GA4 Funnels for Ecommerce: The Complete Guide

A revenue total tells you how the month went. A funnel tells you why. The difference between a store that improves and one that guesses is usually whether anyone is looking at where shoppers drop out, step by step, in a way that points at a specific, fixable problem.

GA4's funnel exploration is the tool for exactly that. Built correctly, it shows you how many shoppers move from viewing a product to adding it, to starting checkout, to buying, and it makes the largest leak in that sequence impossible to miss. This guide covers how to build a funnel exploration, how to read it without fooling yourself, and how to turn each drop-off into a priced revenue leak you can rank and fix.

What a GA4 funnel exploration is

A GA4 funnel exploration is a report in the Explore section of Google Analytics 4 that shows how many users complete a defined sequence of steps, and how many drop out at each one.

You define the steps, usually the ecommerce events of your buying journey, and GA4 shows the count and conversion rate between each. Unlike a single conversion number, a funnel exposes the shape of the loss: not just that you lose sales, but exactly where. That is what makes it the single most useful ecommerce report for finding revenue leaks.

Funnels only work if the underlying events are correct, so if you have not already, start with our guide to GA4 ecommerce tracking. A funnel built on broken events produces confident, wrong conclusions.

Why funnels are the leak-finding tool

A funnel turns "we lose sales" into "we lose 57% of shoppers between checkout and purchase, and that step costs us more than any other." It converts a general problem into a specific, located, measurable one, which is the first requirement for fixing anything.

Every revenue leak lives at a step. Shoppers view a product and do not add it, add it and do not start checkout, start checkout and do not pay. Each of those transitions has a conversion rate, and a poor one is a leak with a location. Without a funnel you can see the final total and guess; with one you can point at the exact step that is costing the most.

This is why the funnel is the entry point to Revenue Intelligence in practice. The Revenue Intelligence framework begins by identifying and pricing leaks, and the funnel is how you identify them and gather the drop-off numbers you need to price them.

How to build a funnel exploration in GA4

Step 1: Open a funnel exploration

In GA4, go to Explore, start a new exploration, and choose the Funnel exploration technique. This gives you a blank funnel you will define step by step.

Step 2: Define the steps of your journey

Add a step for each stage of the buying journey, using your ecommerce events in order:

1. view_item (viewed a product)
2. add_to_cart (added to cart)
3. begin_checkout (started checkout)
4. purchase (completed the order)

For a deeper checkout view, insert add_shipping_info and add_payment_info between begin_checkout and purchase. Those two steps isolate friction inside the checkout itself, which is often where the largest fixable leaks hide, as covered in checkout optimization.

Step 3: Choose open or closed, and add a breakdown

Two settings shape what the funnel tells you.

• Open versus closed. A closed funnel counts only shoppers who move through the steps in order, giving a strict sequential view. An open funnel counts anyone who reached a step regardless of how they got there. Use closed for a clean journey analysis, and open when you want to see the true volume at each step including shoppers who entered mid-way.
• Breakdown. Add a breakdown dimension, most importantly device category. This splits the funnel by mobile, desktop and tablet, and it is where the single most common hidden leak appears.

Step 4: Add elapsed time and segments

Turn on elapsed time to see how long shoppers take between steps. A long gap before purchase can signal hesitation or a slow checkout. Then apply segments, by traffic source or new versus returning, to see whether a leak affects everyone or a specific group. A funnel that is fine for returning customers but terrible for new ones is a very different problem from one that fails for everyone.

Reading the funnel to find revenue leaks

Building the funnel is the easy part. Reading it well is where the value is.

Find the biggest drop, then price it

Start with the largest percentage drop between steps, because that is usually the biggest opportunity. But do not stop at the percentage. A 30% drop on a step that 40,000 shoppers reach can cost more than a 60% drop on a step only 2,000 reach. Rank by revenue, not by percentage.

Here is an illustrative funnel to show the reading. The figures are illustrative, built to demonstrate the method rather than drawn from a real store:

Two steps stand out. The view-to-cart step loses 80% of shoppers, which points at the product page. The checkout-to-purchase step loses 57%, which points at checkout friction. To decide which to fix first, price each: multiply the recoverable share of the drop by the traffic and your average order value. The revenue loss calculator and conversion rate lift calculator turn these percentages into dollars, and revenue leak detection shows the full method.

Segment by device, always

Run the same funnel split by device. It is common to find a checkout that converts acceptably on desktop and falls off a cliff on mobile, where most traffic now is. The blended funnel hides it; the device breakdown reveals it. A weak mobile step is one of the most reliably recoverable leaks in ecommerce.

Diagnose the step, then act

Each step points at a different kind of problem, so the funnel tells you not just where to look but what to look for.

• A weak view-to-cart step points at the product page: images, reviews, price clarity, covered in product page optimization.
• A weak cart-to-checkout step points at surprise costs or a cart that does not reassure.
• A weak checkout-to-purchase step points at form friction, trust, or payment problems.

Beyond the standard funnel

The funnel exploration is the workhorse, but two related tools help.

• Path exploration shows the routes shoppers actually take, forward or backward from any event. It is useful when a funnel drop is puzzling and you want to see where those shoppers went instead.
• Segment overlap helps when you suspect a leak affects a specific group, by showing how segments intersect.

