# Retention Analytics MCP > Cohort Retention Analytics MCP processes user data to map out retention curves and calculate average user lifetime. This tool determines exactly when your users start churning, helping you benchmark performance against industry standards for SaaS, E-Commerce, or Gaming models. ## Overview - **Category:** product-management - **Price:** Free - **Tags:** retention, cohorts, analytics, saas, benchmarking ## Description Understanding why users leave is key to product health. This connector takes raw cohort data and turns it into specific actions. Instead of guessing where the biggest leaks are, your agent generates clear retention curves showing how a group of users decays over time. You'll know exactly if you're meeting industry standards—whether you run an e-commerce site or B2B SaaS. It also calculates the expected average lifetime value for any user cohort. When you connect this MCP via Vinkius, your agent can compare your current metrics against predefined benchmarks for multiple industries, giving immediate feedback on where to focus product efforts. ## Tools ### calculate_average_lifetime Predicts the total number of months a user is expected to remain active in your service. ### compare_performance_to_benchmark Checks if a specific metric you provide performs well against known industry standards for your product category. ### evaluate_retention_milestone Retrieves the exact retention percentage for any user cohort at a specified point in time. ### calculate_retention_curve Generates a data sequence that maps out how rapidly your users lose activity over successive months. ## Prompt Examples **Prompt:** ``` Calculate the retention curve for a cohort with these monthly rates: 1.0, 0.8, 0.6, 0.4. ``` **Response:** ``` The retention curve points are: Month 0: 100%, Month 1: 80%, Month 2: 60%, Month 3: 40%. ``` **Prompt:** ``` Is a 25% retention rate for E-Commerce considered good? ``` **Response:** ``` The status is 'At Risk' because the benchmark for E-Commerce is 0.30 (30%). ``` **Prompt:** ``` What is the expected lifetime if my retention rates are 1.0, 0.5, and 0.2? ``` **Response:** ``` The calculated average lifetime is 1.7 months. ``` ## Capabilities ### Map User Decay Rates Generates a sequence of data points that show how quickly a group of users loses activity over months. ### Forecast Average User Value Calculates the expected number of months an average user will remain active and engaged with your product. ### Check Milestone Performance Determines if a specific retention rate (like 3-month or 6-month) meets industry expectations for your category. ### Compare Against Benchmarks Compares any metric you provide against hardcoded standards for various industries, flagging gaps immediately. ## Use Cases ### The 'Early Drop-Off' Problem A PM notices a high sign-up rate but low 30-day activity. They ask their agent to run `evaluate_retention_milestone` for the 1-month mark, identifying that users drop off sharply after onboarding, leading them to prioritize better first-week feature guidance. ### The 'Is This Good?' Question A Growth Lead gets a 20% retention rate and wonders if it's acceptable. They use `compare_performance_to_benchmark` to confirm their industry standard is 30%, immediately showing they have a clear gap to fix. ### The 'How Long Will They Stay?' Question A Product Owner needs to justify long-term feature investment. They ask the agent to calculate the average user lifetime, giving them a 2.5-year projection that validates the roadmap's financial viability. ### The 'What If?' Scenario A Data Analyst wants to model the effect of losing specific features. They use `calculate_retention_curve` with hypothetical decay rates, visualizing how quickly their user base would shrink without a core product pillar. ## Benefits - Stop guessing about churn. Use `calculate_retention_curve` to generate a data map that shows the exact monthly decay rate, letting you pinpoint retention leaks. - Instantly gauge performance against peers. The `compare_performance_to_benchmark` tool checks your current metrics (e.g., 3-month retention) against industry standards like SaaS or E-commerce. - Forecast future value with confidence. Calculating the average user lifetime tells you how long your investment in a new user is actually worth. - Target specific product fixes. The `evaluate_retention_milestone` tool lets you check if users are surviving key periods, like 60 or 120 days post-signup. - Focus efforts where they count. By combining the decay curve with benchmark checks, you move beyond simple reporting to actionable resource allocation. ## How It Works The bottom line is that your agent takes complex retention numbers and outputs simple answers about where you need to focus. 1. You feed the MCP raw cohort data or a specific retention rate to your agent. 2. The service processes this data, running calculations to generate curves and compare metrics against industry standards. 3. Your agent receives actionable insights, such as a decay curve visualization or a 'performance status' (e.g., At Risk) compared to the benchmark. ## Frequently Asked Questions **How does calculate_average_lifetime work with Vinkius?** It predicts the total expected active time for a user cohort by analyzing your retention rates. By connecting through Vinkius, you'll have access to this analysis alongside hundreds of other data services. **What kind of metrics does compare_performance_to_benchmark use?** It compares specific figures—like 3-month retention or average LTV—against hardcoded industry standards for categories like SaaS, E-Commerce, and Gaming. **Can I check a specific month using evaluate_retention_milestone?** Yes. You can input any desired time point (the milestone) to see the exact retention percentage for your cohort at that moment without needing to calculate it manually. **Is calculate_retention_curve better than just viewing raw data?** The curve is much better. It smooths out random noise and shows a clear, predictable decay pattern over time, making the drop-off points obvious to diagnose. **What input data structure does `calculate_retention_curve` require?** It requires a sequence of raw percentage inputs, representing decay over time. You must provide these as an ordered array or list; simply describing the rates is not enough for calculation. **What happens if I run `calculate_average_lifetime` with insufficient data points?** The tool will return a specific error message indicating that it needs more time periods to calculate an accurate average. Providing just two or three months of activity won't give you a reliable estimate. **Can `compare_performance_to_benchmark` adjust for different product verticals?** No. The tool uses hardcoded industry standards linked to specific categories like SaaS and E-Commerce. If your niche falls outside those predefined sectors, you'll have to manually adjust the benchmark percentage. **When using `evaluate_retention_milestone`, how precise is the time frame? Can I check for quarterly milestones?** It checks retention against a specific point in time, which means it works best with month-over-month data. While you can request broader periods, the calculation relies on accurate monthly inputs.