GA4 funnel exploration: finding where shoppers drop out
It shows where people leave. It does not show why. Keep that in mind, and the rest of this page is about using it well. Where this report sits in the wider ecommerce marketing technology stack is on the technologies index.
What does it do?
Each step is a condition. Google's Analytics Help says conditions can be events users trigger or dimension values they share, but not metrics1. The report counts how many users reach each step and how many carry on to the next one. In an open funnel, users can enter at any step. In a closed funnel, they must enter at the first one1.
It is one of several explorations in GA4. A separate report, path exploration, works the other way round: it has either a starting point or an ending point, not both, and shows the screens or events around it2. Use the funnel when you know the steps. Use paths when you do not.
For checkout, Google also offers a ready-made checkout journey report. It is a closed funnel that starts at Begin checkout, and you can't customise it3.
The funnel only works if the events exist. Google's ecommerce Q&A lists begin_checkout, add_shipping_info, add_payment_info and purchase as the events to send for a purchase funnel4. The checkout journey report gets each of its steps from one of these events, so a missing event means a missing step3. Google notes that Shopify collects some of them automatically but not add_shipping_info3.
Why do e-commerce brands use it?
Because "the conversion rate is low" is not a diagnosis. Conversion rate is one number for the whole journey. A funnel splits it into steps, so you can see whether the problem is the product page, the basket or the payment step.
It also gives a first look at where carts are abandoned, a step-by-step view of cart abandonment rate, without any extra tool. And it is the natural first step of CRO (conversion rate optimisation): look at the data before you decide what to test.
For the next step, read how to improve ecommerce conversion rate.
How do we implement it?
Check the events
Make sure the shop sends the full set of ecommerce events with the right parameters. A missing event is a missing step.
Reconcile first
Compare GA4 purchases with your shop's orders before you trust any funnel.
Define the steps
Choose the steps, decide whether people must go through them in order, and say whether you count sessions or users.
Turn drops into hypotheses
A drop shows where, not why. Pair it with session recordings, customer feedback or a test.
When should you not use it?
Do not rely on it when your ecommerce events are broken or partial. Audit the tracking first.
Do not use it to judge small changes on low traffic. A step with few people in it can swing a lot for no real reason.
If you want the cause rather than the location, you need qualitative research or a controlled test, not another report.
What does it cost in effort?
No figures. The effort is:
- Events. Make sure the shop sends the full set of ecommerce events, with the right parameters.
- Funnel set-up. Choose the steps, and decide whether people must go through them in order.
- Reconciliation. Compare GA4 purchases with your shop's orders before you trust a funnel.
- Reading. Someone has to turn a drop-off into a hypothesis.
Pitfalls
Missing events mean missing steps
The checkout steps come from specific events3. A sudden drop may be a tracking gap, not a customer problem.
Purchases that do not match the shop
If GA4 and the shop disagree on orders, fix that first.
Consent gaps
Our view: when visitors decline analytics consent, parts of the journey disappear from the data. A hole in the funnel can be a hole in consent. What is legal for your site is a question for your lawyer.
Reading a drop as a verdict
A drop shows where, not why. Pair it with session recordings, customer feedback or a test.
A/B testing inside GA4
GA4 is not a test tool. Google's developer documentation describes integrating an external experiment tool with Google Analytics, which then lets you read results there5.
Mixing up units
A session and a user are different things. Say which one your funnel counts.
Our take: Core
In our view, GA4 funnel exploration is Core. That is our rating, not a market fact.
Our reasoning: if you run GA4, the report is already there. It needs no extra tool, only correct events. That makes it the sensible first diagnostic when a checkout looks leaky.
For ZFD: We start conversion work here, then check the numbers against the shop before we draw conclusions.
FAQ
What is GA4 funnel exploration?
It is a report in GA4 where you set up to 10 steps and see how many users move through them. Funnels can be open (enter at any step) or closed (enter at the first step)1.
Which events do I need for a checkout funnel?
What is the difference between funnel and path exploration?
A funnel follows steps you define. Path exploration has a starting point or an ending point you pick and shows what users did after or before it2.
Is a [key event](/glossary/key-event) the same as a funnel step?
No. A key event is an event you mark as important to your business. A funnel step is a condition you choose for the report, such as an event, and it does not need to be a key event.
Can I run A/B tests in GA4?
Not on its own. Google documents an integration for external experiment tools, with results read in Google Analytics5.
Sources
- [GA4] Funnel exploration - Analytics HelpPrimary source
- [GA4] Path exploration - Analytics HelpPrimary source
- [GA4] Checkout journey report - Analytics HelpPrimary source
- Ecommerce setup Q&A - Analytics HelpPrimary source
- Create an experiment integration with Google Analytics - Google for DevelopersPrimary source
Let’s find what’s actually working.
Next step. Want someone to check your events and read the funnel with you? Talk to us about your setup. We also offer conversion rate optimisation for ecommerce, including audits and implementation.