What is win rate, and how do you measure it in Metabase?

Win rate is the share of deals that close won out of the deals that reach a decision. It's the headline efficiency metric for a sales team — but it's only meaningful once you fix what goes in the denominator. Measure it in Metabase from CRM data synced into a database (HubSpot, Salesforce, Pipedrive, and others).

How is win rate defined?

The formula is simple; the definition of the denominator is where teams disagree:

• Won ÷ closed — of the deals that reached a decision (won or lost), how many were won. This is the most common and most stable.
• Won ÷ created — of all deals created in a cohort, how many eventually won. Lower and slower to settle, but useful for top-of-funnel quality.

Both are valid. What breaks reporting is switching between them, or leaving it ambiguous.

Choosing and holding a cohort

A win rate is meaningless without a cohort. Decide whether you're grouping byclose date (deals that closed in the period) or creation date (deals created in the period, measured once they settle), and keep it consistent.

• Close-date cohorts answer "how are we converting deals right now?" — good for operational tracking.
• Creation-date cohorts answer "how good was this batch of pipeline?" — good for lead-quality analysis, but recent cohorts are incomplete.

What data does win rate need?

• A deal/opportunity table with a clear won/lost status.
• A closed_at (or created_at) timestamp for the cohort.
• Segmentation columns: pipeline, source, owner, segment, and stage entered.
• A consistent rule for excluding open deals from the denominator.

SQL patterns

SELECT
  date_trunc('month', closed_at) AS month,
  COUNT(*) FILTER (WHERE is_won)                 AS won,
  COUNT(*) FILTER (WHERE is_closed)              AS closed,
  ROUND(
    100.0 * COUNT(*) FILTER (WHERE is_won)
      / NULLIF(COUNT(*) FILTER (WHERE is_closed), 0),
    1
  ) AS win_rate_pct
FROM modeled_deals
WHERE is_closed
  AND closed_at >= CURRENT_DATE - INTERVAL '12 months'
GROUP BY 1
ORDER BY 1;

SELECT
  pipeline,
  source,
  COUNT(*) FILTER (WHERE is_won)    AS won,
  COUNT(*) FILTER (WHERE is_closed) AS closed,
  ROUND(
    100.0 * COUNT(*) FILTER (WHERE is_won)
      / NULLIF(COUNT(*) FILTER (WHERE is_closed), 0),
    1
  ) AS win_rate_pct
FROM modeled_deals
WHERE is_closed
GROUP BY pipeline, source
ORDER BY closed DESC;

Pitfalls

Where this metric applies

• HubSpot + Metabase — deal stage and is_closed_won
• Salesforce + Metabase — opportunity IsWon/IsClosed
• Pipedrive + Metabase — deal status (won/lost)
• Close + Metabase — opportunity status_type

Related

Analytics

Metrics

FAQ