Funnel Analysis in GA4: Find Where Visitors Drop Off

Funnel analysis shows you how many visitors move through each step towards a conversion and, more importantly, where they give up. In GA4 you can build a conversion funnel in a few minutes using a funnel exploration. The hard part is setting it up so the numbers mean something, then working out why people leave.

Quick answer

To run funnel analysis in GA4, go to Explore, create a Funnel exploration, and add each step of your journey as an event or page condition (for example landing page, product view, add to cart, begin checkout, purchase). GA4 shows how many users complete each step and the drop-off percentage between them. Use breakdowns by device or source to find where the biggest funnel drop off happens, then use session recordings to see why.

What a conversion funnel is

A conversion funnel is the sequence of steps a visitor takes before converting. Examples:

• Ecommerce: product page, add to cart, begin checkout, add payment info, purchase
• Lead generation: landing page, service page, contact page, form start, form submit
• Booking: pricing page, booking page, select date, confirm booking

Every step loses some people. That is normal. Funnel analysis is about finding the step that loses more than it should.

Before you build: check your events

A funnel is only as good as the events behind it. Before building anything:

• Confirm each step is tracked. GA4 enhanced measurement captures page views, scrolls, outbound clicks and some form interactions, but ecommerce steps and reliable form submissions usually need setup via Google Tag Manager or your platform.
• Track successful submissions, not button clicks. A click on "Submit" fires even when validation fails. Fire your event on the success message or a confirmation page that cannot be revisited.
• Test in DebugView. Walk through the journey yourself and confirm each event fires once, in order.
• Use consistent naming. If your ecommerce events follow GA4's recommended names (view_item, add_to_cart, begin_checkout, purchase), funnels are far easier to build.

Also be aware that consent choices, ad blockers and cross-device journeys mean GA4 will never see every visitor. Funnels show relative drop-off well, even if absolute numbers are incomplete.

How to build a funnel exploration in GA4

1. In GA4, open Explore and choose Funnel exploration.
2. Set a date range long enough to give meaningful volume.
3. Under Steps, click edit and add each step. Each step can be an event (such as add_to_cart) or a condition such as page_view where page location contains /contact.
4. Decide on an open or closed funnel. A closed funnel only counts users who enter at step one. An open funnel lets users join at any step, which is useful when people arrive directly on deeper pages.
5. Optionally, mark a step as directly followed by the previous step if you need strict sequence, or set a time constraint between steps.
6. Add a breakdown dimension such as device category, session default channel group or landing page.
7. Switch on Make funnel trended to see how each step performs over time.

GA4 shows completion and abandonment counts and rates for each step. You can right-click a step to create a segment of users who abandoned there, and use it elsewhere in your analysis.

Standard reports

If you use GA4 ecommerce events, the built-in Purchase journey and Checkout journey reports give a quick view without building an exploration. Explorations give far more control, so most analysis happens there.

How to read funnel drop off properly

Raw drop-off percentages can mislead. Keep these points in mind.

Compare segments, not just totals

A checkout step losing many users on mobile but few on desktop points to a mobile-specific problem, perhaps a payment button off-screen or a form field that is awkward to complete on a phone. The total hides that.

Expect natural drop-off early

Big losses between landing and the next page are common because many visitors are just browsing. The most actionable leaks are usually later, when people have shown intent and then stop.

Watch for tracking artefacts

A step showing almost no drop-off, or more users than the step before, often means an event fires in the wrong place. Investigate before drawing conclusions.

Look at trends

A step that suddenly worsens after a site update is a strong lead. Trended funnels make this obvious.

Tracking form abandonment

For lead generation sites, the form is often the most important step, and form abandonment is common. To measure it in GA4:

1. Track form start. GA4 enhanced measurement can record form_start when someone first interacts with a form, though it does not work reliably on every form setup. A custom GTM trigger is often more dependable.
2. Track successful submission. Use a confirmation-based event as described above.
3. Build a two-step funnel: form_start then your submission event. The gap is your abandonment rate.
4. Find the problem field. GA4 does not show field-level behaviour by default. Session recordings from tools such as Microsoft Clarity reveal which fields people hesitate on, get errors on or quit at. Our comparison of Microsoft Clarity vs Hotjar can help you choose a tool.

Common causes of form abandonment include:

• Asking for too much information too early
• Unclear or unhelpful error messages
• Phone number or postcode fields with strict formatting rules
• Required fields that feel intrusive, such as budget before any context
• Forms that clear entered data after an error
• Captchas that are difficult on mobile

From drop-off to fixes

Finding the leak is half the job. To fix it:

1. Isolate the segment. Which device, source or landing page has the worst drop-off at that step?
2. Watch recordings. Filter to users who reached the step and left. Look for repeated patterns.
3. Review the step yourself. Complete it on a real phone, on mobile data.
4. Form a hypothesis. "Users drop off at checkout on mobile because the delivery cost appears late and surprises them."
5. Make one change at a time where possible, so you know what worked.
6. Re-run the funnel after a sensible period and compare like with like.

This is the core of a structured CRO process, explained further in our CRO audit guide.

Common funnel analysis mistakes

• Building funnels on events that do not fire reliably
• Using button clicks as a proxy for submissions
• Reviewing only blended data rather than device and channel segments
• Using a date range too short to be meaningful
• Treating a single week's movement as a trend
• Ignoring that cross-domain journeys (for example to a third-party checkout or booking tool) can break the funnel unless cross-domain tracking is configured

FAQs

How many steps can a GA4 funnel exploration have?

GA4 funnel explorations support up to 10 steps, which is more than enough for most journeys.

What is the difference between an open and closed funnel in GA4?

A closed funnel only counts users who start at the first step. An open funnel counts users entering at any step, which suits sites where visitors land deep in the journey.

Why does my GA4 funnel show more users at a later step?

In an open funnel users can join part way through. In a closed funnel it usually indicates a tracking issue, such as an event firing on the wrong page.

Can GA4 show which form field people abandon on?

Not by default. You can build custom field-level events in GTM, but session recordings are usually the quickest way to see where people quit.

Find your biggest leak

Our CRO & Behaviour Audit costs a fixed £229. We map your funnel drop-off in GA4, analyse session behaviour in Microsoft Clarity and identify the UX friction costing you conversions, with clear, prioritised recommendations. If your events need checking first, Conversion Clarity covers tracking and CRO together for £379. Get My Audit.

Related reading: What a good conversion rate looks like.