# LTV Cohort Calculator MCP

> The LTV Cohort Calculator determines the long-term value of specific customer groups based on their acquisition date. It analyzes historical spending trends, calculates accumulated Customer Lifetime Value (LTV) at set time points, and projects future revenue using statistical models. This lets you understand which initial cohorts are driving your most stable, predictable growth.

## Overview
- **Category:** data-analytics
- **Price:** Free
- **Tags:** ltv, cohort-analysis, revenue-projection, customer-value, lifetime-value

## Description

Figuring out the true worth of a customer group goes way beyond looking at last month's sales numbers. You need to know what happened to customers who signed up back in January versus those who joined last week. This MCP lets you calculate and project Customer Lifetime Value (LTV) for any specific acquisition cohort.

It first pulls the raw, accumulated monthly revenue history for a group of users defined by their sign-up date. Then, it calculates how much that group has generated up to key milestones—say, 6 months or 12 months after joining. For long-term planning, you can feed this historical data into the system to generate LTV projections out to thirty-six months using both linear and logarithmic methods.

Because financial modeling requires absolute certainty, everything that passes through this MCP is tracked by Vinkius AI Analytics. This means every tool call, from fetching raw metrics to projecting future revenue, generates a cryptographically signed audit trail. You always know exactly how the data moved and what inputs drove the final LTV number.

## Tools

### calculate_cumulative_ltv
Determines the total accumulated LTV for a cohort up to specific, fixed months after acquisition.

### fetch_cohort_metrics
Retrieves raw accumulated monthly revenue data for a given customer group and historical time frame.

### generate_ltv_projections
Projects the LTV out to 36 months using linear or logarithmic statistical methods based on provided history.

## Prompt Examples

**Prompt:** 
```
Please analyze the '2023-01' cohort. First, fetch all historical revenue data for 36 months, and then calculate its LTV at 3, 6, and 12 months.
```

**Response:** 
```
First, I will call `fetch_cohort_metrics(acquireMonth='2023-01', cohortLengthMonths=36)` to get the full revenue history. Then, using that data, I will run `calculate_cumulative_ltv(cohortIdentifier='2023-01', targetMilestones=[3, 6, 12])`.
```

**Prompt:** 
```
I need a long-term forecast for the '2022-11' cohort. Project its LTV to 36 months using both linear and logarithmic methods.
```

**Response:** 
```
To project the LTV, I must first compile the historical data by calling `fetch_cohort_metrics(acquireMonth='2022-11', cohortLengthMonths=36)`. Once I have the full metrics JSON array, I will pass it to `generate_ltv_projections` for the final forecast.
```

**Prompt:** 
```
What is the current LTV status of the '2023-04' cohort at 24 months?
```

**Response:** 
```
I will use the `calculate_cumulative_ltv` tool, specifying `cohortIdentifier='2023-04'` and `targetMilestones=[24]`. This provides the accurate cumulative LTV up to the 24-month mark.
```

## Capabilities

### Fetch Raw Cohort Data
You can pull the complete history of accumulated monthly revenues for any specified group of customers.

### Calculate Milestone LTV
The system computes the total, cumulative LTV at specific, pre-set points in time for a cohort.

### Predict Long-Term Value
You can forecast LTV up to 36 months out using both linear and logarithmic extrapolation methods.

### Identify Cohort Health
Determine if a cohort is performing better or worse than expected at various points in its lifecycle.

## Use Cases

### We need to know if our Q3 marketing push was worth the spend.
A Growth Manager runs `fetch_cohort_metrics(acquireMonth='2024-07', cohortLengthMonths=36)` to pull all historical revenue for the July cohort. They then pass that data into `calculate_cumulative_ltv` to prove the group hit a strong 12-month value, justifying a higher ad spend next quarter.

### Our current monthly reporting is misleading about long-term health.
A Product Lead uses `generate_ltv_projections` on an older cohort to see how the revenue curve stabilizes over 36 months. This shows that while month-to-month numbers are volatile, the actual retained value remains high.

