Mixpanel model

Funnel conversion daily

Your saved Mixpanel funnels, step by step per entry day: how many users reached each step, the conversion from the previous step and from the start, and the time it took.

Grain
one row per funnel x day (users' entry day) x step
Built from
  1. 1
    Raw tables
  2. 1
    Staging
  3. 1
    Core
  4. 1
    Analytics
Reads from Mixpanel
funnels
Ships as
Weld and dbt SQL, open source on GitHub
core.mixpanel_source.funnel_conversion_daily
1-- core.mixpanel_source.funnel_conversion_daily
2-- Your saved Mixpanel funnels, step by step per entry day: how many users reached each
3-- step, the conversion from the previous step and from the start, and the time it took.
4--
5-- Grain: one row per funnel x day (users' entry day) x step.
6-- Depends on: staging.mixpanel_source.funnel_step
7--
8-- These are Mixpanel's own funnel numbers, computed with each funnel's saved conversion
9-- window and identity rules, so they match the Funnels report for the same days.
10-- `drop_off` is users who reached the previous step and not this one.
11--
12-- To report a funnel over a week or a month, sum `users` per step and divide step by
13-- step one; averaging the daily conversion rates weights a quiet day like a busy one.
14-- Recent days keep changing until the conversion window has passed.
15
16SELECT
17    funnel_id,
18    funnel_name,
19    date,
20    step_number,
21    step_label,
22    step_event,
23    users,
24    FIRST_VALUE(users) OVER w                               AS users_entered,
25    LAG(users) OVER w - users                               AS drop_off,
26    step_conversion,
27    overall_conversion,
28    avg_seconds_from_previous_step,
29    avg_seconds_from_start
30FROM {{staging.mixpanel_source.funnel_step}}
31WINDOW w AS (PARTITION BY funnel_id, date ORDER BY step_number)

This model and the 2 it depends on. Deploy them together; refs like {{staging.…}} resolve to the matching folders in Weld, and the dbt view uses source() and ref().

Notes

How it counts. And where it stops.

These are Mixpanel's own funnel numbers, computed with each funnel's saved conversion window and identity rules, so they match the Funnels report for the same days. `drop_off` is users who reached the previous step and not this one. To report a funnel over a week or a month, sum `users` per step and divide step by step one; averaging the daily conversion rates weights a quiet day like a busy one. Recent days keep changing until the conversion window has passed.

Raw tables it reads

  • raw.mixpanel_source.funnels

Put Mixpanel in front
of your agents.

Connect Mixpanel, deploy the models, and ask questions in Claude, Cursor or ChatGPT. Free for 14 days, no credit card.