What are net new leads, and how do you measure them in Metabase?

Net new leads is the count of genuinely new leads created in a period — leads created minus duplicates, junk, and disqualified records. It's the honest version of top-of-funnel volume: the number that should agree with cost per lead and hold up when sales audits it. Measure it in Metabase from CRM data synced from HubSpot, Salesforce, or Pipedrive.

What does a net new leads chart look like in Metabase?

Chart net new leads — created minus merged and disqualified — per month as bars, and trust the growth because the junk is already subtracted. A one-month drop like Feb 2026's often traces to a dedup cleanup or a paused campaign rather than shrinking demand, so check the merge counts before sounding the alarm.

What net new leads measures

It measures how many new potential buyers actually entered the funnel. That makes it the volume input for everything downstream — conversion rates need a trustworthy denominator, capacity plans need a real inflow number, and channel comparisons need each source's count cleaned the same way. It also keeps marketing honest with itself: a list import or a viral junk-form day inflates leads created and leaves net new leads unmoved, which is exactly the point.

What counts as net new

Three rules define the metric, and all three belong in a versioned SQL model rather than in each analyst's head. First, the dedup level: same email, same person, or same account — each stricter level yields a smaller, more sales-shaped number. Second, the exclusion list: merged duplicates, spam and test submissions, competitors, and out-of-ICP records marked disqualified. Third, the timing convention: count the lead in its creation month regardless of when it's later merged or disqualified, or accept that history restates as hygiene catches up — either works, but the dashboard must say which. Teams that skip this step get a trend line that quietly shrinks backward every time someone cleans the CRM.

What data does it need?

• A CRM leads table with created_at, source, status, and a merge pointer like merged_into_id — HubSpot, Salesforce, and Pipedrive all expose these through their standard sync schemas.
• A meetings or activities table keyed by lead for follow-through analysis.
• A fixed source-attribution rule, so "organic" and "paid" mean the same thing in every month of the trend.

SQL patterns

SELECT
  date_trunc('month', created_at) AS month,
  source,
  COUNT(*) AS leads_created,
  COUNT(*) FILTER (WHERE merged_into_id IS NOT NULL) AS duplicates,
  COUNT(*) FILTER (WHERE status = 'disqualified') AS disqualified,
  COUNT(*) FILTER (
    WHERE merged_into_id IS NULL
      AND status <> 'disqualified'
  ) AS net_new_leads
FROM leads
WHERE created_at >= CURRENT_DATE - INTERVAL '12 months'
GROUP BY 1, 2
ORDER BY 1, 2;

WITH net_new AS (
  SELECT
    id,
    date_trunc('month', created_at) AS cohort_month
  FROM leads
  WHERE merged_into_id IS NULL
    AND status <> 'disqualified'
    AND created_at >= CURRENT_DATE - INTERVAL '12 months'
),
met AS (
  SELECT DISTINCT lead_id
  FROM meetings
  WHERE status = 'held'
)
SELECT
  n.cohort_month,
  COUNT(*) AS net_new_leads,
  COUNT(m.lead_id) AS leads_with_meeting,
  ROUND(
    100.0 * COUNT(m.lead_id) / NULLIF(COUNT(*), 0), 1
  ) AS lead_to_meeting_pct
FROM net_new n
LEFT JOIN met m ON m.lead_id = n.id
GROUP BY 1
ORDER BY 1;

Pitfalls

Where this metric applies

• HubSpot + Metabase — contacts with lifecycle stages and merge history
• Salesforce + Metabase — lead objects with status and converted flags
• Pipedrive + Metabase — persons and deals for source-level funnels
• Attio + Metabase — people and companies for account-level dedup

Related

Metrics

Dashboards

FAQ