What is sales cycle length, and how do you measure it in Metabase?

Sales cycle length is how long a deal takes to close — typically from creation (or first activity) to won. It's the core velocity metric for a sales team, and it's heavily right-skewed, so how you summarize it matters as much as how you measure it. Track it in Metabase from CRM data synced into a database (HubSpot, Salesforce, Pipedrive, and others).

How is sales cycle length defined?

The elapsed days between a deal's start and its win. Teams vary on the start point:

• Created → won — simplest and most common; uses the deal's creation timestamp.
• First activity → won — starts the clock at genuine engagement, ignoring dormant records.
• Qualified → won — starts when a deal enters a qualified stage; needs stage history.

Pick one start definition and apply it consistently.

Why sales cycle length needs stage history

A current CRM snapshot tells you where a deal is now, not when it entered or left each stage. Without stage-change history you can compute overall cycle length (created → won) but nottime-in-stage or where deals stall. Sync the stage/deal-change log:

• HubSpot: deal property history (stage change timestamps).
• Salesforce:OpportunityFieldHistory orOpportunityHistory for stage transitions.
• Pipedrive: deal flow / changes log.
• Close: opportunity status changes (webhooks or activity log).

What data does sales cycle length need?

• A created_at (or first-activity) timestamp for the start.
• A closed_at / won timestamp and a clear "won" definition.
• For time-in-stage: full stage-change history per deal.
• Segmentation columns: pipeline, segment, deal size band, owner.

SQL patterns

SELECT
  date_trunc('month', closed_at) AS month,
  percentile_cont(0.5) WITHIN GROUP (
    ORDER BY EXTRACT(EPOCH FROM (closed_at - created_at)) / 86400.0
  ) AS median_cycle_days,
  percentile_cont(0.9) WITHIN GROUP (
    ORDER BY EXTRACT(EPOCH FROM (closed_at - created_at)) / 86400.0
  ) AS p90_cycle_days
FROM modeled_deals
WHERE is_won
GROUP BY 1
ORDER BY 1;

-- Time-in-stage from stage-change history
WITH ordered AS (
  SELECT
    deal_id,
    stage,
    changed_at,
    LEAD(changed_at) OVER (
      PARTITION BY deal_id ORDER BY changed_at
    ) AS next_changed_at
  FROM modeled_stage_history
)
SELECT
  stage,
  percentile_cont(0.5) WITHIN GROUP (
    ORDER BY EXTRACT(EPOCH FROM (COALESCE(next_changed_at, now()) - changed_at)) / 86400.0
  ) AS median_days_in_stage
FROM ordered
GROUP BY stage
ORDER BY median_days_in_stage DESC;

Pitfalls

Where this metric applies

• HubSpot + Metabase — deal create date and property history
• Salesforce + Metabase — CreatedDate and OpportunityFieldHistory
• Pipedrive + Metabase — add_time and deal flow
• Close + Metabase — date_created and status changes

Related

Analytics

Metrics

FAQ