What goes in a customer health dashboard?

A customer health dashboard is the shared answer to "which accounts are in trouble, and how do we know?" — the score distribution, who moved this week, and whether the score agrees with what customers actually do in the product. Build it on health-score components you re-derive in the warehouse, not on a vendor score you cannot explain.

Which cards belong on this dashboard?

• Health score distribution across the book of business
• Accounts and ARR by risk tier (green, yellow, red)
• Biggest score movers over the last 30 days
• Health score vs. weekly active users, plotted per account
• Accounts red on usage but green on score (and the reverse)
• Open risks with owner, playbook, and age
• Coverage: accounts with no touchpoint in 60+ days
• Score component breakdown for a selected account

What data does this dashboard need?

• An account dimension with segment, CSM owner, and contract dates
• Health-score components stored as columns, not just the composite
• Score history with an as-of date so movement is measurable
• Product usage at daily account grain (active users, key events)
• Touchpoints: meetings, tickets, QBRs, emails logged per account
• Open risks or playbook tasks with status and owner

How do you build it?

1. Land CS platform, CRM, billing, and product-usage data in a database and keep account IDs, contract dates, ARR, health-score components, and event timestamps.
2. Build account-grain models at the grain this dashboard needs, with health components and revenue definitions as explicit columns.
3. Create one saved question per card and certify the shared models and definitions.
4. Add dashboard filters for Segment, CSM owner, risk tier, renewal quarter, region, plan.
5. Reconcile ARR against billing rather than the CS tool, and show data freshness so every viewer knows how current the syncs are.

Example card SQL

WITH latest_score AS (
  SELECT DISTINCT ON (account_id)
    account_id,
    scored_at,
    health_score
  FROM account_health_scores
  ORDER BY account_id, scored_at DESC
), usage_28d AS (
  SELECT
    account_id,
    SUM(active_users) AS active_user_days
  FROM product_usage_daily
  WHERE usage_date >= CURRENT_DATE - INTERVAL '28 days'
  GROUP BY account_id
)
SELECT
  CASE
    WHEN s.health_score >= 70 THEN 'green'
    WHEN s.health_score >= 40 THEN 'yellow'
    ELSE 'red'
  END AS risk_tier,
  COUNT(*) AS accounts,
  ROUND(SUM(a.arr_usd), 2) AS arr,
  COUNT(*) FILTER (
    WHERE COALESCE(u.active_user_days, 0) = 0
  ) AS accounts_with_no_usage_28d
FROM cs_accounts a
JOIN latest_score s USING (account_id)
LEFT JOIN usage_28d u USING (account_id)
WHERE a.status = 'active'
GROUP BY 1
ORDER BY MIN(s.health_score);

Related

Dashboards

Integrations

Metrics

Analytics

FAQ