KPI Depot MCP - KPI Intelligence for Hermes Agent¶
KPI Depot MCP gives your AI agent access to the world's largest structured KPI database - 20,000+ KPI definitions with formulas, 30,000+ industry benchmarks with source attribution. Plan strategies, set targets, and evaluate performance against real industry data.
What It Does¶
KPI Depot MCP brings benchmark intelligence to agent workflows:
- KPI definitions - 20,000+ KPIs with exact formulas, data sources, and calculation methods
- Industry benchmarks - 30,000+ source-attributed benchmarks across sectors, company sizes, and regions
- KPI discovery - Search by business function, industry, or metric category
- Target setting - Set data-backed targets based on peer performance percentiles
- Performance gap analysis - Compare your metrics to industry quartiles
Quick Setup¶
Prerequisites¶
- KPI Depot account: Sign up at kpidepot.com
- API key: Generate from your KPI Depot dashboard
- The MCP server is remote-hosted - no local installation required
Add to Hermes Agent¶
{
"mcpServers": {
"kpidepot": {
"command": "npx",
"args": ["-y", "@kpidepot/mcp-server"],
"env": {
"KPIDEPOT_API_KEY": "your_api_key_here"
}
}
}
}
Key Capabilities¶
| Tool | Description |
|---|---|
search_kpis |
Search 20,000+ KPI definitions by keyword, function, or industry |
get_kpi_detail |
Get full KPI detail: definition, formula, data sources, calculation guide |
get_benchmarks |
Pull industry benchmarks for a specific KPI (by sector, size, region) |
compare_to_benchmark |
Compare your metric to industry quartiles (25th/50th/75th percentile) |
list_categories |
Browse KPI categories: Marketing, Sales, Finance, Operations, HR, Product |
get_peer_group |
Find peer companies by industry, revenue band, employee count |
benchmark_report |
Generate a full benchmark report for a set of KPIs |
Use Cases for Business Operators¶
1. Board Deck Data¶
Generate data-backed performance context:
Agent prompt: "I'm preparing our Q3 board deck. For our 5 core metrics
(CAC, LTV, churn, NPS, gross margin), pull SaaS industry benchmarks
for companies our size ($5-20M ARR). Show our actuals vs 50th and 75th
percentile. Format as a comparison table."
2. Investor Pitch Metrics¶
Validate your metrics against the market:
Agent prompt: "We pitch investors next week. Pull benchmarks for
B2B SaaS Series A companies: CAC payback period, net dollar retention,
gross margin, and magic number. Where do we need to improve to be
in the top quartile? What's table stakes for Series A?"
3. KPI Program Design¶
Design a measurement framework for a new initiative:
Agent prompt: "We're launching a customer success function. What are
the standard KPIs for B2B SaaS customer success? Pull the top 10 with
formulas and benchmarks. Recommend which 5 we should track first
and what good looks like at our stage."
4. Quarterly Business Review Prep¶
Automate QBR data preparation:
Agent prompt: "Pull benchmarks for all 12 KPIs in our Q3 scorecard.
For each, show: our Q3 actual, industry median, top quartile, and
our percentile rank. Flag any metric where we're below the 25th
percentile - those need a remediation plan."
Integration with CorpusIQ¶
KPI Depot MCP + CorpusIQ = data-backed strategy execution:
- CorpusIQ GA4 connector → pull your actual marketing metrics
- CorpusIQ Stripe connector → pull your actual financial metrics
- KPI Depot MCP → compare your actuals to industry benchmarks
- AI agent → produce "where we stand" reports automatically
Your agent becomes a strategy analyst that doesn't just report your numbers - it tells you how your numbers compare to the market and what to do about it.
Pricing¶
- KPI Depot MCP: Remote MCP server (no local install)
- KPI Depot: Check kpidepot.com for current API pricing
- Free tier: Typically offers limited benchmark access for evaluation
Limitations¶
- Remote MCP server requires internet connectivity at all times
- Benchmark data currency depends on KPI Depot's update frequency
- Industry coverage strongest in: SaaS, ecommerce, financial services, healthcare
- Regional benchmarks primarily US and Western Europe; emerging markets have thinner data