Growth benchmark

Grades your growth metrics against companies like yours and writes a report with three fixes, ranked by impact.

  • Free
  • No login
  • Read-only, stores nothing
  • B2B SaaS and consumer apps
Open in browser →

Connect your AI assistant

MCP server https://benchmark.scilla.studio/mcp

Run this in your terminal, then start Claude Code and say "Benchmark my product".

claude mcp add --scope user --transport http scilla-benchmark https://benchmark.scilla.studio/mcp

Run this in your terminal, then start Codex and say "Benchmark my product".

codex mcp add scilla-benchmark --url https://benchmark.scilla.studio/mcp

In claude.ai or the Claude desktop app:

  1. Open Customize, then Connectors.
  2. Click +, choose Add custom connector, and paste https://benchmark.scilla.studio/mcp.
  3. Click Add. You can leave the advanced settings empty.
  4. Start a new chat and say "Benchmark my product".

The free plan allows one custom connector.

In ChatGPT:

  1. Open Settings, then Apps & Connectors, then Advanced settings, and turn on Developer mode.
  2. Back in Apps & Connectors, click Create. Name it Growth benchmark, paste https://benchmark.scilla.studio/mcp, and choose No authentication.
  3. Start a new chat, add the connector from the + menu, and say "Benchmark my product".

Developer mode needs a paid ChatGPT plan.

Paste this into any coding agent. It adds the server and starts asking for your numbers.

Add the scilla.studio growth benchmark MCP server: https://benchmark.scilla.studio/mcp (Streamable HTTP, no login). In Claude Code, run: claude mcp add --scope user --transport http scilla-benchmark https://benchmark.scilla.studio/mcp. In Codex, run: codex mcp add scilla-benchmark --url https://benchmark.scilla.studio/mcp. In any other tool, add it to your MCP settings. Then benchmark my product right away: call the server's get_metric_checklist tool, over HTTP if its tools are not loaded in this session yet, and ask me for the numbers.

Tools on the server

get_metric_checklist Lists the numbers the benchmark needs and where to find them, so your assistant knows what to ask you.
get_benchmarks Returns the healthy range for each metric and the sources behind it.
grade_metrics Grades your numbers against those ranges: below, inside or above, per metric and for acquisition, retention and growth.
project_growth Projects your monthly active users 12 months ahead with the benchmark's cohort model.

Every answer ends with a link to your report.

What a session looks like

Benchmark my product

get_metric_checklist

Is it B2B SaaS or a consumer app? And what does it cost you to win a customer, and what is one customer worth over their lifetime?

B2B SaaS. CAC is about €850 and LTV about €2,600.

Of the users who signed up three months ago, what share was still active at Day 30 and at Day 90?

6% and 1.8%.

grade_metrics

Acquisition is inside the range. Retention is critical: Day 90 sits at 1.8% against a B2B range of 2.5% to 15.6%. Your report is ready.

A growth diagnostic report: the verdict, three grades and the 12-month projection Sample report · B2B SaaS

What it grades against

Metrics

  • LTV:CAC
  • CAC payback
  • K-factor
  • Cohort retention at Day 1, 7, 30 and 90

Sources include

  • Bessemer State of the Cloud
  • OpenView SaaS Benchmarks
  • Amplitude
  • Pendo Product Benchmarks
  • Reforge
  • Andrew Chen
  • Adjust Benchmarks
  • AppsFlyer Reports
Where the numbers come from →

Want to talk the numbers through? Book 15 minutes with Joni →