Web and product analytics
How to build a funnel exploration in GA4
The short answer
In GA4 go to Explore and pick the Funnel exploration template, then define up to 10 steps, each from an event or dimension condition. Choose a closed funnel (everyone must start at step one) to measure a checkout or signup, or an open funnel to see people entering midway. Turn on elapsed time, and add a breakdown such as device category to see where each group drops out.
On this page 5 sections
Step by step
- Open Explore, then under Start a new exploration choose Funnel exploration.
- Click the pencil beside Steps. Name each step and give it a condition: an event (page_view where page_location contains /pricing, generate_lead, purchase) or a dimension value. Steps can use events and dimensions, not metrics.
- Decide between open and closed. In a closed funnel users must enter at step one; in an open funnel they can enter at any step. In both, a user only counts for steps they complete in order.
- For each step after the first, choose indirectly followed by (other actions may happen between steps) or directly followed by (the next step must be the very next action). Indirect is right for most journeys.
- Switch on Show elapsed time to see the average time between steps.
- Add a Breakdown dimension, such as Device category or Session default channel group. Each user is attributed to the first breakdown value they had in the funnel.
- Add a Next action dimension (Event name or Page path) to see the top five actions after each step: where people go instead of continuing.
- Switch the visualisation to Trended funnel to see whether a step's completion rate changed after a site release.
A lead-generation example
| Step | Condition | What a drop here usually means |
|---|---|---|
| 1. Service page | page_view, page path begins with /services/ | Wrong traffic or a weak page |
| 2. Contact or quote page | page_view, page path equals /quote | No clear call to action, or pricing worries |
| 3. Form started | form_start (enhanced measurement) | Form looks long or asks too much |
| 4. Lead | generate_lead | Validation errors, broken form, or spam filters |
A big fall between steps 3 and 4 is the cheapest problem to fix: people wanted to enquire and the form stopped them. Watch session recordings of that step before redesigning anything.
Reading it without fooling yourself
- Explorations are not retroactive beyond your data retention setting (two months by default for event-level data, up to 14 months on standard properties). Raise retention on day one.
- Small funnels are noisy. A step that converts 20% from 40 users could plausibly be anywhere from about 10% to 35%; use the conversion rate calculator for the interval before declaring a problem.
- Thresholding can hide rows when Google signals or demographics are involved; small sites see it most.
- Save a useful exploration as a funnel report so others can open it from Reports.
Questions
- How many steps can a GA4 funnel exploration have?
- Up to 10 steps. Each step is defined by events or dimension values, and a user only counts for steps completed in the order you set.
- What is the difference between an open and a closed funnel in GA4?
- In a closed funnel users must enter at the first step to be counted. In an open funnel they can enter at any step, which shows people who arrive midway, for example straight on a pricing page from an ad.
Sources
Next step
Try the conversion rate calculator
Related guides
How to set up event tracking in GA4
GA4 event tracking step by step: automatic, enhanced measurement, recommended and custom events, key events, custom dimensions, and how to test them.
Heatmaps and session recordings explained
What click, scroll and move heatmaps and session recordings show, what they cannot tell you, how much traffic you need, privacy and masking, and free options.
Cohort analysis explained, with GA4 steps
What a cohort retention table shows, how to read the triangle, standard vs rolling vs cumulative retention, and how to build a cohort exploration in GA4.
A/B testing for small business: traffic and significance
How to run an A/B test on a small site, how much traffic you need, how to read statistical significance, and when not to test. With a free calculator.