### We need to compare two completely different customer bases.
A Finance Analyst runs `fetch_cohort_metrics` for both '2022-10' and '2023-05'. By comparing the raw data, they can spot fundamental differences in spending patterns that single dashboards hide.

### We just launched a premium tier and need to prove its value.
The team uses `calculate_cumulative_ltv` on all users who adopted the new tier. By checking milestones at 6 months, they can quantify the increased revenue contribution from the upgrade.

## Benefits

- Pinpoint retention issues. Instead of just seeing a dip in total revenue, you can use `fetch_cohort_metrics` to show exactly *when* and *where* a specific acquisition group's spending starts to drop off.
- Build better budgets. Run the full 36-month forecast using `generate_ltv_projections`. This moves your team past guessing and gives you data for long-term financial planning.
- Measure marketing ROI accurately. Compare cohorts from different campaigns (e.g., 'Q1 Paid' vs. 'Q2 Organic') to prove which acquisition source provides the highest sustained, lasting value.
- Understand maturity curves. Use `calculate_cumulative_ltv` to see how a cohort’s value builds up over time at set checkpoints like 3 months or 12 months post-signup.
- Audit your numbers. The platform tracks every single data point that flows through, giving you an unbreakable audit trail for financial reporting.

## How It Works

The bottom line is you get clear, auditable financial figures showing how valuable your initial customer groups are today and years down the road.

1. First, input the acquisition month and optionally specify how many months of historical data you need to pull.
2. Next, use that raw revenue history to either calculate LTV at specific milestones or generate a full 36-month projection.
3. You receive the calculated LTV figure, showing the total accumulated value for that customer group over time.

## Frequently Asked Questions

**How does `calculate_cumulative_ltv` work?**
It takes a cohort identifier and a list of target months. It then calculates the total accumulated value for that group specifically at those requested milestones, avoiding guesswork.

**Can I use this MCP to forecast beyond 36 months?**
No, the current tools are capped at a 36-month projection limit. For longer timelines, you'd need a custom financial modeling setup.

**What data does `fetch_cohort_metrics` require?**
You must provide an acquisition month in YYYY-MM format and specify the total number of historical months you want to retrieve for that cohort.

**Is the LTV calculation accurate enough for investor decks?**
Yes. Because every step, from fetching raw metrics to running the projection, is recorded in a cryptographically signed audit trail by Vinkius, you have full visibility and confidence in the data's integrity.

**How is the revenue data handled and secured when I use `fetch_cohort_metrics`?**
The MCP uses a zero-trust proxy for all data calls. Your credentials pass through in transit but are never stored on disk, and every action generates a cryptographically signed audit trail.

**If I send bad input to `generate_ltv_projections`, what happens?**
The agent receives an immediate validation error. This message pinpoints exactly which field in the required historical metrics JSON array is malformed or missing, so you know precisely where to fix your data.

**Can I use the output from `calculate_cumulative_ltv` for other automations?**
Yes. The result of calculating cumulative LTV provides structured, actionable data. This means any subsequent agent or MCP can read that output and build complex, multi-step workflows around it.

**What if I try to run too many calls to `fetch_cohort_metrics` quickly?**
The Vinkius platform manages throttling for you. If your usage approaches a limit, the agent receives an immediate rate-limit error code. This allows your script or workflow to implement smart retries without failing completely.

**How do I get the raw historical revenue data needed for LTV calculations?**
Use the `fetch_cohort_metrics` tool. This function retrieves the accumulated monthly revenues for a specific cohort, providing the foundational time-series data required for all subsequent LTV analyses.

**What is the difference between calculating LTV at fixed milestones (e.g., 6 months) and projecting future value?**
First, use `calculate_cumulative_ltv` to determine the exact LTV at fixed points (like 3 or 6 months). Then, if you need a longer-term forecast, feed those historical metrics into `generate_ltv_projections`. The projection tool uses advanced math to estimate value beyond observed data.

**What inputs are required for LTV projections?**
The `generate_ltv_projections` tool requires two key pieces of information: the unique cohort identifier and a structured JSON array containing historical revenue metrics. This data is typically sourced from running `fetch_cohort_metrics` first.