What goes in a cohort retention dashboard in Metabase?

A cohort retention dashboard groups customers by when they signed up and follows each cohort's revenue over time. It answers the question a blended churn rate can't: is retention actually improving for newer customers? Build it from billing data synced into a database — see Stripe or Chargebee for the connection.

What does a cohort retention dashboard look like?

Here's the layout this guide builds: lifetime value and month-12 retention at the top, then one retention curve per signup cohort with the same numbers as a cohort-by-month table, then lifetime value and cumulative revenue by cohort and channel. Read it when you want to know whether the customers you are acquiring now are better than the ones you acquired a year ago.

Which cards belong on a cohort retention dashboard?

Headline KPIs

• LTV — average lifetime value per customer
• Month-12 revenue retention
• Net revenue retention
• Average customer lifetime (months)

Cohort views

• Revenue-retention heatmap — cohort × months since signup
• Retention curves by cohort (line)
• LTV by acquisition cohort or channel
• Cumulative revenue per cohort (table)

What data does a cohort retention dashboard need?

• A monthly MRR model keyed to subscriptions.
• A signup or first-paid date per customer to define the cohort.
• Optionally an acquisition channel or plan for cohort segmentation and LTV by source.

How do you build a cohort retention dashboard?

1. Sync your billing tool into a database (Stripe or Chargebee).
2. Assign each customer a cohort month from their first paid subscription.
3. Compute MRR per cohort per months-since-signup; retention is that MRR ÷ the cohort's month-0 MRR.
4. Visualize as a heatmap and curves; add filters for channel, plan, and cohort range.

Example card SQL

-- Revenue retention by signup cohort and months-since-signup.
WITH base AS (
  SELECT
    date_trunc('month', s.started_at)                    AS cohort_month,
    date_trunc('month', m.month)                         AS active_month,
    m.mrr,
    s.customer_id
  FROM subscriptions s
  JOIN modeled_mrr m ON m.subscription_id = s.id
)
SELECT
  cohort_month,
  (EXTRACT(YEAR FROM active_month) - EXTRACT(YEAR FROM cohort_month)) * 12
    + (EXTRACT(MONTH FROM active_month) - EXTRACT(MONTH FROM cohort_month)) AS month_index,
  ROUND(SUM(mrr), 2)                                     AS mrr
FROM base
GROUP BY cohort_month, month_index
ORDER BY cohort_month, month_index;

Related

Integrations

Analytics

Dashboards

Metrics

FAQ