I'm calculating the following, but I'm not sure how to best calculate C and get to use it in a formula without calculating it in a formula itself. A: Users who did Event A followed by Event B B: Users who did Event A followed by Event C C: Users who did Event A but not Event B or C D Total users who did Event A E: (B + C) / D How is the best way for me to get C?
You can’t define C directly as “A but not B or C” inside a Formula; you first need to get that user set via Funnels or Cohorts, then use it in a Formula. From the docs, the only explicit “did not do” logic on sequences is:
Funnels exclusion steps: exclude users who did a specific event between funnel steps. [Funnels exclusion]
Cohorts with “did not do”: cohorts can be built on “did not do” conditions. [Cohorts FAQ]
Given that, the most reliable way to get C is:
Create a cohort for D (users who did Event A)
Users → Create cohort: “did Event A” in your desired time range. [Cohorts FAQ]
Create cohorts for A and B
A-cohort: users who did Event A then Event B (via a Funnel, then “View Users → Create Cohort”, or via Users page sequence). [Cohorts FAQ]
B-cohort: users who did Event A then Event C (same approach).
Define C as a cohort using AND / NOT logic
In the Cohorts builder, create a new cohort:
Condition 1: in cohort “D (did Event A)”
Condition 2: NOT in cohort “A (A→B)”
Condition 3: NOT in cohort “B (A→C)”
Use AND between these conditions. [Cohorts FAQ]
Use A, B, C, D in an Insights Formula
In Insights, add metrics for each cohort (e.g., Count users in cohort A, B, C, D). [Measurements]
Add a Formula metric: E = (B + C) / D. [Insights formulas]
The knowledge sources don’t describe a way to compute “C” purely as a one-line Formula without first defining it via cohorts or funnel segments, so the cohort-based approach above is the supported path.
ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.
