Funnel Analysis in Apache Superset: 2026 Guide

Funnel analysis measures user conversion between successive steps of a product or purchase journey. Apache Superset offers a native Funnel Chart visualization that makes this analysis accessible. This guide details SQL modeling, visualization, and interpretation in 2026.

1. What is a funnel for?

A funnel identifies the steps where you lose the most users. A massive drop between step 2 and step 3 indicates a precise UX, technical, or communication problem to solve.

If you want to start immediately, TVL Managed Superset offers pre-configured funnel templates on Pro+ instances.

2. Typical use cases

• Onboarding: signup → activation → first value;
• E-commerce: visit → cart → checkout → payment;
• Marketing: impression → click → lead → SQL → opportunity → won;
• Support: ticket opened → first response → resolved → CSAT.

3. Data model

A typed events table:

CREATE TABLE fct_events (
  user_id INT,
  event_name VARCHAR,
  event_at TIMESTAMP,
  properties JSONB
);

4. SQL funnel query

WITH steps AS (
  SELECT
    user_id,
    MIN(CASE WHEN event_name='signup' THEN event_at END)        AS step1_at,
    MIN(CASE WHEN event_name='email_verified' THEN event_at END) AS step2_at,
    MIN(CASE WHEN event_name='first_login' THEN event_at END)    AS step3_at,
    MIN(CASE WHEN event_name='first_value' THEN event_at END)    AS step4_at
  FROM fct_events
  WHERE event_at >= CURRENT_DATE - INTERVAL '30 days'
  GROUP BY user_id
)
SELECT
  'Signup' AS step, COUNT(*) AS users FROM steps WHERE step1_at IS NOT NULL
UNION ALL
SELECT 'Email verified', COUNT(*) FROM steps
WHERE step1_at IS NOT NULL AND step2_at IS NOT NULL AND step2_at >= step1_at
UNION ALL
SELECT 'First login', COUNT(*) FROM steps
WHERE step3_at IS NOT NULL AND step3_at >= step2_at
UNION ALL
SELECT 'First value', COUNT(*) FROM steps
WHERE step4_at IS NOT NULL AND step4_at >= step3_at;

5. Funnel Chart visualization

1. Create a virtual dataset with the above query;
2. Chart type: Funnel Chart;
3. Dimension: step;
4. Metric: SUM(users);
5. Sort by: users DESC;
6. Run query → Save.

This configuration is applied by default on TVL Managed Superset, which follows community best practices.

6. Metrics to expose

Metric	Calculation
Global conversion	Final step / Initial step
Drop-off per step	(Step N − Step N+1) / Step N
Average time to convert	AVG(final_step − step_1)
Funnel by segment	Funnel broken down by cohort / channel

7. Common pitfalls

• Order not respected: a user may "skip" a step, to filter out;
• Window too small: 30-day conversion must wait 30 days, otherwise outgoing bias;
• No segmentation: a global funnel hides huge gaps per channel;
• Step poorly defined: activation must be the product's key event, not a weak proxy.

8. Going further

• Funnel by cohort: see if conversion improves over time;
• Funnel A/B test: compare two variants on the same funnel;
• Behavioral funnel: conditional on a behavior (e.g., opened the app 3 times in M+1).

9. Conclusion

Funnel analysis is one of the most actionable indicators of a product or e-commerce site. Apache Superset makes this analysis accessible with a native visualization, provided events are properly modeled upstream (with dbt or directly in SQL).

Want the benefits of Apache Superset without the friction of installation and maintenance? Deploy your instance in 3 clicks with TVL Managed Superset, hosted in Europe (OVHcloud, Roubaix, France).

For more: cohort analysis, SaaS metrics, A/B testing.