Use these to explain a funnel drop, not to replace the funnel. The funnel tells you where the leak is; these tell you more about who and why.

The limits of GA4 funnels

Read funnels with their limitations in mind, so you do not over-trust a number.

• Modeled and consented data. GA4 reports on consented, observed activity and models part of the gap. Your funnel represents most shoppers, not literally all of them.
• Thresholding and sampling. On smaller segments GA4 may withhold or estimate data to protect privacy, which can make very granular funnels noisy.
• Attribution is separate. A funnel shows on-site drop-off; it does not tell you which channel deserves credit for the shoppers who did convert. That is revenue attribution, a different question.
• A funnel is only as good as its events. This is the big one. If begin_checkout misfires, your checkout funnel is wrong in a way that looks completely normal.

Common GA4 funnel mistakes

Even careful teams repeat these. Treat the list as a pre-flight check.

• Ranking by percentage instead of revenue. The scariest percentage is not always the biggest leak.
• Never segmenting by device. The blended funnel hides the mobile step that is actually failing.
• Building on unvalidated events. A funnel on broken tracking is confidently wrong.
• Confusing open and closed funnels. They answer different questions; know which you are reading.
• Stopping at "conversion is low." The funnel exists to locate the loss at a step, not to restate the total.
• Ignoring the limits. Treating modeled, thresholded data as exact leads to over-precise, brittle conclusions.

The GA4 funnel checklist

Use this to audit any store's funnel analysis in a single pass.

• A funnel exploration covers the core journey: view, add to cart, begin checkout, purchase.
• Mid-checkout steps are included where relevant: shipping info and payment info.
• The underlying events are validated, so the funnel reflects reality.
• The funnel is segmented by device, and by new versus returning where useful.
• Open versus closed is set deliberately for the question being asked.
• Each step-to-step drop is priced in revenue, not just read as a percentage.
• The largest priced drop is fixed first, and the funnel is re-checked after.

Frequently asked questions

What is a GA4 funnel exploration?

It is a report in the Explore section of GA4 that shows how many users complete a defined sequence of steps and how many drop out at each. For ecommerce, the steps are usually the buying journey events, from viewing a product to purchasing, which makes step-by-step drop-off visible.

How do I build an ecommerce funnel in GA4?

Open Explore, choose the Funnel exploration technique, and add a step for each stage using your ecommerce events in order: view_item, add_to_cart, begin_checkout, purchase. Add a device breakdown, and insert add_shipping_info and add_payment_info for a deeper checkout view.

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

A closed funnel counts only users who move through the steps in order, giving a strict sequential view. An open funnel counts anyone who reached a step regardless of the path they took, including shoppers who entered mid-way. Use closed for clean journey analysis and open to see true volume at each step.

How do I find where customers drop off in GA4?

Build a funnel across your journey events and read the conversion rate between steps. The step with the largest drop-off is where you lose the most shoppers. Segment by device to reveal drops that the blended view hides, then price each drop to decide which to fix first.

Which funnel step should I fix first?

The one that costs the most, not the one with the scariest percentage. Multiply each step's drop-off by the traffic reaching it and your average order value to price it, then fix the most expensive leak first. A smaller drop on high traffic can outweigh a larger drop on low traffic.

Why does my GA4 funnel look wrong?

Usually because the underlying events are misfiring: a missing begin_checkout, a double-counted purchase, or an inconsistent item_id. A funnel is only as accurate as its events, so validate them in DebugView first. See GA4 ecommerce tracking for how.

Can GA4 funnels track across sessions?

Yes. GA4 is user-centric rather than session-centric, so a funnel can include shoppers who complete steps across multiple visits, which reflects how ecommerce purchases actually happen. Elapsed time between steps helps you see when shoppers return later to convert.

How is a GA4 funnel different from attribution?

A funnel shows where on your site shoppers drop out. Attribution shows which channels deserve credit for the shoppers who convert. They answer different questions, and you need both: the funnel to find leaks, attribution to allocate acquisition.

Are GA4 funnels accurate?

They are accurate enough to find leaks, with caveats. GA4 reports consented, observed activity and models part of the gap, and it may threshold small segments for privacy. Treat funnels as a reliable guide to where the largest leaks are, not as an exact census of every shopper.

What is a revenue leak?

A revenue leak is a specific, fixable point where ready buyers drop out and the sale is lost. A funnel is how you locate one, by showing the step where shoppers leave. See our definition of a revenue leak for the full concept.

Conclusion and next steps

A funnel is how a store stops guessing. It takes a vague sense that conversion could be better and turns it into a specific, located, priceable leak at a named step. The teams that improve fastest are the ones that build a clean funnel, read it by device, rank the drops in dollars, and fix the most expensive one first.

Your next three moves:

1. Validate your events, then build a funnel across your core journey in GA4, segmented by device.
2. Price the biggest drop with the revenue loss calculator, so you fix by impact rather than by which percentage looks worst.
3. Turn the funnel into action with the Revenue Intelligence framework, or run a free revenue audit that finds and prices your funnel leaks for you.

Build the funnel, read it in revenue, and the next thing to fix stops being a matter of opinion.