the average customer.
Score a session event against a customer's behavioral baseline for anomaly detection.
Customer base-level behavioral dashboard for a given date.
TAM/SAM/SOM sizing for a proposed feature from your actual customer base.
Predict how a cohort reacts to a feature across 5 dimensions before you build it.
Score customer-facing copy against a cohort across relevance, tone, urgency, trust, CTA.
Rigorous TAM/SAM/SOM from your real customer data with segment breakdown.
Test a hypothesis against your customer data. Returns SUPPORTED/REFUTED/INCONCLUSIVE.
Pre/post signal movement for a campaign or product launch.
Best action for a specific customer or cohort right now — individual or batch mode.
Save a research finding to research_log.json. Persists across Claude sessions.
Retrieve and search prior findings. Your research memory across sessions.
Clear all or partial research log with preview mode and tag-based filtering.
Turn saved findings into a PRD background, Slack post, exec summary, or research review.
Prioritized list of research questions based on your log and open follow-ups.
Weekly synthesis: customer base health, research summary, open questions, and focus for next week.
One config file. One install script. Works with any BigQuery MCP.
git clone https://github.com/gnavada/customer-intel-mcp.git && cd customer-intel-mcp
cp config.example.yaml config.yaml — then edit with your BigQuery project, dataset, table names, and column mappings. Takes ~10 minutes.
bash setup.sh — creates a Python venv, installs dependencies, and registers the MCP in Claude Desktop automatically.
Quit and reopen Claude Desktop. Ask: "Are you connected to customer-intel-mcp?" — you're ready.
Copy any of these into Claude Desktop once customer-intel-mcp is connected.
"Get an opportunity map of my entire customer base"
"What do my churning customers have in common?"
"Compare my champions segment to my at-risk segment"
"Find customers who are highly engaged but haven't upgraded yet"
"How would my at-risk customers react to a win-back offer?"
"Score this email subject line against my churning segment: 'We miss you — here's what's new'"
"How big is the upsell opportunity in my engaged-but-not-premium segment?"
"Test this hypothesis: customers who engage in the first 7 days retain better at 30 days"
"Generate an exec summary of everything I've learned about churn this quarter"