All skills
nimrodfisher avatar

/cohort-analysis

@2e18ac4

Time-based cohort analysis with retention and behaviour tracking. Activate when you need to measure how groups of users/customers behave over time — retention rates, revenue by cohort, or feature adoption curves.

Use this Skill: https://skilld.dev/gh/nimrodfisher/data-analytics-skills/cohort-analysis

This session only. Nothing lands on disk.

referencesretention_metrics_glossary.md

≈929 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Retention Metrics Glossary

Definitions for retention metrics used in cohort analysis. Use these as the canonical definitions when writing reports.


Core Retention Metrics

Day-N / Week-N / Month-N Retention

The percentage of users from a cohort who were active on (or within) period N after joining.

N-period Retention = (Users active in period N) / (Cohort size at period 0) × 100

Variants:

  • Unbounded (on or after): User is retained if they were active any time up to period N
  • Bounded (exactly at N): User is retained only if they were active within period N (the standard definition)

Always specify which variant you're using.


D1 / D7 / D30 Retention (Day-based)

Common benchmarks for consumer apps. Measured using bounded windows.

Product Category D1 Target D7 Target D30 Target
Top-quartile mobile apps > 40% > 20% > 10%
Median mobile apps ~25% ~10% ~5%
SaaS / B2B products > 70% (W1) > 50% (W4) —

Benchmarks vary widely by product type — use industry-specific sources.


Rolling Retention

A user is counted as retained in period N if they were active at any point from period N onwards.

Rolling Retention (N) = Users who returned on period N or later / Cohort size

Rolling retention never decreases — it can only stay flat or increase as you extend the window. Useful for showing "did this user ever come back?"


Week-over-Week (WoW) Retention

Percentage of weekly active users from week N who are also active in week N+1.

WoW Retention = WAU(this week) who were also WAU(last week) / WAU(last week)

Month-over-Month (MoM) Retention

Same concept for monthly active users.

MoM Retention = MAU(this month) who were also MAU(last month) / MAU(last month)

Revenue Retention Metrics

Gross Revenue Retention (GRR)

The percentage of recurring revenue retained from existing customers, excluding expansion.

GRR = (Starting MRR − Churn MRR − Contraction MRR) / Starting MRR × 100

GRR is capped at 100% (cannot exceed starting revenue).

Best-in-class: > 90% for SaaS; > 85% for SMB-focused SaaS.


Net Revenue Retention (NRR) / Net Dollar Retention (NDR)

Revenue retained including expansion (upsell/cross-sell), minus churn and contraction.

NRR = (Starting MRR − Churn MRR − Contraction MRR + Expansion MRR) / Starting MRR × 100

NRR can exceed 100% if expansion outweighs churn.

Best-in-class: > 120% for enterprise SaaS; > 100% for any healthy SaaS.


Logo Retention / Customer Retention Rate (CRR)

The percentage of customers (accounts, not revenue) retained over a period.

CRR = (Customers at end − New customers acquired) / Customers at start × 100

Derived Metrics

Churn Rate

Monthly Churn Rate = Churned customers this month / Customers at start of month × 100

Relationship: Monthly Retention Rate + Monthly Churn Rate = 100%

Expected Lifetime

Expected Lifetime (months) = 1 / Monthly Churn Rate

Example: 5% monthly churn → 20-month expected lifetime.

Lifetime Value (LTV) via Retention

LTV = ARPU × (1 / Churn Rate)   [for constant churn rate]

Period Labels Explained

Label Meaning Example (monthly)
Period 0 The cohort's first period — baseline (always 100%) January cohort in January
Period 1 One period after joining January cohort in February
Period N N periods after joining January cohort in month N+1

Period 0 should always be 100% unless there is a data quality issue (users who had the retention event before their cohort date).

Source: SKILL.md on GitHub

No alerts16d4 checks · Risk SAFE
  • Gen Agent Trust Hub16d

    The skill provides comprehensive cohort analysis capabilities but includes an HTML heatmap template that is vulnerable to Cross-Site Scripting (XSS) if populated with unsanitized data labels from external sources.

  • Socket16d

    No alerts

  • Snyk16d

    Risk: LOW · No issues

  • ZeroLeaks5mo

    Score: 93/100 · 2 sections analyzed

Signed by skilld at 2e18ac4. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 5 days ago.

Activeupdated 5 months ago

README badge

README badge for nimrodfisher/data-analytics-skills/cohort-analysis