# Analyze Startup Program Performance Metrics AI Agent Connect

> Accelerator Graduation Metrics calculates deep performance insights for startup cohorts. This MCP lets your AI client analyze graduation rates, drop rates, and time-to-graduation distributions. You can identify exactly why companies exit programs and how program intensity affects outcomes. Use the tools to get high-level success percentages, understand the full time-to-graduation spread, or analyze the core reasons for attrition in any specific cohort.

## Overview
- **Category:** education
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_v5RiOHzHplcpbABx1uMTKTedSHZUAJqsedikFb3r/ai-agent-connect
- **Tags:** accelerator, metrics, attrition, graduation, cohort-analysis

## Description

Need to prove the ROI of your accelerator program? This MCP gives you the data. It lets your AI agent dig into cohort performance, moving beyond simple success rates. You can calculate graduation rates and drop rates, and even map out the full distribution of time it takes for companies to graduate. The tools help you identify the root causes of attrition, letting you know if the problem is selection fit, funding, or something else. It’s the data you need to adjust program intensity or prove value to stakeholders. Just connect your AI client to the Vinkius catalog and start running these complex analyses immediately.

## Tools

### analyze_attrition_drivers
This tool analyzes the primary reasons companies drop out of a specific cohort, giving you insight into why they exit the program.

### get_completion_time_distribution
Use this tool to get the full spread of completion times for a cohort, showing you the average and median graduation timelines.

### get_graduation_metrics
This tool provides high-level success and attrition percentages for a specific cohort, giving you an immediate snapshot of program health.

## Prompt Examples

**Prompt:** 
```
What is the graduation rate for cohort C2023?
```

**Response:** 
```
The graduation rate for cohort C2023 is 85%, with a drop rate of 15%.
```

**Prompt:** 
```
How long does it take for companies in cohort C2023 to graduate?
```

**Response:** 
```
The average completion time for cohort C2023 is 6 months, with a median of 5.5 months.
```

**Prompt:** 
```
Why are companies dropping out of cohort C2023?
```

**Response:** 
```
The primary reason for attrition in cohort C2023 is poor selection fit, with a correlation score of 0.75.
```

## Capabilities

### Calculate Success Rates
Your agent uses this to get immediate, high-level percentages for graduation and attrition for any given cohort.

### Analyze Drop-Off Causes
This tool helps you pinpoint the primary reasons companies are leaving the program, moving beyond just knowing they left.

### Map Completion Timelines
You can get the full distribution of time-to-graduation, showing the average, median, and spread for a cohort.

## Use Cases

### Quarterly Program Review
You need to present to investors. Use the MCP to pull graduation rates and attrition drivers for the last quarter's cohort.

### Curriculum Adjustment
The data shows high attrition due to 'poor selection fit.' You use the MCP to prove that the initial screening process needs tightening.

### Benchmarking Cohorts
Compare the completion time distribution of Cohort A versus Cohort B to see if a change in program structure actually sped up graduation.

## Benefits

- You get concrete metrics on program success, moving beyond qualitative reports.
- The MCP identifies specific drivers of attrition, allowing you to fix the root problem.
- You can map the full time-to-graduation distribution, giving a clearer picture than just an average.

## How It Works

Connecting is simple. You connect your AI client to the Vinkius catalog, and the MCP is immediately available. You then prompt your agent with a specific question about a cohort, and the MCP runs the necessary calculations.

1. Connect your preferred AI client (Claude, Cursor, etc.) to the Vinkius catalog.
2. Specify the cohort ID and the metric you need (e.g., 'graduation rate').
3. Your agent invokes the appropriate tool within the MCP.
4. The MCP returns the calculated, data-backed metric directly to your prompt.

## Frequently Asked Questions

**Does this MCP handle data for multiple cohorts?**
Yes. All the tools are designed to accept a specific cohort identifier, allowing you to compare performance across different groups or time periods.

**What kind of data does it provide about attrition?**
It doesn't just say companies left. The analyze_attrition_drivers tool identifies the primary reasons for exit, such as poor selection fit or funding issues, giving you actionable data.

**Is this better than using a spreadsheet?**
It's faster and more reliable. The MCP handles the complex statistical calculations and distributions instantly, which is difficult to manage manually in a spreadsheet.

**Can I find the average completion time?**
Yes. The get_completion_time_distribution tool provides both the average and the median completion time, giving you a much clearer picture of the timeline spread.