• BENCHMARK
  • COMPARISON
  • ANALYSIS
  • HUMAN

How to calculate accelerator performance scores with your AI

Connect your data once. Your AI handles the math to create a standardized benchmark for every cohort.

Ask AI about this page

Short answer

How can I get a normalized performance score for my accelerator?

Your AI uses the compute_track_record_score tool to process your raw cohort data into a single benchmark. This gives you a consistent way to compare different years or programs side-by-side. You'll get a clear breakdown of how each cohort performed.

The scoring outcomes.

Where the data lands.

The AI processes your raw data into these specific outputs.

BENCHMARK

Normalized score

The AI uses compute_track_record_score to generate a single number for each cohort. This lets you compare different years on the same scale.

COMPARISON

Cohort rankings

Your agent sorts every cohort based on the new score. You can see exactly which years outperformed others instantly.

ANALYSIS

Trend identification

The AI looks for patterns in the scores over time. It identifies if your program quality is rising or falling.

HUMAN

Strategic decisions

You take the final scores and decide how to adjust your selection criteria or curriculum. The AI provides the math, but you provide the direction.

The workflow

What your AI does when the data arrives.

The AI moves from raw data to a finished benchmark through these steps.

  1. Data ingestion

    Your agent reads your historical cohort data, including funding amounts and exit values.

    compute_track_record_score

  2. Normalization

    The AI applies the scoring logic to ensure different cohort sizes don't skew the results.

    compute_track_record_score

  3. Score generation

    The tool calculates the final performance metric for every group in your dataset.

    compute_track_record_score

  4. Reporting

    Your agent presents the final scores in a format you can use for your next board meeting.

    compute_track_record_score

Try it

Copy these to start.

Use these prompts to kick off the calculation process.

Starting points

Replace the bracketed text with your actual file names or data sources.

Accelerator Track Record Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_ycqMOekfQi9B5f3SZDHLpw46siRrpNpBThWadHSB/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — Accelerator Track Record capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "accelerator-track-record-metrics-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_ycqMOekfQi9B5f3SZDHLpw46siRrpNpBThWadHSB/mcp"
    }
  }
}
Copy into chat04

  • Calculate the performance scores for the cohorts in my '2023_data.csv' file.

  • Compare the normalized scores between the 'Summer 2022' and 'Winter 2023' cohorts.

  • Based on the scores in my 'accelerator_metrics.xlsx' sheet, which year had the best ROI?

  • Generate a ranking of all cohorts listed in my recent data upload.

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect Accelerator Track Record Metrics once, then ask.

Just link the MCP to your preferred client. Your credentials stay encrypted, and you can start asking questions immediately.

Connect Accelerator Track Record Metrics to your AI

FAQ

How the scoring works.

  • 01

    Can the AI change my original data files?

    No. The AI only reads your data to perform calculations. It cannot modify or delete your source files.

  • 02

    Does the AI decide which cohorts were 'successful'?

    No. The AI calculates the score based on the math provided by the tool. You decide how to interpret those results.

  • 03

    What kind of data do I need to provide?

    You need to provide data that includes cohort identifiers and the specific performance metrics you want to benchmark.

  • 04

    Can I use this with Claude?

    Yes, as long as you are using an MCP-compatible client like Claude Desktop.

  • 05

    Will this work if my data is in a Google Sheet?

    Yes, once you provide the data to your agent, it can run the scoring tool on that information.

  • More questions about Accelerator Track Record Metrics? The Connector page answers them. See everything the Accelerator Track Record Metrics Connector can do

Connect Accelerator Track Record to Claude, Cursor, ChatGPT & more