How do you analyze customer success data in Metabase?
Customer success platforms hold the relationship record: who owns each account, what it is worth, how healthy it looks, and what the team has done about it. The value in Metabase is not in rebuilding those views — it is in the join. Land the CS record in a SQL warehouse next to product usage and billing, and you can finally test whether health scores predict renewal, which behaviors precede churn, and what retention looks like measured against invoiced revenue rather than against the CS tool's own copy of it.
TL;DR — Sync accounts, dated health snapshots, and activity into the warehouse, then join them to product usage and billing. Re-derive health score components rather than trusting the vendor composite, snapshot renewal forecasts at period start, and reconcile ARR against billing — a CS platform holds a projection of CRM, billing, and product data, not the source of truth for any of them.
Which tools does this cover?
- Gainsight — companies and their scorecards, CTAs and playbooks, success plans, timeline activities
- Vitally — accounts and their health scores, users, conversations and notes, tasks and playbooks
- Planhat — companies and licenses, endusers, conversations and notes, invoices and NPS responses
Two scheduling tools share this cluster's revenue-ops data model but sit in a different category: Calendly and Cal.com are filed under Sales & CRM, because a booked meeting is a pipeline signal before it is a health one. Their guides use the same account, owner, and activity modeling as the three above.
What is the shared customer success data model?
| Concept | Common terms | Used for |
|---|---|---|
| Account | Company, customer, organization | The spine every other table hangs off |
| Health score snapshot | Scorecard, health score, score measure | Risk tiers, score movement, trend |
| Activity | CTA, task, conversation, note, timeline entry | Coverage, workload, engagement frequency |
| License or subscription | License, subscription, contract line | ARR, renewal dates, auto-renew share |
| Renewal opportunity | Renewal, contract term, forecast | Renewal exposure and forecast accuracy |
| Product usage rollup | Active users, events, feature adoption | The reality check on every health score |
The three platforms name these differently — Gainsight has companies, scorecards, and CTAs; Vitally has accounts, named scores, and tasks; Planhat has companies, licenses, and conversations — but they map onto the same six concepts. Model them once and a dashboard built for one platform survives a migration to another.
How do you connect customer success tools to Metabase?
- Managed connector route — the durable anchor where a connector exists. Read its support level and sync mode before you rely on it: several connectors in this category are full-refresh only, so deletions never propagate and yesterday's values are gone.
- Custom pipeline route — scheduled API pulls give you control over snapshot grain and cadence, which matters here because health scores and renewal dates are current-state fields. If you don't append dated snapshots, history simply never exists.
- MCP + CLI route — pull a scoped account or scorecard slice through a vendor MCP server, save CSV, and load it with
mb upload csvfor fast exploration. Gainsight and Planhat publish their own servers; Vitally does not, so its guide leads with a scripted API export instead.
What can you analyze across these tools?
- Customer health score— the composite, re-derived from components you control
- Gross revenue retention — retained revenue with expansion excluded, capped at 100%
- Net revenue retention — the same cohort including expansion, which can exceed 100%
- Churn rate — logo churn and revenue churn, always labeled as one or the other
- Renewal forecast accuracy — forecast against outcome, measured from a frozen snapshot
- Time to value — days from contract start to the first realized outcome
Which dashboards should you build?
- Customer health — risk tiers, ARR by tier, score movement, and coverage gaps
- Renewal forecast — exposure by quarter, at-risk ARR, and forecast against outcome
- Churn and retention — GRR and NRR by cohort, churn reasons, and the leading signals
- Subscription analytics — ARR movement, expansion, downgrades, and contract terms
Common mistakes
Treating the vendor health score as a metric definition.→ It is an opaque composite that gets reweighted without notice. Re-derive the components in the warehouse and keep the vendor score beside yours as a labeled column.
Reporting ARR from the CS platform.→ CS data is a projection of CRM, billing, and product data, synced on someone else's schedule. Reconcile ARR against billing, and treat a gap as a pipeline bug rather than a rounding difference.
Charting health trends from current-state fields.→ Scores and renewal dates describe today. Without dated snapshots appended on a schedule, a trend line is the same number repeated.
Measuring renewal forecast accuracy against a moving forecast.→ Freeze the forecast at period start in its own snapshot table. A forecast that keeps updating will always look accurate, and will never be useful.
Reading GRR and NRR as if they were the same metric.→ Gross revenue retention excludes expansion and can never exceed 100%; net revenue retention includes it and can. Put both on the dashboard and label which is which.
Letting a full-refresh connector stand in for history.→ Several connectors in this category replace tables wholesale. Snapshot each sync into a dated table if you want to answer any question about movement.
Related
Analytics
Integrations
Metrics
FAQ
What is customer success analytics?
Customer success analytics is reporting built on the account record — health scores, renewals, CS activity, and subscription value — modeled in a SQL warehouse and analyzed in Metabase. A CS platform answers "which accounts should my CSM call today?"; customer success analytics answers "is health predictive of renewal, what is our gross revenue retention by cohort, and which product behaviors precede churn?" See the customer success analytics overview for the full build.
Does Metabase connect natively to Gainsight, Vitally, or Planhat?
No. Metabase reads SQL databases and warehouses, so CS data has to land in one first — via a managed connector, a scheduled API pull, or a CSV loaded with the Metabase CLI. That is less of a detour than it sounds: the questions worth asking need billing and product usage tables sitting next to the CS record anyway, and those already live in your warehouse.
Should we report from the CS platform or from the warehouse?
From the warehouse, for anything that leaves the CS team. The platform is the system of action — playbooks, alerts, and the CSM's queue — and its native reports are built for that job. The warehouse is where health data joins product usage, support tickets, and invoiced revenue, which is exactly what a retention or renewal-risk question needs. Keep the platform for operating; keep the reporting where the joins are.
Can I trust the health score that comes out of the box?
Trust it as a signal, not as a metric definition. Vendor health scores are opaque composites, and they get reweighted or redefined without the analytics team hearing about it — which quietly breaks every trend built on them. Re-derive the components in the warehouse, keep the vendor score beside your own as a labeled column, and validate both against what actually happened at renewal. See customer health score for the modeling pattern.
Which tools does the customer success category cover?
Three guides: Gainsight, the enterprise platform with scorecards, CTAs, and success plans; Vitally, the mid-market B2B SaaS favorite with multiple named health scores and a clean REST API; and Planhat, built around the company object with licenses, invoices, and NPS in the same model. The two scheduling tools that share this cluster's revenue-ops data model, Calendly and Cal.com, are filed under Sales & CRM since meetings are a pipeline signal rather than a health one.
Which dashboards should you build first from CS data?
Start with the customer health dashboard — it needs only accounts and a dated health snapshot. Add the renewal forecast once contract dates and subscription values are in the warehouse, then churn and retention and subscription analytics as billing data joins the model.